Merge "Update to libxml2 2.9.9 (2019-01-03)."
am: 3f2b932875
Change-Id: I41a032276eb5ee4c99c8b875d01ac29107d0b903
diff --git a/Android.bp b/Android.bp
index e3a02f5..a09eeb4 100644
--- a/Android.bp
+++ b/Android.bp
@@ -84,6 +84,17 @@
cflags: [
"-DLIBXML_THREAD_ENABLED=1",
+ // Poison various things that should not be defined.
+ "-DLIBXML_LEGACY_ENABLED=NO",
+ "-DLIBXML_FTP_ENABLED=NO",
+ "-DLIBXML_HTML_ENABLED=NO",
+ "-DLIBXML_HTTP_ENABLED=NO",
+ "-DLIBXML_ICONV_ENABLED=NO",
+ "-DLIBXML_LZMA_ENABLED=NO",
+ "-DLIBXML_ZLIB_ENABLED=NO",
+ // And this one depends on whether or not we're building for the VNDK.
+ "-DLIBXML_ICU_ENABLED=NO",
+
"-Wall",
"-Werror",
"-Wno-error=ignored-attributes",
diff --git a/HTMLparser.c b/HTMLparser.c
new file mode 100644
index 0000000..9e60e27
--- /dev/null
+++ b/HTMLparser.c
@@ -0,0 +1,7171 @@
+/*
+ * HTMLparser.c : an HTML 4.0 non-verifying parser
+ *
+ * See Copyright for the status of this software.
+ *
+ * daniel@veillard.com
+ */
+
+#define IN_LIBXML
+#include "libxml.h"
+#ifdef LIBXML_HTML_ENABLED
+
+#include <string.h>
+#ifdef HAVE_CTYPE_H
+#include <ctype.h>
+#endif
+#ifdef HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
+#ifdef HAVE_SYS_STAT_H
+#include <sys/stat.h>
+#endif
+#ifdef HAVE_FCNTL_H
+#include <fcntl.h>
+#endif
+#ifdef HAVE_UNISTD_H
+#include <unistd.h>
+#endif
+#ifdef LIBXML_ZLIB_ENABLED
+#include <zlib.h>
+#endif
+
+#include <libxml/xmlmemory.h>
+#include <libxml/tree.h>
+#include <libxml/parser.h>
+#include <libxml/parserInternals.h>
+#include <libxml/xmlerror.h>
+#include <libxml/HTMLparser.h>
+#include <libxml/HTMLtree.h>
+#include <libxml/entities.h>
+#include <libxml/encoding.h>
+#include <libxml/valid.h>
+#include <libxml/xmlIO.h>
+#include <libxml/globals.h>
+#include <libxml/uri.h>
+
+#include "buf.h"
+#include "enc.h"
+
+#define HTML_MAX_NAMELEN 1000
+#define HTML_PARSER_BIG_BUFFER_SIZE 1000
+#define HTML_PARSER_BUFFER_SIZE 100
+
+/* #define DEBUG */
+/* #define DEBUG_PUSH */
+
+static int htmlOmittedDefaultValue = 1;
+
+xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
+ xmlChar end, xmlChar end2, xmlChar end3);
+static void htmlParseComment(htmlParserCtxtPtr ctxt);
+
+/************************************************************************
+ * *
+ * Some factorized error routines *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlErrMemory:
+ * @ctxt: an HTML parser context
+ * @extra: extra informations
+ *
+ * Handle a redefinition of attribute error
+ */
+static void
+htmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
+{
+ if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
+ (ctxt->instate == XML_PARSER_EOF))
+ return;
+ if (ctxt != NULL) {
+ ctxt->errNo = XML_ERR_NO_MEMORY;
+ ctxt->instate = XML_PARSER_EOF;
+ ctxt->disableSAX = 1;
+ }
+ if (extra)
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
+ XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
+ NULL, NULL, 0, 0,
+ "Memory allocation failed : %s\n", extra);
+ else
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
+ XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
+ NULL, NULL, 0, 0, "Memory allocation failed\n");
+}
+
+/**
+ * htmlParseErr:
+ * @ctxt: an HTML parser context
+ * @error: the error number
+ * @msg: the error message
+ * @str1: string infor
+ * @str2: string infor
+ *
+ * Handle a fatal parser error, i.e. violating Well-Formedness constraints
+ */
+static void LIBXML_ATTR_FORMAT(3,0)
+htmlParseErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
+ const char *msg, const xmlChar *str1, const xmlChar *str2)
+{
+ if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
+ (ctxt->instate == XML_PARSER_EOF))
+ return;
+ if (ctxt != NULL)
+ ctxt->errNo = error;
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
+ XML_ERR_ERROR, NULL, 0,
+ (const char *) str1, (const char *) str2,
+ NULL, 0, 0,
+ msg, str1, str2);
+ if (ctxt != NULL)
+ ctxt->wellFormed = 0;
+}
+
+/**
+ * htmlParseErrInt:
+ * @ctxt: an HTML parser context
+ * @error: the error number
+ * @msg: the error message
+ * @val: integer info
+ *
+ * Handle a fatal parser error, i.e. violating Well-Formedness constraints
+ */
+static void LIBXML_ATTR_FORMAT(3,0)
+htmlParseErrInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
+ const char *msg, int val)
+{
+ if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
+ (ctxt->instate == XML_PARSER_EOF))
+ return;
+ if (ctxt != NULL)
+ ctxt->errNo = error;
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
+ XML_ERR_ERROR, NULL, 0, NULL, NULL,
+ NULL, val, 0, msg, val);
+ if (ctxt != NULL)
+ ctxt->wellFormed = 0;
+}
+
+/************************************************************************
+ * *
+ * Parser stacks related functions and macros *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlnamePush:
+ * @ctxt: an HTML parser context
+ * @value: the element name
+ *
+ * Pushes a new element name on top of the name stack
+ *
+ * Returns 0 in case of error, the index in the stack otherwise
+ */
+static int
+htmlnamePush(htmlParserCtxtPtr ctxt, const xmlChar * value)
+{
+ if ((ctxt->html < 3) && (xmlStrEqual(value, BAD_CAST "head")))
+ ctxt->html = 3;
+ if ((ctxt->html < 10) && (xmlStrEqual(value, BAD_CAST "body")))
+ ctxt->html = 10;
+ if (ctxt->nameNr >= ctxt->nameMax) {
+ ctxt->nameMax *= 2;
+ ctxt->nameTab = (const xmlChar * *)
+ xmlRealloc((xmlChar * *)ctxt->nameTab,
+ ctxt->nameMax *
+ sizeof(ctxt->nameTab[0]));
+ if (ctxt->nameTab == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ return (0);
+ }
+ }
+ ctxt->nameTab[ctxt->nameNr] = value;
+ ctxt->name = value;
+ return (ctxt->nameNr++);
+}
+/**
+ * htmlnamePop:
+ * @ctxt: an HTML parser context
+ *
+ * Pops the top element name from the name stack
+ *
+ * Returns the name just removed
+ */
+static const xmlChar *
+htmlnamePop(htmlParserCtxtPtr ctxt)
+{
+ const xmlChar *ret;
+
+ if (ctxt->nameNr <= 0)
+ return (NULL);
+ ctxt->nameNr--;
+ if (ctxt->nameNr < 0)
+ return (NULL);
+ if (ctxt->nameNr > 0)
+ ctxt->name = ctxt->nameTab[ctxt->nameNr - 1];
+ else
+ ctxt->name = NULL;
+ ret = ctxt->nameTab[ctxt->nameNr];
+ ctxt->nameTab[ctxt->nameNr] = NULL;
+ return (ret);
+}
+
+/**
+ * htmlNodeInfoPush:
+ * @ctxt: an HTML parser context
+ * @value: the node info
+ *
+ * Pushes a new element name on top of the node info stack
+ *
+ * Returns 0 in case of error, the index in the stack otherwise
+ */
+static int
+htmlNodeInfoPush(htmlParserCtxtPtr ctxt, htmlParserNodeInfo *value)
+{
+ if (ctxt->nodeInfoNr >= ctxt->nodeInfoMax) {
+ if (ctxt->nodeInfoMax == 0)
+ ctxt->nodeInfoMax = 5;
+ ctxt->nodeInfoMax *= 2;
+ ctxt->nodeInfoTab = (htmlParserNodeInfo *)
+ xmlRealloc((htmlParserNodeInfo *)ctxt->nodeInfoTab,
+ ctxt->nodeInfoMax *
+ sizeof(ctxt->nodeInfoTab[0]));
+ if (ctxt->nodeInfoTab == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ return (0);
+ }
+ }
+ ctxt->nodeInfoTab[ctxt->nodeInfoNr] = *value;
+ ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
+ return (ctxt->nodeInfoNr++);
+}
+
+/**
+ * htmlNodeInfoPop:
+ * @ctxt: an HTML parser context
+ *
+ * Pops the top element name from the node info stack
+ *
+ * Returns 0 in case of error, the pointer to NodeInfo otherwise
+ */
+static htmlParserNodeInfo *
+htmlNodeInfoPop(htmlParserCtxtPtr ctxt)
+{
+ if (ctxt->nodeInfoNr <= 0)
+ return (NULL);
+ ctxt->nodeInfoNr--;
+ if (ctxt->nodeInfoNr < 0)
+ return (NULL);
+ if (ctxt->nodeInfoNr > 0)
+ ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr - 1];
+ else
+ ctxt->nodeInfo = NULL;
+ return &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
+}
+
+/*
+ * Macros for accessing the content. Those should be used only by the parser,
+ * and not exported.
+ *
+ * Dirty macros, i.e. one need to make assumption on the context to use them
+ *
+ * CUR_PTR return the current pointer to the xmlChar to be parsed.
+ * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
+ * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
+ * in UNICODE mode. This should be used internally by the parser
+ * only to compare to ASCII values otherwise it would break when
+ * running with UTF-8 encoding.
+ * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
+ * to compare on ASCII based substring.
+ * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
+ * it should be used only to compare on ASCII based substring.
+ * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
+ * strings without newlines within the parser.
+ *
+ * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
+ *
+ * CURRENT Returns the current char value, with the full decoding of
+ * UTF-8 if we are using this mode. It returns an int.
+ * NEXT Skip to the next character, this does the proper decoding
+ * in UTF-8 mode. It also pop-up unfinished entities on the fly.
+ * NEXTL(l) Skip the current unicode character of l xmlChars long.
+ * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
+ */
+
+#define UPPER (toupper(*ctxt->input->cur))
+
+#define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val),ctxt->input->col+=(val)
+
+#define NXT(val) ctxt->input->cur[(val)]
+
+#define UPP(val) (toupper(ctxt->input->cur[(val)]))
+
+#define CUR_PTR ctxt->input->cur
+#define BASE_PTR ctxt->input->base
+
+#define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
+ (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
+ xmlParserInputShrink(ctxt->input)
+
+#define GROW if ((ctxt->progressive == 0) && \
+ (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
+ xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
+
+#define CURRENT ((int) (*ctxt->input->cur))
+
+#define SKIP_BLANKS htmlSkipBlankChars(ctxt)
+
+/* Inported from XML */
+
+/* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
+#define CUR ((int) (*ctxt->input->cur))
+#define NEXT xmlNextChar(ctxt)
+
+#define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
+
+
+#define NEXTL(l) do { \
+ if (*(ctxt->input->cur) == '\n') { \
+ ctxt->input->line++; ctxt->input->col = 1; \
+ } else ctxt->input->col++; \
+ ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \
+ } while (0)
+
+/************
+ \
+ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
+ if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
+ ************/
+
+#define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
+#define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
+
+#define COPY_BUF(l,b,i,v) \
+ if (l == 1) b[i++] = (xmlChar) v; \
+ else i += xmlCopyChar(l,&b[i],v)
+
+/**
+ * htmlFindEncoding:
+ * @the HTML parser context
+ *
+ * Ty to find and encoding in the current data available in the input
+ * buffer this is needed to try to switch to the proper encoding when
+ * one face a character error.
+ * That's an heuristic, since it's operating outside of parsing it could
+ * try to use a meta which had been commented out, that's the reason it
+ * should only be used in case of error, not as a default.
+ *
+ * Returns an encoding string or NULL if not found, the string need to
+ * be freed
+ */
+static xmlChar *
+htmlFindEncoding(xmlParserCtxtPtr ctxt) {
+ const xmlChar *start, *cur, *end;
+
+ if ((ctxt == NULL) || (ctxt->input == NULL) ||
+ (ctxt->input->encoding != NULL) || (ctxt->input->buf == NULL) ||
+ (ctxt->input->buf->encoder != NULL))
+ return(NULL);
+ if ((ctxt->input->cur == NULL) || (ctxt->input->end == NULL))
+ return(NULL);
+
+ start = ctxt->input->cur;
+ end = ctxt->input->end;
+ /* we also expect the input buffer to be zero terminated */
+ if (*end != 0)
+ return(NULL);
+
+ cur = xmlStrcasestr(start, BAD_CAST "HTTP-EQUIV");
+ if (cur == NULL)
+ return(NULL);
+ cur = xmlStrcasestr(cur, BAD_CAST "CONTENT");
+ if (cur == NULL)
+ return(NULL);
+ cur = xmlStrcasestr(cur, BAD_CAST "CHARSET=");
+ if (cur == NULL)
+ return(NULL);
+ cur += 8;
+ start = cur;
+ while (((*cur >= 'A') && (*cur <= 'Z')) ||
+ ((*cur >= 'a') && (*cur <= 'z')) ||
+ ((*cur >= '0') && (*cur <= '9')) ||
+ (*cur == '-') || (*cur == '_') || (*cur == ':') || (*cur == '/'))
+ cur++;
+ if (cur == start)
+ return(NULL);
+ return(xmlStrndup(start, cur - start));
+}
+
+/**
+ * htmlCurrentChar:
+ * @ctxt: the HTML parser context
+ * @len: pointer to the length of the char read
+ *
+ * The current char value, if using UTF-8 this may actually span multiple
+ * bytes in the input buffer. Implement the end of line normalization:
+ * 2.11 End-of-Line Handling
+ * If the encoding is unspecified, in the case we find an ISO-Latin-1
+ * char, then the encoding converter is plugged in automatically.
+ *
+ * Returns the current char value and its length
+ */
+
+static int
+htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
+ if (ctxt->instate == XML_PARSER_EOF)
+ return(0);
+
+ if (ctxt->token != 0) {
+ *len = 0;
+ return(ctxt->token);
+ }
+ if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
+ /*
+ * We are supposed to handle UTF8, check it's valid
+ * From rfc2044: encoding of the Unicode values on UTF-8:
+ *
+ * UCS-4 range (hex.) UTF-8 octet sequence (binary)
+ * 0000 0000-0000 007F 0xxxxxxx
+ * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
+ * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
+ *
+ * Check for the 0x110000 limit too
+ */
+ const unsigned char *cur = ctxt->input->cur;
+ unsigned char c;
+ unsigned int val;
+
+ c = *cur;
+ if (c & 0x80) {
+ if (cur[1] == 0) {
+ xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
+ cur = ctxt->input->cur;
+ }
+ if ((cur[1] & 0xc0) != 0x80)
+ goto encoding_error;
+ if ((c & 0xe0) == 0xe0) {
+
+ if (cur[2] == 0) {
+ xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
+ cur = ctxt->input->cur;
+ }
+ if ((cur[2] & 0xc0) != 0x80)
+ goto encoding_error;
+ if ((c & 0xf0) == 0xf0) {
+ if (cur[3] == 0) {
+ xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
+ cur = ctxt->input->cur;
+ }
+ if (((c & 0xf8) != 0xf0) ||
+ ((cur[3] & 0xc0) != 0x80))
+ goto encoding_error;
+ /* 4-byte code */
+ *len = 4;
+ val = (cur[0] & 0x7) << 18;
+ val |= (cur[1] & 0x3f) << 12;
+ val |= (cur[2] & 0x3f) << 6;
+ val |= cur[3] & 0x3f;
+ } else {
+ /* 3-byte code */
+ *len = 3;
+ val = (cur[0] & 0xf) << 12;
+ val |= (cur[1] & 0x3f) << 6;
+ val |= cur[2] & 0x3f;
+ }
+ } else {
+ /* 2-byte code */
+ *len = 2;
+ val = (cur[0] & 0x1f) << 6;
+ val |= cur[1] & 0x3f;
+ }
+ if (!IS_CHAR(val)) {
+ htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
+ "Char 0x%X out of allowed range\n", val);
+ }
+ return(val);
+ } else {
+ if ((*ctxt->input->cur == 0) &&
+ (ctxt->input->cur < ctxt->input->end)) {
+ htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
+ "Char 0x%X out of allowed range\n", 0);
+ *len = 1;
+ return(' ');
+ }
+ /* 1-byte code */
+ *len = 1;
+ return((int) *ctxt->input->cur);
+ }
+ }
+ /*
+ * Assume it's a fixed length encoding (1) with
+ * a compatible encoding for the ASCII set, since
+ * XML constructs only use < 128 chars
+ */
+ *len = 1;
+ if ((int) *ctxt->input->cur < 0x80)
+ return((int) *ctxt->input->cur);
+
+ /*
+ * Humm this is bad, do an automatic flow conversion
+ */
+ {
+ xmlChar * guess;
+ xmlCharEncodingHandlerPtr handler;
+
+ guess = htmlFindEncoding(ctxt);
+ if (guess == NULL) {
+ xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
+ } else {
+ if (ctxt->input->encoding != NULL)
+ xmlFree((xmlChar *) ctxt->input->encoding);
+ ctxt->input->encoding = guess;
+ handler = xmlFindCharEncodingHandler((const char *) guess);
+ if (handler != NULL) {
+ xmlSwitchToEncoding(ctxt, handler);
+ } else {
+ htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
+ "Unsupported encoding %s", guess, NULL);
+ }
+ }
+ ctxt->charset = XML_CHAR_ENCODING_UTF8;
+ }
+
+ return(xmlCurrentChar(ctxt, len));
+
+encoding_error:
+ /*
+ * If we detect an UTF8 error that probably mean that the
+ * input encoding didn't get properly advertized in the
+ * declaration header. Report the error and switch the encoding
+ * to ISO-Latin-1 (if you don't like this policy, just declare the
+ * encoding !)
+ */
+ {
+ char buffer[150];
+
+ if (ctxt->input->end - ctxt->input->cur >= 4) {
+ snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
+ ctxt->input->cur[0], ctxt->input->cur[1],
+ ctxt->input->cur[2], ctxt->input->cur[3]);
+ } else {
+ snprintf(buffer, 149, "Bytes: 0x%02X\n", ctxt->input->cur[0]);
+ }
+ htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
+ "Input is not proper UTF-8, indicate encoding !\n",
+ BAD_CAST buffer, NULL);
+ }
+
+ ctxt->charset = XML_CHAR_ENCODING_8859_1;
+ *len = 1;
+ return((int) *ctxt->input->cur);
+}
+
+/**
+ * htmlSkipBlankChars:
+ * @ctxt: the HTML parser context
+ *
+ * skip all blanks character found at that point in the input streams.
+ *
+ * Returns the number of space chars skipped
+ */
+
+static int
+htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
+ int res = 0;
+
+ while (IS_BLANK_CH(*(ctxt->input->cur))) {
+ if ((*ctxt->input->cur == 0) &&
+ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
+ xmlPopInput(ctxt);
+ } else {
+ if (*(ctxt->input->cur) == '\n') {
+ ctxt->input->line++; ctxt->input->col = 1;
+ } else ctxt->input->col++;
+ ctxt->input->cur++;
+ ctxt->nbChars++;
+ if (*ctxt->input->cur == 0)
+ xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
+ }
+ res++;
+ }
+ return(res);
+}
+
+
+
+/************************************************************************
+ * *
+ * The list of HTML elements and their properties *
+ * *
+ ************************************************************************/
+
+/*
+ * Start Tag: 1 means the start tag can be ommited
+ * End Tag: 1 means the end tag can be ommited
+ * 2 means it's forbidden (empty elements)
+ * 3 means the tag is stylistic and should be closed easily
+ * Depr: this element is deprecated
+ * DTD: 1 means that this element is valid only in the Loose DTD
+ * 2 means that this element is valid only in the Frameset DTD
+ *
+ * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
+ , subElements , impliedsubelt , Attributes, userdata
+ */
+
+/* Definitions and a couple of vars for HTML Elements */
+
+#define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
+#define NB_FONTSTYLE 8
+#define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
+#define NB_PHRASE 10
+#define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
+#define NB_SPECIAL 16
+#define INLINE FONTSTYLE, PHRASE, SPECIAL, FORMCTRL
+#define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
+#define BLOCK HEADING, LIST, "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
+#define NB_BLOCK NB_HEADING + NB_LIST + 14
+#define FORMCTRL "input", "select", "textarea", "label", "button"
+#define NB_FORMCTRL 5
+#define PCDATA
+#define NB_PCDATA 0
+#define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
+#define NB_HEADING 6
+#define LIST "ul", "ol", "dir", "menu"
+#define NB_LIST 4
+#define MODIFIER
+#define NB_MODIFIER 0
+#define FLOW BLOCK,INLINE
+#define NB_FLOW NB_BLOCK + NB_INLINE
+#define EMPTY NULL
+
+
+static const char* const html_flow[] = { FLOW, NULL } ;
+static const char* const html_inline[] = { INLINE, NULL } ;
+
+/* placeholders: elts with content but no subelements */
+static const char* const html_pcdata[] = { NULL } ;
+#define html_cdata html_pcdata
+
+
+/* ... and for HTML Attributes */
+
+#define COREATTRS "id", "class", "style", "title"
+#define NB_COREATTRS 4
+#define I18N "lang", "dir"
+#define NB_I18N 2
+#define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
+#define NB_EVENTS 9
+#define ATTRS COREATTRS,I18N,EVENTS
+#define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
+#define CELLHALIGN "align", "char", "charoff"
+#define NB_CELLHALIGN 3
+#define CELLVALIGN "valign"
+#define NB_CELLVALIGN 1
+
+static const char* const html_attrs[] = { ATTRS, NULL } ;
+static const char* const core_i18n_attrs[] = { COREATTRS, I18N, NULL } ;
+static const char* const core_attrs[] = { COREATTRS, NULL } ;
+static const char* const i18n_attrs[] = { I18N, NULL } ;
+
+
+/* Other declarations that should go inline ... */
+static const char* const a_attrs[] = { ATTRS, "charset", "type", "name",
+ "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
+ "tabindex", "onfocus", "onblur", NULL } ;
+static const char* const target_attr[] = { "target", NULL } ;
+static const char* const rows_cols_attr[] = { "rows", "cols", NULL } ;
+static const char* const alt_attr[] = { "alt", NULL } ;
+static const char* const src_alt_attrs[] = { "src", "alt", NULL } ;
+static const char* const href_attrs[] = { "href", NULL } ;
+static const char* const clear_attrs[] = { "clear", NULL } ;
+static const char* const inline_p[] = { INLINE, "p", NULL } ;
+
+static const char* const flow_param[] = { FLOW, "param", NULL } ;
+static const char* const applet_attrs[] = { COREATTRS , "codebase",
+ "archive", "alt", "name", "height", "width", "align",
+ "hspace", "vspace", NULL } ;
+static const char* const area_attrs[] = { "shape", "coords", "href", "nohref",
+ "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
+static const char* const basefont_attrs[] =
+ { "id", "size", "color", "face", NULL } ;
+static const char* const quote_attrs[] = { ATTRS, "cite", NULL } ;
+static const char* const body_contents[] = { FLOW, "ins", "del", NULL } ;
+static const char* const body_attrs[] = { ATTRS, "onload", "onunload", NULL } ;
+static const char* const body_depr[] = { "background", "bgcolor", "text",
+ "link", "vlink", "alink", NULL } ;
+static const char* const button_attrs[] = { ATTRS, "name", "value", "type",
+ "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
+
+
+static const char* const col_attrs[] = { ATTRS, "span", "width", CELLHALIGN, CELLVALIGN, NULL } ;
+static const char* const col_elt[] = { "col", NULL } ;
+static const char* const edit_attrs[] = { ATTRS, "datetime", "cite", NULL } ;
+static const char* const compact_attrs[] = { ATTRS, "compact", NULL } ;
+static const char* const dl_contents[] = { "dt", "dd", NULL } ;
+static const char* const compact_attr[] = { "compact", NULL } ;
+static const char* const label_attr[] = { "label", NULL } ;
+static const char* const fieldset_contents[] = { FLOW, "legend" } ;
+static const char* const font_attrs[] = { COREATTRS, I18N, "size", "color", "face" , NULL } ;
+static const char* const form_contents[] = { HEADING, LIST, INLINE, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL } ;
+static const char* const form_attrs[] = { ATTRS, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL } ;
+static const char* const frame_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL } ;
+static const char* const frameset_attrs[] = { COREATTRS, "rows", "cols", "onload", "onunload", NULL } ;
+static const char* const frameset_contents[] = { "frameset", "frame", "noframes", NULL } ;
+static const char* const head_attrs[] = { I18N, "profile", NULL } ;
+static const char* const head_contents[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL } ;
+static const char* const hr_depr[] = { "align", "noshade", "size", "width", NULL } ;
+static const char* const version_attr[] = { "version", NULL } ;
+static const char* const html_content[] = { "head", "body", "frameset", NULL } ;
+static const char* const iframe_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL } ;
+static const char* const img_attrs[] = { ATTRS, "longdesc", "name", "height", "width", "usemap", "ismap", NULL } ;
+static const char* const embed_attrs[] = { COREATTRS, "align", "alt", "border", "code", "codebase", "frameborder", "height", "hidden", "hspace", "name", "palette", "pluginspace", "pluginurl", "src", "type", "units", "vspace", "width", NULL } ;
+static const char* const input_attrs[] = { ATTRS, "type", "name", "value", "checked", "disabled", "readonly", "size", "maxlength", "src", "alt", "usemap", "ismap", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", "accept", NULL } ;
+static const char* const prompt_attrs[] = { COREATTRS, I18N, "prompt", NULL } ;
+static const char* const label_attrs[] = { ATTRS, "for", "accesskey", "onfocus", "onblur", NULL } ;
+static const char* const legend_attrs[] = { ATTRS, "accesskey", NULL } ;
+static const char* const align_attr[] = { "align", NULL } ;
+static const char* const link_attrs[] = { ATTRS, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL } ;
+static const char* const map_contents[] = { BLOCK, "area", NULL } ;
+static const char* const name_attr[] = { "name", NULL } ;
+static const char* const action_attr[] = { "action", NULL } ;
+static const char* const blockli_elt[] = { BLOCK, "li", NULL } ;
+static const char* const meta_attrs[] = { I18N, "http-equiv", "name", "scheme", "charset", NULL } ;
+static const char* const content_attr[] = { "content", NULL } ;
+static const char* const type_attr[] = { "type", NULL } ;
+static const char* const noframes_content[] = { "body", FLOW MODIFIER, NULL } ;
+static const char* const object_contents[] = { FLOW, "param", NULL } ;
+static const char* const object_attrs[] = { ATTRS, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL } ;
+static const char* const object_depr[] = { "align", "border", "hspace", "vspace", NULL } ;
+static const char* const ol_attrs[] = { "type", "compact", "start", NULL} ;
+static const char* const option_elt[] = { "option", NULL } ;
+static const char* const optgroup_attrs[] = { ATTRS, "disabled", NULL } ;
+static const char* const option_attrs[] = { ATTRS, "disabled", "label", "selected", "value", NULL } ;
+static const char* const param_attrs[] = { "id", "value", "valuetype", "type", NULL } ;
+static const char* const width_attr[] = { "width", NULL } ;
+static const char* const pre_content[] = { PHRASE, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL } ;
+static const char* const script_attrs[] = { "charset", "src", "defer", "event", "for", NULL } ;
+static const char* const language_attr[] = { "language", NULL } ;
+static const char* const select_content[] = { "optgroup", "option", NULL } ;
+static const char* const select_attrs[] = { ATTRS, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL } ;
+static const char* const style_attrs[] = { I18N, "media", "title", NULL } ;
+static const char* const table_attrs[] = { ATTRS, "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL } ;
+static const char* const table_depr[] = { "align", "bgcolor", NULL } ;
+static const char* const table_contents[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL} ;
+static const char* const tr_elt[] = { "tr", NULL } ;
+static const char* const talign_attrs[] = { ATTRS, CELLHALIGN, CELLVALIGN, NULL} ;
+static const char* const th_td_depr[] = { "nowrap", "bgcolor", "width", "height", NULL } ;
+static const char* const th_td_attr[] = { ATTRS, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN, CELLVALIGN, NULL } ;
+static const char* const textarea_attrs[] = { ATTRS, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL } ;
+static const char* const tr_contents[] = { "th", "td", NULL } ;
+static const char* const bgcolor_attr[] = { "bgcolor", NULL } ;
+static const char* const li_elt[] = { "li", NULL } ;
+static const char* const ul_depr[] = { "type", "compact", NULL} ;
+static const char* const dir_attr[] = { "dir", NULL} ;
+
+#define DECL (const char**)
+
+static const htmlElemDesc
+html40ElementTable[] = {
+{ "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
+ DECL html_inline , NULL , DECL a_attrs , DECL target_attr, NULL
+},
+{ "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "acronym", 0, 0, 0, 0, 0, 0, 1, "",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
+ DECL inline_p , NULL , DECL html_attrs, NULL, NULL
+},
+{ "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
+ DECL flow_param , NULL , NULL , DECL applet_attrs, NULL
+},
+{ "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
+ EMPTY , NULL , DECL area_attrs , DECL target_attr, DECL alt_attr
+},
+{ "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
+ EMPTY , NULL , NULL , DECL target_attr, DECL href_attrs
+},
+{ "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
+ EMPTY , NULL , NULL, DECL basefont_attrs, NULL
+},
+{ "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
+ DECL html_inline , NULL , DECL core_i18n_attrs, NULL, DECL dir_attr
+},
+{ "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
+ DECL html_flow , NULL , DECL quote_attrs , NULL, NULL
+},
+{ "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
+ DECL body_contents , "div" , DECL body_attrs, DECL body_depr, NULL
+},
+{ "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
+ EMPTY , NULL , DECL core_attrs, DECL clear_attrs , NULL
+},
+{ "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
+ DECL html_flow MODIFIER , NULL , DECL button_attrs, NULL, NULL
+},
+{ "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
+ DECL html_flow , NULL , NULL, DECL html_attrs, NULL
+},
+{ "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
+ EMPTY , NULL , DECL col_attrs , NULL, NULL
+},
+{ "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
+ DECL col_elt , "col" , DECL col_attrs , NULL, NULL
+},
+{ "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
+ DECL html_flow , NULL , DECL html_attrs, NULL, NULL
+},
+{ "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
+ DECL html_flow , NULL , DECL edit_attrs , NULL, NULL
+},
+{ "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
+ DECL html_inline , NULL , DECL html_attrs, NULL, NULL
+},
+{ "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
+ DECL blockli_elt, "li" , NULL, DECL compact_attrs, NULL
+},
+{ "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
+ DECL html_flow, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
+ DECL dl_contents , "dd" , DECL html_attrs, DECL compact_attr, NULL
+},
+{ "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "embed", 0, 1, 0, 0, 1, 1, 1, "generic embedded object ",
+ EMPTY, NULL, DECL embed_attrs, NULL, NULL
+},
+{ "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
+ DECL fieldset_contents , NULL, DECL html_attrs, NULL, NULL
+},
+{ "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
+ DECL html_inline, NULL, NULL, DECL font_attrs, NULL
+},
+{ "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
+ DECL form_contents, "fieldset", DECL form_attrs , DECL target_attr, DECL action_attr
+},
+{ "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
+ EMPTY, NULL, NULL, DECL frame_attrs, NULL
+},
+{ "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
+ DECL frameset_contents, "noframes" , NULL , DECL frameset_attrs, NULL
+},
+{ "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
+ DECL head_contents, NULL, DECL head_attrs, NULL, NULL
+},
+{ "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
+ EMPTY, NULL, DECL html_attrs, DECL hr_depr, NULL
+},
+{ "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
+ DECL html_content , NULL , DECL i18n_attrs, DECL version_attr, NULL
+},
+{ "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
+ DECL html_flow, NULL, NULL, DECL iframe_attrs, NULL
+},
+{ "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
+ EMPTY, NULL, DECL img_attrs, DECL align_attr, DECL src_alt_attrs
+},
+{ "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
+ EMPTY, NULL, DECL input_attrs , DECL align_attr, NULL
+},
+{ "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
+ DECL html_flow, NULL, DECL edit_attrs, NULL, NULL
+},
+{ "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
+ EMPTY, NULL, NULL, DECL prompt_attrs, NULL
+},
+{ "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
+ DECL html_inline MODIFIER, NULL, DECL label_attrs , NULL, NULL
+},
+{ "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
+ DECL html_inline, NULL, DECL legend_attrs , DECL align_attr, NULL
+},
+{ "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
+ DECL html_flow, NULL, DECL html_attrs, NULL, NULL
+},
+{ "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
+ EMPTY, NULL, DECL link_attrs, DECL target_attr, NULL
+},
+{ "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
+ DECL map_contents , NULL, DECL html_attrs , NULL, DECL name_attr
+},
+{ "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
+ DECL blockli_elt , NULL, NULL, DECL compact_attrs, NULL
+},
+{ "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
+ EMPTY, NULL, DECL meta_attrs , NULL , DECL content_attr
+},
+{ "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
+ DECL noframes_content, "body" , DECL html_attrs, NULL, NULL
+},
+{ "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
+ DECL html_flow, "div", DECL html_attrs, NULL, NULL
+},
+{ "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
+ DECL object_contents , "div" , DECL object_attrs, DECL object_depr, NULL
+},
+{ "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
+ DECL li_elt , "li" , DECL html_attrs, DECL ol_attrs, NULL
+},
+{ "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
+ DECL option_elt , "option", DECL optgroup_attrs, NULL, DECL label_attr
+},
+{ "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
+ DECL html_pcdata, NULL, DECL option_attrs, NULL, NULL
+},
+{ "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
+ DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
+},
+{ "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
+ EMPTY, NULL, DECL param_attrs, NULL, DECL name_attr
+},
+{ "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
+ DECL pre_content, NULL, DECL html_attrs, DECL width_attr, NULL
+},
+{ "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
+ DECL html_inline, NULL, DECL quote_attrs, NULL, NULL
+},
+{ "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
+ DECL html_inline, NULL, NULL, DECL html_attrs, NULL
+},
+{ "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
+ DECL html_cdata, NULL, DECL script_attrs, DECL language_attr, DECL type_attr
+},
+{ "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
+ DECL select_content, NULL, DECL select_attrs, NULL, NULL
+},
+{ "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
+ DECL html_inline, NULL, NULL, DECL html_attrs, NULL
+},
+{ "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
+ DECL html_cdata, NULL, DECL style_attrs, NULL, DECL type_attr
+},
+{ "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "table", 0, 0, 0, 0, 0, 0, 0, "",
+ DECL table_contents , "tr" , DECL table_attrs , DECL table_depr, NULL
+},
+{ "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
+ DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
+},
+{ "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
+ DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
+},
+{ "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
+ DECL html_pcdata, NULL, DECL textarea_attrs, NULL, DECL rows_cols_attr
+},
+{ "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
+ DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
+},
+{ "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
+ DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
+},
+{ "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
+ DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
+},
+{ "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
+ DECL html_pcdata, NULL, DECL i18n_attrs, NULL, NULL
+},
+{ "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
+ DECL tr_contents , "td" , DECL talign_attrs, DECL bgcolor_attr, NULL
+},
+{ "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+},
+{ "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
+ DECL html_inline, NULL, NULL, DECL html_attrs, NULL
+},
+{ "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
+ DECL li_elt , "li" , DECL html_attrs, DECL ul_depr, NULL
+},
+{ "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
+ DECL html_inline, NULL, DECL html_attrs, NULL, NULL
+}
+};
+
+/*
+ * start tags that imply the end of current element
+ */
+static const char * const htmlStartClose[] = {
+"form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
+ "dl", "ul", "ol", "menu", "dir", "address", "pre",
+ "listing", "xmp", "head", NULL,
+"head", "p", NULL,
+"title", "p", NULL,
+"body", "head", "style", "link", "title", "p", NULL,
+"frameset", "head", "style", "link", "title", "p", NULL,
+"li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
+ "pre", "listing", "xmp", "head", "li", NULL,
+"hr", "p", "head", NULL,
+"h1", "p", "head", NULL,
+"h2", "p", "head", NULL,
+"h3", "p", "head", NULL,
+"h4", "p", "head", NULL,
+"h5", "p", "head", NULL,
+"h6", "p", "head", NULL,
+"dir", "p", "head", NULL,
+"address", "p", "head", "ul", NULL,
+"pre", "p", "head", "ul", NULL,
+"listing", "p", "head", NULL,
+"xmp", "p", "head", NULL,
+"blockquote", "p", "head", NULL,
+"dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
+ "xmp", "head", NULL,
+"dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
+ "head", "dd", NULL,
+"dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
+ "head", "dt", NULL,
+"ul", "p", "head", "ol", "menu", "dir", "address", "pre",
+ "listing", "xmp", NULL,
+"ol", "p", "head", "ul", NULL,
+"menu", "p", "head", "ul", NULL,
+"p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", FONTSTYLE, NULL,
+"div", "p", "head", NULL,
+"noscript", "script", NULL,
+"center", "font", "b", "i", "p", "head", NULL,
+"a", "a", "head", NULL,
+"caption", "p", NULL,
+"colgroup", "caption", "colgroup", "col", "p", NULL,
+"col", "caption", "col", "p", NULL,
+"table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
+ "listing", "xmp", "a", NULL,
+"th", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
+"td", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
+"tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
+"thead", "caption", "col", "colgroup", NULL,
+"tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
+ "tbody", "p", NULL,
+"tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
+ "tfoot", "tbody", "p", NULL,
+"optgroup", "option", NULL,
+"option", "option", NULL,
+"fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
+ "pre", "listing", "xmp", "a", NULL,
+/* most tags in in FONTSTYLE, PHRASE and SPECIAL should close <head> */
+"tt", "head", NULL,
+"i", "head", NULL,
+"b", "head", NULL,
+"u", "head", NULL,
+"s", "head", NULL,
+"strike", "head", NULL,
+"big", "head", NULL,
+"small", "head", NULL,
+
+"em", "head", NULL,
+"strong", "head", NULL,
+"dfn", "head", NULL,
+"code", "head", NULL,
+"samp", "head", NULL,
+"kbd", "head", NULL,
+"var", "head", NULL,
+"cite", "head", NULL,
+"abbr", "head", NULL,
+"acronym", "head", NULL,
+
+/* "a" */
+"img", "head", NULL,
+/* "applet" */
+/* "embed" */
+/* "object" */
+"font", "head", NULL,
+/* "basefont" */
+"br", "head", NULL,
+/* "script" */
+"map", "head", NULL,
+"q", "head", NULL,
+"sub", "head", NULL,
+"sup", "head", NULL,
+"span", "head", NULL,
+"bdo", "head", NULL,
+"iframe", "head", NULL,
+NULL
+};
+
+/*
+ * The list of HTML elements which are supposed not to have
+ * CDATA content and where a p element will be implied
+ *
+ * TODO: extend that list by reading the HTML SGML DTD on
+ * implied paragraph
+ */
+static const char *const htmlNoContentElements[] = {
+ "html",
+ "head",
+ NULL
+};
+
+/*
+ * The list of HTML attributes which are of content %Script;
+ * NOTE: when adding ones, check htmlIsScriptAttribute() since
+ * it assumes the name starts with 'on'
+ */
+static const char *const htmlScriptAttributes[] = {
+ "onclick",
+ "ondblclick",
+ "onmousedown",
+ "onmouseup",
+ "onmouseover",
+ "onmousemove",
+ "onmouseout",
+ "onkeypress",
+ "onkeydown",
+ "onkeyup",
+ "onload",
+ "onunload",
+ "onfocus",
+ "onblur",
+ "onsubmit",
+ "onreset",
+ "onchange",
+ "onselect"
+};
+
+/*
+ * This table is used by the htmlparser to know what to do with
+ * broken html pages. By assigning different priorities to different
+ * elements the parser can decide how to handle extra endtags.
+ * Endtags are only allowed to close elements with lower or equal
+ * priority.
+ */
+
+typedef struct {
+ const char *name;
+ int priority;
+} elementPriority;
+
+static const elementPriority htmlEndPriority[] = {
+ {"div", 150},
+ {"td", 160},
+ {"th", 160},
+ {"tr", 170},
+ {"thead", 180},
+ {"tbody", 180},
+ {"tfoot", 180},
+ {"table", 190},
+ {"head", 200},
+ {"body", 200},
+ {"html", 220},
+ {NULL, 100} /* Default priority */
+};
+
+static const char** htmlStartCloseIndex[100];
+static int htmlStartCloseIndexinitialized = 0;
+
+/************************************************************************
+ * *
+ * functions to handle HTML specific data *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlInitAutoClose:
+ *
+ * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
+ * This is not reentrant. Call xmlInitParser() once before processing in
+ * case of use in multithreaded programs.
+ */
+void
+htmlInitAutoClose(void) {
+ int indx, i = 0;
+
+ if (htmlStartCloseIndexinitialized) return;
+
+ for (indx = 0;indx < 100;indx ++) htmlStartCloseIndex[indx] = NULL;
+ indx = 0;
+ while ((htmlStartClose[i] != NULL) && (indx < 100 - 1)) {
+ htmlStartCloseIndex[indx++] = (const char**) &htmlStartClose[i];
+ while (htmlStartClose[i] != NULL) i++;
+ i++;
+ }
+ htmlStartCloseIndexinitialized = 1;
+}
+
+/**
+ * htmlTagLookup:
+ * @tag: The tag name in lowercase
+ *
+ * Lookup the HTML tag in the ElementTable
+ *
+ * Returns the related htmlElemDescPtr or NULL if not found.
+ */
+const htmlElemDesc *
+htmlTagLookup(const xmlChar *tag) {
+ unsigned int i;
+
+ for (i = 0; i < (sizeof(html40ElementTable) /
+ sizeof(html40ElementTable[0]));i++) {
+ if (!xmlStrcasecmp(tag, BAD_CAST html40ElementTable[i].name))
+ return((htmlElemDescPtr) &html40ElementTable[i]);
+ }
+ return(NULL);
+}
+
+/**
+ * htmlGetEndPriority:
+ * @name: The name of the element to look up the priority for.
+ *
+ * Return value: The "endtag" priority.
+ **/
+static int
+htmlGetEndPriority (const xmlChar *name) {
+ int i = 0;
+
+ while ((htmlEndPriority[i].name != NULL) &&
+ (!xmlStrEqual((const xmlChar *)htmlEndPriority[i].name, name)))
+ i++;
+
+ return(htmlEndPriority[i].priority);
+}
+
+
+/**
+ * htmlCheckAutoClose:
+ * @newtag: The new tag name
+ * @oldtag: The old tag name
+ *
+ * Checks whether the new tag is one of the registered valid tags for
+ * closing old.
+ * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
+ *
+ * Returns 0 if no, 1 if yes.
+ */
+static int
+htmlCheckAutoClose(const xmlChar * newtag, const xmlChar * oldtag)
+{
+ int i, indx;
+ const char **closed = NULL;
+
+ if (htmlStartCloseIndexinitialized == 0)
+ htmlInitAutoClose();
+
+ /* inefficient, but not a big deal */
+ for (indx = 0; indx < 100; indx++) {
+ closed = htmlStartCloseIndex[indx];
+ if (closed == NULL)
+ return (0);
+ if (xmlStrEqual(BAD_CAST * closed, newtag))
+ break;
+ }
+
+ i = closed - htmlStartClose;
+ i++;
+ while (htmlStartClose[i] != NULL) {
+ if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
+ return (1);
+ }
+ i++;
+ }
+ return (0);
+}
+
+/**
+ * htmlAutoCloseOnClose:
+ * @ctxt: an HTML parser context
+ * @newtag: The new tag name
+ * @force: force the tag closure
+ *
+ * The HTML DTD allows an ending tag to implicitly close other tags.
+ */
+static void
+htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
+{
+ const htmlElemDesc *info;
+ int i, priority;
+
+ priority = htmlGetEndPriority(newtag);
+
+ for (i = (ctxt->nameNr - 1); i >= 0; i--) {
+
+ if (xmlStrEqual(newtag, ctxt->nameTab[i]))
+ break;
+ /*
+ * A missplaced endtag can only close elements with lower
+ * or equal priority, so if we find an element with higher
+ * priority before we find an element with
+ * matching name, we just ignore this endtag
+ */
+ if (htmlGetEndPriority(ctxt->nameTab[i]) > priority)
+ return;
+ }
+ if (i < 0)
+ return;
+
+ while (!xmlStrEqual(newtag, ctxt->name)) {
+ info = htmlTagLookup(ctxt->name);
+ if ((info != NULL) && (info->endTag == 3)) {
+ htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
+ "Opening and ending tag mismatch: %s and %s\n",
+ newtag, ctxt->name);
+ }
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, ctxt->name);
+ htmlnamePop(ctxt);
+ }
+}
+
+/**
+ * htmlAutoCloseOnEnd:
+ * @ctxt: an HTML parser context
+ *
+ * Close all remaining tags at the end of the stream
+ */
+static void
+htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt)
+{
+ int i;
+
+ if (ctxt->nameNr == 0)
+ return;
+ for (i = (ctxt->nameNr - 1); i >= 0; i--) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, ctxt->name);
+ htmlnamePop(ctxt);
+ }
+}
+
+/**
+ * htmlAutoClose:
+ * @ctxt: an HTML parser context
+ * @newtag: The new tag name or NULL
+ *
+ * The HTML DTD allows a tag to implicitly close other tags.
+ * The list is kept in htmlStartClose array. This function is
+ * called when a new tag has been detected and generates the
+ * appropriates closes if possible/needed.
+ * If newtag is NULL this mean we are at the end of the resource
+ * and we should check
+ */
+static void
+htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
+{
+ while ((newtag != NULL) && (ctxt->name != NULL) &&
+ (htmlCheckAutoClose(newtag, ctxt->name))) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, ctxt->name);
+ htmlnamePop(ctxt);
+ }
+ if (newtag == NULL) {
+ htmlAutoCloseOnEnd(ctxt);
+ return;
+ }
+ while ((newtag == NULL) && (ctxt->name != NULL) &&
+ ((xmlStrEqual(ctxt->name, BAD_CAST "head")) ||
+ (xmlStrEqual(ctxt->name, BAD_CAST "body")) ||
+ (xmlStrEqual(ctxt->name, BAD_CAST "html")))) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, ctxt->name);
+ htmlnamePop(ctxt);
+ }
+}
+
+/**
+ * htmlAutoCloseTag:
+ * @doc: the HTML document
+ * @name: The tag name
+ * @elem: the HTML element
+ *
+ * The HTML DTD allows a tag to implicitly close other tags.
+ * The list is kept in htmlStartClose array. This function checks
+ * if the element or one of it's children would autoclose the
+ * given tag.
+ *
+ * Returns 1 if autoclose, 0 otherwise
+ */
+int
+htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
+ htmlNodePtr child;
+
+ if (elem == NULL) return(1);
+ if (xmlStrEqual(name, elem->name)) return(0);
+ if (htmlCheckAutoClose(elem->name, name)) return(1);
+ child = elem->children;
+ while (child != NULL) {
+ if (htmlAutoCloseTag(doc, name, child)) return(1);
+ child = child->next;
+ }
+ return(0);
+}
+
+/**
+ * htmlIsAutoClosed:
+ * @doc: the HTML document
+ * @elem: the HTML element
+ *
+ * The HTML DTD allows a tag to implicitly close other tags.
+ * The list is kept in htmlStartClose array. This function checks
+ * if a tag is autoclosed by one of it's child
+ *
+ * Returns 1 if autoclosed, 0 otherwise
+ */
+int
+htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
+ htmlNodePtr child;
+
+ if (elem == NULL) return(1);
+ child = elem->children;
+ while (child != NULL) {
+ if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
+ child = child->next;
+ }
+ return(0);
+}
+
+/**
+ * htmlCheckImplied:
+ * @ctxt: an HTML parser context
+ * @newtag: The new tag name
+ *
+ * The HTML DTD allows a tag to exists only implicitly
+ * called when a new tag has been detected and generates the
+ * appropriates implicit tags if missing
+ */
+static void
+htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
+ int i;
+
+ if (ctxt->options & HTML_PARSE_NOIMPLIED)
+ return;
+ if (!htmlOmittedDefaultValue)
+ return;
+ if (xmlStrEqual(newtag, BAD_CAST"html"))
+ return;
+ if (ctxt->nameNr <= 0) {
+ htmlnamePush(ctxt, BAD_CAST"html");
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+ ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
+ }
+ if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
+ return;
+ if ((ctxt->nameNr <= 1) &&
+ ((xmlStrEqual(newtag, BAD_CAST"script")) ||
+ (xmlStrEqual(newtag, BAD_CAST"style")) ||
+ (xmlStrEqual(newtag, BAD_CAST"meta")) ||
+ (xmlStrEqual(newtag, BAD_CAST"link")) ||
+ (xmlStrEqual(newtag, BAD_CAST"title")) ||
+ (xmlStrEqual(newtag, BAD_CAST"base")))) {
+ if (ctxt->html >= 3) {
+ /* we already saw or generated an <head> before */
+ return;
+ }
+ /*
+ * dropped OBJECT ... i you put it first BODY will be
+ * assumed !
+ */
+ htmlnamePush(ctxt, BAD_CAST"head");
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+ ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
+ } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) &&
+ (!xmlStrEqual(newtag, BAD_CAST"frame")) &&
+ (!xmlStrEqual(newtag, BAD_CAST"frameset"))) {
+ if (ctxt->html >= 10) {
+ /* we already saw or generated a <body> before */
+ return;
+ }
+ for (i = 0;i < ctxt->nameNr;i++) {
+ if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) {
+ return;
+ }
+ if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) {
+ return;
+ }
+ }
+
+ htmlnamePush(ctxt, BAD_CAST"body");
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+ ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
+ }
+}
+
+/**
+ * htmlCheckParagraph
+ * @ctxt: an HTML parser context
+ *
+ * Check whether a p element need to be implied before inserting
+ * characters in the current element.
+ *
+ * Returns 1 if a paragraph has been inserted, 0 if not and -1
+ * in case of error.
+ */
+
+static int
+htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
+ const xmlChar *tag;
+ int i;
+
+ if (ctxt == NULL)
+ return(-1);
+ tag = ctxt->name;
+ if (tag == NULL) {
+ htmlAutoClose(ctxt, BAD_CAST"p");
+ htmlCheckImplied(ctxt, BAD_CAST"p");
+ htmlnamePush(ctxt, BAD_CAST"p");
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+ ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
+ return(1);
+ }
+ if (!htmlOmittedDefaultValue)
+ return(0);
+ for (i = 0; htmlNoContentElements[i] != NULL; i++) {
+ if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
+ htmlAutoClose(ctxt, BAD_CAST"p");
+ htmlCheckImplied(ctxt, BAD_CAST"p");
+ htmlnamePush(ctxt, BAD_CAST"p");
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+ ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
+ return(1);
+ }
+ }
+ return(0);
+}
+
+/**
+ * htmlIsScriptAttribute:
+ * @name: an attribute name
+ *
+ * Check if an attribute is of content type Script
+ *
+ * Returns 1 is the attribute is a script 0 otherwise
+ */
+int
+htmlIsScriptAttribute(const xmlChar *name) {
+ unsigned int i;
+
+ if (name == NULL)
+ return(0);
+ /*
+ * all script attributes start with 'on'
+ */
+ if ((name[0] != 'o') || (name[1] != 'n'))
+ return(0);
+ for (i = 0;
+ i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
+ i++) {
+ if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
+ return(1);
+ }
+ return(0);
+}
+
+/************************************************************************
+ * *
+ * The list of HTML predefined entities *
+ * *
+ ************************************************************************/
+
+
+static const htmlEntityDesc html40EntitiesTable[] = {
+/*
+ * the 4 absolute ones, plus apostrophe.
+ */
+{ 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
+{ 38, "amp", "ampersand, U+0026 ISOnum" },
+{ 39, "apos", "single quote" },
+{ 60, "lt", "less-than sign, U+003C ISOnum" },
+{ 62, "gt", "greater-than sign, U+003E ISOnum" },
+
+/*
+ * A bunch still in the 128-255 range
+ * Replacing them depend really on the charset used.
+ */
+{ 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
+{ 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
+{ 162, "cent", "cent sign, U+00A2 ISOnum" },
+{ 163, "pound","pound sign, U+00A3 ISOnum" },
+{ 164, "curren","currency sign, U+00A4 ISOnum" },
+{ 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
+{ 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
+{ 167, "sect", "section sign, U+00A7 ISOnum" },
+{ 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
+{ 169, "copy", "copyright sign, U+00A9 ISOnum" },
+{ 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
+{ 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
+{ 172, "not", "not sign, U+00AC ISOnum" },
+{ 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
+{ 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
+{ 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
+{ 176, "deg", "degree sign, U+00B0 ISOnum" },
+{ 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
+{ 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
+{ 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
+{ 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
+{ 181, "micro","micro sign, U+00B5 ISOnum" },
+{ 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
+{ 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
+{ 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
+{ 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
+{ 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
+{ 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
+{ 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
+{ 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
+{ 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
+{ 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
+{ 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
+{ 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
+{ 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
+{ 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
+{ 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
+{ 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
+{ 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
+{ 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
+{ 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
+{ 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
+{ 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
+{ 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
+{ 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
+{ 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
+{ 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
+{ 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
+{ 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
+{ 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
+{ 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
+{ 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
+{ 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
+{ 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
+{ 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
+{ 215, "times","multiplication sign, U+00D7 ISOnum" },
+{ 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
+{ 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
+{ 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
+{ 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
+{ 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
+{ 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
+{ 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
+{ 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
+{ 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
+{ 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
+{ 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
+{ 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
+{ 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
+{ 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
+{ 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
+{ 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
+{ 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
+{ 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
+{ 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
+{ 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
+{ 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
+{ 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
+{ 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
+{ 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
+{ 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
+{ 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
+{ 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
+{ 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
+{ 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
+{ 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
+{ 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
+{ 247, "divide","division sign, U+00F7 ISOnum" },
+{ 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
+{ 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
+{ 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
+{ 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
+{ 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
+{ 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
+{ 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
+{ 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
+
+{ 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
+{ 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
+{ 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
+{ 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
+{ 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
+
+/*
+ * Anything below should really be kept as entities references
+ */
+{ 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
+
+{ 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
+{ 732, "tilde","small tilde, U+02DC ISOdia" },
+
+{ 913, "Alpha","greek capital letter alpha, U+0391" },
+{ 914, "Beta", "greek capital letter beta, U+0392" },
+{ 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
+{ 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
+{ 917, "Epsilon","greek capital letter epsilon, U+0395" },
+{ 918, "Zeta", "greek capital letter zeta, U+0396" },
+{ 919, "Eta", "greek capital letter eta, U+0397" },
+{ 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
+{ 921, "Iota", "greek capital letter iota, U+0399" },
+{ 922, "Kappa","greek capital letter kappa, U+039A" },
+{ 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
+{ 924, "Mu", "greek capital letter mu, U+039C" },
+{ 925, "Nu", "greek capital letter nu, U+039D" },
+{ 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
+{ 927, "Omicron","greek capital letter omicron, U+039F" },
+{ 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
+{ 929, "Rho", "greek capital letter rho, U+03A1" },
+{ 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
+{ 932, "Tau", "greek capital letter tau, U+03A4" },
+{ 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
+{ 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
+{ 935, "Chi", "greek capital letter chi, U+03A7" },
+{ 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
+{ 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
+
+{ 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
+{ 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
+{ 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
+{ 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
+{ 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
+{ 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
+{ 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
+{ 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
+{ 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
+{ 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
+{ 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
+{ 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
+{ 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
+{ 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
+{ 959, "omicron","greek small letter omicron, U+03BF NEW" },
+{ 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
+{ 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
+{ 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
+{ 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
+{ 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
+{ 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
+{ 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
+{ 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
+{ 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
+{ 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
+{ 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
+{ 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
+{ 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
+
+{ 8194, "ensp", "en space, U+2002 ISOpub" },
+{ 8195, "emsp", "em space, U+2003 ISOpub" },
+{ 8201, "thinsp","thin space, U+2009 ISOpub" },
+{ 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
+{ 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
+{ 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
+{ 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
+{ 8211, "ndash","en dash, U+2013 ISOpub" },
+{ 8212, "mdash","em dash, U+2014 ISOpub" },
+{ 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
+{ 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
+{ 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
+{ 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
+{ 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
+{ 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
+{ 8224, "dagger","dagger, U+2020 ISOpub" },
+{ 8225, "Dagger","double dagger, U+2021 ISOpub" },
+
+{ 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
+{ 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
+
+{ 8240, "permil","per mille sign, U+2030 ISOtech" },
+
+{ 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
+{ 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
+
+{ 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
+{ 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
+
+{ 8254, "oline","overline = spacing overscore, U+203E NEW" },
+{ 8260, "frasl","fraction slash, U+2044 NEW" },
+
+{ 8364, "euro", "euro sign, U+20AC NEW" },
+
+{ 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
+{ 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
+{ 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
+{ 8482, "trade","trade mark sign, U+2122 ISOnum" },
+{ 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
+{ 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
+{ 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
+{ 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
+{ 8595, "darr", "downwards arrow, U+2193 ISOnum" },
+{ 8596, "harr", "left right arrow, U+2194 ISOamsa" },
+{ 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
+{ 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
+{ 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
+{ 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
+{ 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
+{ 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
+
+{ 8704, "forall","for all, U+2200 ISOtech" },
+{ 8706, "part", "partial differential, U+2202 ISOtech" },
+{ 8707, "exist","there exists, U+2203 ISOtech" },
+{ 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
+{ 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
+{ 8712, "isin", "element of, U+2208 ISOtech" },
+{ 8713, "notin","not an element of, U+2209 ISOtech" },
+{ 8715, "ni", "contains as member, U+220B ISOtech" },
+{ 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
+{ 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
+{ 8722, "minus","minus sign, U+2212 ISOtech" },
+{ 8727, "lowast","asterisk operator, U+2217 ISOtech" },
+{ 8730, "radic","square root = radical sign, U+221A ISOtech" },
+{ 8733, "prop", "proportional to, U+221D ISOtech" },
+{ 8734, "infin","infinity, U+221E ISOtech" },
+{ 8736, "ang", "angle, U+2220 ISOamso" },
+{ 8743, "and", "logical and = wedge, U+2227 ISOtech" },
+{ 8744, "or", "logical or = vee, U+2228 ISOtech" },
+{ 8745, "cap", "intersection = cap, U+2229 ISOtech" },
+{ 8746, "cup", "union = cup, U+222A ISOtech" },
+{ 8747, "int", "integral, U+222B ISOtech" },
+{ 8756, "there4","therefore, U+2234 ISOtech" },
+{ 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
+{ 8773, "cong", "approximately equal to, U+2245 ISOtech" },
+{ 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
+{ 8800, "ne", "not equal to, U+2260 ISOtech" },
+{ 8801, "equiv","identical to, U+2261 ISOtech" },
+{ 8804, "le", "less-than or equal to, U+2264 ISOtech" },
+{ 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
+{ 8834, "sub", "subset of, U+2282 ISOtech" },
+{ 8835, "sup", "superset of, U+2283 ISOtech" },
+{ 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
+{ 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
+{ 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
+{ 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
+{ 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
+{ 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
+{ 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
+{ 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
+{ 8969, "rceil","right ceiling, U+2309 ISOamsc" },
+{ 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
+{ 8971, "rfloor","right floor, U+230B ISOamsc" },
+{ 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
+{ 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
+{ 9674, "loz", "lozenge, U+25CA ISOpub" },
+
+{ 9824, "spades","black spade suit, U+2660 ISOpub" },
+{ 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
+{ 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
+{ 9830, "diams","black diamond suit, U+2666 ISOpub" },
+
+};
+
+/************************************************************************
+ * *
+ * Commodity functions to handle entities *
+ * *
+ ************************************************************************/
+
+/*
+ * Macro used to grow the current buffer.
+ */
+#define growBuffer(buffer) { \
+ xmlChar *tmp; \
+ buffer##_size *= 2; \
+ tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
+ if (tmp == NULL) { \
+ htmlErrMemory(ctxt, "growing buffer\n"); \
+ xmlFree(buffer); \
+ return(NULL); \
+ } \
+ buffer = tmp; \
+}
+
+/**
+ * htmlEntityLookup:
+ * @name: the entity name
+ *
+ * Lookup the given entity in EntitiesTable
+ *
+ * TODO: the linear scan is really ugly, an hash table is really needed.
+ *
+ * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
+ */
+const htmlEntityDesc *
+htmlEntityLookup(const xmlChar *name) {
+ unsigned int i;
+
+ for (i = 0;i < (sizeof(html40EntitiesTable)/
+ sizeof(html40EntitiesTable[0]));i++) {
+ if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
+ return((htmlEntityDescPtr) &html40EntitiesTable[i]);
+ }
+ }
+ return(NULL);
+}
+
+/**
+ * htmlEntityValueLookup:
+ * @value: the entity's unicode value
+ *
+ * Lookup the given entity in EntitiesTable
+ *
+ * TODO: the linear scan is really ugly, an hash table is really needed.
+ *
+ * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
+ */
+const htmlEntityDesc *
+htmlEntityValueLookup(unsigned int value) {
+ unsigned int i;
+
+ for (i = 0;i < (sizeof(html40EntitiesTable)/
+ sizeof(html40EntitiesTable[0]));i++) {
+ if (html40EntitiesTable[i].value >= value) {
+ if (html40EntitiesTable[i].value > value)
+ break;
+ return((htmlEntityDescPtr) &html40EntitiesTable[i]);
+ }
+ }
+ return(NULL);
+}
+
+/**
+ * UTF8ToHtml:
+ * @out: a pointer to an array of bytes to store the result
+ * @outlen: the length of @out
+ * @in: a pointer to an array of UTF-8 chars
+ * @inlen: the length of @in
+ *
+ * Take a block of UTF-8 chars in and try to convert it to an ASCII
+ * plus HTML entities block of chars out.
+ *
+ * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
+ * The value of @inlen after return is the number of octets consumed
+ * as the return value is positive, else unpredictable.
+ * The value of @outlen after return is the number of octets consumed.
+ */
+int
+UTF8ToHtml(unsigned char* out, int *outlen,
+ const unsigned char* in, int *inlen) {
+ const unsigned char* processed = in;
+ const unsigned char* outend;
+ const unsigned char* outstart = out;
+ const unsigned char* instart = in;
+ const unsigned char* inend;
+ unsigned int c, d;
+ int trailing;
+
+ if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
+ if (in == NULL) {
+ /*
+ * initialization nothing to do
+ */
+ *outlen = 0;
+ *inlen = 0;
+ return(0);
+ }
+ inend = in + (*inlen);
+ outend = out + (*outlen);
+ while (in < inend) {
+ d = *in++;
+ if (d < 0x80) { c= d; trailing= 0; }
+ else if (d < 0xC0) {
+ /* trailing byte in leading position */
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(-2);
+ } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
+ else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
+ else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
+ else {
+ /* no chance for this in Ascii */
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(-2);
+ }
+
+ if (inend - in < trailing) {
+ break;
+ }
+
+ for ( ; trailing; trailing--) {
+ if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
+ break;
+ c <<= 6;
+ c |= d & 0x3F;
+ }
+
+ /* assertion: c is a single UTF-4 value */
+ if (c < 0x80) {
+ if (out + 1 >= outend)
+ break;
+ *out++ = c;
+ } else {
+ int len;
+ const htmlEntityDesc * ent;
+ const char *cp;
+ char nbuf[16];
+
+ /*
+ * Try to lookup a predefined HTML entity for it
+ */
+
+ ent = htmlEntityValueLookup(c);
+ if (ent == NULL) {
+ snprintf(nbuf, sizeof(nbuf), "#%u", c);
+ cp = nbuf;
+ }
+ else
+ cp = ent->name;
+ len = strlen(cp);
+ if (out + 2 + len >= outend)
+ break;
+ *out++ = '&';
+ memcpy(out, cp, len);
+ out += len;
+ *out++ = ';';
+ }
+ processed = in;
+ }
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(0);
+}
+
+/**
+ * htmlEncodeEntities:
+ * @out: a pointer to an array of bytes to store the result
+ * @outlen: the length of @out
+ * @in: a pointer to an array of UTF-8 chars
+ * @inlen: the length of @in
+ * @quoteChar: the quote character to escape (' or ") or zero.
+ *
+ * Take a block of UTF-8 chars in and try to convert it to an ASCII
+ * plus HTML entities block of chars out.
+ *
+ * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
+ * The value of @inlen after return is the number of octets consumed
+ * as the return value is positive, else unpredictable.
+ * The value of @outlen after return is the number of octets consumed.
+ */
+int
+htmlEncodeEntities(unsigned char* out, int *outlen,
+ const unsigned char* in, int *inlen, int quoteChar) {
+ const unsigned char* processed = in;
+ const unsigned char* outend;
+ const unsigned char* outstart = out;
+ const unsigned char* instart = in;
+ const unsigned char* inend;
+ unsigned int c, d;
+ int trailing;
+
+ if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL))
+ return(-1);
+ outend = out + (*outlen);
+ inend = in + (*inlen);
+ while (in < inend) {
+ d = *in++;
+ if (d < 0x80) { c= d; trailing= 0; }
+ else if (d < 0xC0) {
+ /* trailing byte in leading position */
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(-2);
+ } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
+ else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
+ else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
+ else {
+ /* no chance for this in Ascii */
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(-2);
+ }
+
+ if (inend - in < trailing)
+ break;
+
+ while (trailing--) {
+ if (((d= *in++) & 0xC0) != 0x80) {
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(-2);
+ }
+ c <<= 6;
+ c |= d & 0x3F;
+ }
+
+ /* assertion: c is a single UTF-4 value */
+ if ((c < 0x80) && (c != (unsigned int) quoteChar) &&
+ (c != '&') && (c != '<') && (c != '>')) {
+ if (out >= outend)
+ break;
+ *out++ = c;
+ } else {
+ const htmlEntityDesc * ent;
+ const char *cp;
+ char nbuf[16];
+ int len;
+
+ /*
+ * Try to lookup a predefined HTML entity for it
+ */
+ ent = htmlEntityValueLookup(c);
+ if (ent == NULL) {
+ snprintf(nbuf, sizeof(nbuf), "#%u", c);
+ cp = nbuf;
+ }
+ else
+ cp = ent->name;
+ len = strlen(cp);
+ if (out + 2 + len > outend)
+ break;
+ *out++ = '&';
+ memcpy(out, cp, len);
+ out += len;
+ *out++ = ';';
+ }
+ processed = in;
+ }
+ *outlen = out - outstart;
+ *inlen = processed - instart;
+ return(0);
+}
+
+/************************************************************************
+ * *
+ * Commodity functions to handle streams *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlNewInputStream:
+ * @ctxt: an HTML parser context
+ *
+ * Create a new input stream structure
+ * Returns the new input stream or NULL
+ */
+static htmlParserInputPtr
+htmlNewInputStream(htmlParserCtxtPtr ctxt) {
+ htmlParserInputPtr input;
+
+ input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
+ if (input == NULL) {
+ htmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
+ return(NULL);
+ }
+ memset(input, 0, sizeof(htmlParserInput));
+ input->filename = NULL;
+ input->directory = NULL;
+ input->base = NULL;
+ input->cur = NULL;
+ input->buf = NULL;
+ input->line = 1;
+ input->col = 1;
+ input->buf = NULL;
+ input->free = NULL;
+ input->version = NULL;
+ input->consumed = 0;
+ input->length = 0;
+ return(input);
+}
+
+
+/************************************************************************
+ * *
+ * Commodity functions, cleanup needed ? *
+ * *
+ ************************************************************************/
+/*
+ * all tags allowing pc data from the html 4.01 loose dtd
+ * NOTE: it might be more apropriate to integrate this information
+ * into the html40ElementTable array but I don't want to risk any
+ * binary incomptibility
+ */
+static const char *allowPCData[] = {
+ "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
+ "blockquote", "body", "button", "caption", "center", "cite", "code",
+ "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
+ "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
+ "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
+ "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
+};
+
+/**
+ * areBlanks:
+ * @ctxt: an HTML parser context
+ * @str: a xmlChar *
+ * @len: the size of @str
+ *
+ * Is this a sequence of blank chars that one can ignore ?
+ *
+ * Returns 1 if ignorable 0 otherwise.
+ */
+
+static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
+ unsigned int i;
+ int j;
+ xmlNodePtr lastChild;
+ xmlDtdPtr dtd;
+
+ for (j = 0;j < len;j++)
+ if (!(IS_BLANK_CH(str[j]))) return(0);
+
+ if (CUR == 0) return(1);
+ if (CUR != '<') return(0);
+ if (ctxt->name == NULL)
+ return(1);
+ if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
+ return(1);
+ if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
+ return(1);
+
+ /* Only strip CDATA children of the body tag for strict HTML DTDs */
+ if (xmlStrEqual(ctxt->name, BAD_CAST "body") && ctxt->myDoc != NULL) {
+ dtd = xmlGetIntSubset(ctxt->myDoc);
+ if (dtd != NULL && dtd->ExternalID != NULL) {
+ if (!xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4.01//EN") ||
+ !xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4//EN"))
+ return(1);
+ }
+ }
+
+ if (ctxt->node == NULL) return(0);
+ lastChild = xmlGetLastChild(ctxt->node);
+ while ((lastChild) && (lastChild->type == XML_COMMENT_NODE))
+ lastChild = lastChild->prev;
+ if (lastChild == NULL) {
+ if ((ctxt->node->type != XML_ELEMENT_NODE) &&
+ (ctxt->node->content != NULL)) return(0);
+ /* keep ws in constructs like ...<b> </b>...
+ for all tags "b" allowing PCDATA */
+ for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
+ if ( xmlStrEqual(ctxt->name, BAD_CAST allowPCData[i]) ) {
+ return(0);
+ }
+ }
+ } else if (xmlNodeIsText(lastChild)) {
+ return(0);
+ } else {
+ /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
+ for all tags "p" allowing PCDATA */
+ for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
+ if ( xmlStrEqual(lastChild->name, BAD_CAST allowPCData[i]) ) {
+ return(0);
+ }
+ }
+ }
+ return(1);
+}
+
+/**
+ * htmlNewDocNoDtD:
+ * @URI: URI for the dtd, or NULL
+ * @ExternalID: the external ID of the DTD, or NULL
+ *
+ * Creates a new HTML document without a DTD node if @URI and @ExternalID
+ * are NULL
+ *
+ * Returns a new document, do not initialize the DTD if not provided
+ */
+htmlDocPtr
+htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
+ xmlDocPtr cur;
+
+ /*
+ * Allocate a new document and fill the fields.
+ */
+ cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
+ if (cur == NULL) {
+ htmlErrMemory(NULL, "HTML document creation failed\n");
+ return(NULL);
+ }
+ memset(cur, 0, sizeof(xmlDoc));
+
+ cur->type = XML_HTML_DOCUMENT_NODE;
+ cur->version = NULL;
+ cur->intSubset = NULL;
+ cur->doc = cur;
+ cur->name = NULL;
+ cur->children = NULL;
+ cur->extSubset = NULL;
+ cur->oldNs = NULL;
+ cur->encoding = NULL;
+ cur->standalone = 1;
+ cur->compression = 0;
+ cur->ids = NULL;
+ cur->refs = NULL;
+ cur->_private = NULL;
+ cur->charset = XML_CHAR_ENCODING_UTF8;
+ cur->properties = XML_DOC_HTML | XML_DOC_USERBUILT;
+ if ((ExternalID != NULL) ||
+ (URI != NULL))
+ xmlCreateIntSubset(cur, BAD_CAST "html", ExternalID, URI);
+ return(cur);
+}
+
+/**
+ * htmlNewDoc:
+ * @URI: URI for the dtd, or NULL
+ * @ExternalID: the external ID of the DTD, or NULL
+ *
+ * Creates a new HTML document
+ *
+ * Returns a new document
+ */
+htmlDocPtr
+htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
+ if ((URI == NULL) && (ExternalID == NULL))
+ return(htmlNewDocNoDtD(
+ BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd",
+ BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN"));
+
+ return(htmlNewDocNoDtD(URI, ExternalID));
+}
+
+
+/************************************************************************
+ * *
+ * The parser itself *
+ * Relates to http://www.w3.org/TR/html40 *
+ * *
+ ************************************************************************/
+
+/************************************************************************
+ * *
+ * The parser itself *
+ * *
+ ************************************************************************/
+
+static const xmlChar * htmlParseNameComplex(xmlParserCtxtPtr ctxt);
+
+/**
+ * htmlParseHTMLName:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML tag or attribute name, note that we convert it to lowercase
+ * since HTML names are not case-sensitive.
+ *
+ * Returns the Tag Name parsed or NULL
+ */
+
+static const xmlChar *
+htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
+ int i = 0;
+ xmlChar loc[HTML_PARSER_BUFFER_SIZE];
+
+ if (!IS_ASCII_LETTER(CUR) && (CUR != '_') &&
+ (CUR != ':') && (CUR != '.')) return(NULL);
+
+ while ((i < HTML_PARSER_BUFFER_SIZE) &&
+ ((IS_ASCII_LETTER(CUR)) || (IS_ASCII_DIGIT(CUR)) ||
+ (CUR == ':') || (CUR == '-') || (CUR == '_') ||
+ (CUR == '.'))) {
+ if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
+ else loc[i] = CUR;
+ i++;
+
+ NEXT;
+ }
+
+ return(xmlDictLookup(ctxt->dict, loc, i));
+}
+
+
+/**
+ * htmlParseHTMLName_nonInvasive:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML tag or attribute name, note that we convert it to lowercase
+ * since HTML names are not case-sensitive, this doesn't consume the data
+ * from the stream, it's a look-ahead
+ *
+ * Returns the Tag Name parsed or NULL
+ */
+
+static const xmlChar *
+htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt) {
+ int i = 0;
+ xmlChar loc[HTML_PARSER_BUFFER_SIZE];
+
+ if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
+ (NXT(1) != ':')) return(NULL);
+
+ while ((i < HTML_PARSER_BUFFER_SIZE) &&
+ ((IS_ASCII_LETTER(NXT(1+i))) || (IS_ASCII_DIGIT(NXT(1+i))) ||
+ (NXT(1+i) == ':') || (NXT(1+i) == '-') || (NXT(1+i) == '_'))) {
+ if ((NXT(1+i) >= 'A') && (NXT(1+i) <= 'Z')) loc[i] = NXT(1+i) + 0x20;
+ else loc[i] = NXT(1+i);
+ i++;
+ }
+
+ return(xmlDictLookup(ctxt->dict, loc, i));
+}
+
+
+/**
+ * htmlParseName:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML name, this routine is case sensitive.
+ *
+ * Returns the Name parsed or NULL
+ */
+
+static const xmlChar *
+htmlParseName(htmlParserCtxtPtr ctxt) {
+ const xmlChar *in;
+ const xmlChar *ret;
+ int count = 0;
+
+ GROW;
+
+ /*
+ * Accelerator for simple ASCII names
+ */
+ in = ctxt->input->cur;
+ if (((*in >= 0x61) && (*in <= 0x7A)) ||
+ ((*in >= 0x41) && (*in <= 0x5A)) ||
+ (*in == '_') || (*in == ':')) {
+ in++;
+ while (((*in >= 0x61) && (*in <= 0x7A)) ||
+ ((*in >= 0x41) && (*in <= 0x5A)) ||
+ ((*in >= 0x30) && (*in <= 0x39)) ||
+ (*in == '_') || (*in == '-') ||
+ (*in == ':') || (*in == '.'))
+ in++;
+
+ if (in == ctxt->input->end)
+ return(NULL);
+
+ if ((*in > 0) && (*in < 0x80)) {
+ count = in - ctxt->input->cur;
+ ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count);
+ ctxt->input->cur = in;
+ ctxt->nbChars += count;
+ ctxt->input->col += count;
+ return(ret);
+ }
+ }
+ return(htmlParseNameComplex(ctxt));
+}
+
+static const xmlChar *
+htmlParseNameComplex(xmlParserCtxtPtr ctxt) {
+ int len = 0, l;
+ int c;
+ int count = 0;
+ const xmlChar *base = ctxt->input->base;
+
+ /*
+ * Handler for more complex cases
+ */
+ GROW;
+ c = CUR_CHAR(l);
+ if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
+ (!IS_LETTER(c) && (c != '_') &&
+ (c != ':'))) {
+ return(NULL);
+ }
+
+ while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */
+ ((IS_LETTER(c)) || (IS_DIGIT(c)) ||
+ (c == '.') || (c == '-') ||
+ (c == '_') || (c == ':') ||
+ (IS_COMBINING(c)) ||
+ (IS_EXTENDER(c)))) {
+ if (count++ > 100) {
+ count = 0;
+ GROW;
+ }
+ len += l;
+ NEXTL(l);
+ c = CUR_CHAR(l);
+ if (ctxt->input->base != base) {
+ /*
+ * We changed encoding from an unknown encoding
+ * Input buffer changed location, so we better start again
+ */
+ return(htmlParseNameComplex(ctxt));
+ }
+ }
+
+ if (ctxt->input->cur - ctxt->input->base < len) {
+ /* Sanity check */
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "unexpected change of input buffer", NULL, NULL);
+ return (NULL);
+ }
+
+ return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
+}
+
+
+/**
+ * htmlParseHTMLAttribute:
+ * @ctxt: an HTML parser context
+ * @stop: a char stop value
+ *
+ * parse an HTML attribute value till the stop (quote), if
+ * stop is 0 then it stops at the first space
+ *
+ * Returns the attribute parsed or NULL
+ */
+
+static xmlChar *
+htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
+ xmlChar *buffer = NULL;
+ int buffer_size = 0;
+ xmlChar *out = NULL;
+ const xmlChar *name = NULL;
+ const xmlChar *cur = NULL;
+ const htmlEntityDesc * ent;
+
+ /*
+ * allocate a translation buffer.
+ */
+ buffer_size = HTML_PARSER_BUFFER_SIZE;
+ buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
+ if (buffer == NULL) {
+ htmlErrMemory(ctxt, "buffer allocation failed\n");
+ return(NULL);
+ }
+ out = buffer;
+
+ /*
+ * Ok loop until we reach one of the ending chars
+ */
+ while ((CUR != 0) && (CUR != stop)) {
+ if ((stop == 0) && (CUR == '>')) break;
+ if ((stop == 0) && (IS_BLANK_CH(CUR))) break;
+ if (CUR == '&') {
+ if (NXT(1) == '#') {
+ unsigned int c;
+ int bits;
+
+ c = htmlParseCharRef(ctxt);
+ if (c < 0x80)
+ { *out++ = c; bits= -6; }
+ else if (c < 0x800)
+ { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
+ else if (c < 0x10000)
+ { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
+ else
+ { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
+
+ for ( ; bits >= 0; bits-= 6) {
+ *out++ = ((c >> bits) & 0x3F) | 0x80;
+ }
+
+ if (out - buffer > buffer_size - 100) {
+ int indx = out - buffer;
+
+ growBuffer(buffer);
+ out = &buffer[indx];
+ }
+ } else {
+ ent = htmlParseEntityRef(ctxt, &name);
+ if (name == NULL) {
+ *out++ = '&';
+ if (out - buffer > buffer_size - 100) {
+ int indx = out - buffer;
+
+ growBuffer(buffer);
+ out = &buffer[indx];
+ }
+ } else if (ent == NULL) {
+ *out++ = '&';
+ cur = name;
+ while (*cur != 0) {
+ if (out - buffer > buffer_size - 100) {
+ int indx = out - buffer;
+
+ growBuffer(buffer);
+ out = &buffer[indx];
+ }
+ *out++ = *cur++;
+ }
+ } else {
+ unsigned int c;
+ int bits;
+
+ if (out - buffer > buffer_size - 100) {
+ int indx = out - buffer;
+
+ growBuffer(buffer);
+ out = &buffer[indx];
+ }
+ c = ent->value;
+ if (c < 0x80)
+ { *out++ = c; bits= -6; }
+ else if (c < 0x800)
+ { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
+ else if (c < 0x10000)
+ { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
+ else
+ { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
+
+ for ( ; bits >= 0; bits-= 6) {
+ *out++ = ((c >> bits) & 0x3F) | 0x80;
+ }
+ }
+ }
+ } else {
+ unsigned int c;
+ int bits, l;
+
+ if (out - buffer > buffer_size - 100) {
+ int indx = out - buffer;
+
+ growBuffer(buffer);
+ out = &buffer[indx];
+ }
+ c = CUR_CHAR(l);
+ if (c < 0x80)
+ { *out++ = c; bits= -6; }
+ else if (c < 0x800)
+ { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
+ else if (c < 0x10000)
+ { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
+ else
+ { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
+
+ for ( ; bits >= 0; bits-= 6) {
+ *out++ = ((c >> bits) & 0x3F) | 0x80;
+ }
+ NEXT;
+ }
+ }
+ *out = 0;
+ return(buffer);
+}
+
+/**
+ * htmlParseEntityRef:
+ * @ctxt: an HTML parser context
+ * @str: location to store the entity name
+ *
+ * parse an HTML ENTITY references
+ *
+ * [68] EntityRef ::= '&' Name ';'
+ *
+ * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
+ * if non-NULL *str will have to be freed by the caller.
+ */
+const htmlEntityDesc *
+htmlParseEntityRef(htmlParserCtxtPtr ctxt, const xmlChar **str) {
+ const xmlChar *name;
+ const htmlEntityDesc * ent = NULL;
+
+ if (str != NULL) *str = NULL;
+ if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL);
+
+ if (CUR == '&') {
+ NEXT;
+ name = htmlParseName(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "htmlParseEntityRef: no name\n", NULL, NULL);
+ } else {
+ GROW;
+ if (CUR == ';') {
+ if (str != NULL)
+ *str = name;
+
+ /*
+ * Lookup the entity in the table.
+ */
+ ent = htmlEntityLookup(name);
+ if (ent != NULL) /* OK that's ugly !!! */
+ NEXT;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING,
+ "htmlParseEntityRef: expecting ';'\n",
+ NULL, NULL);
+ if (str != NULL)
+ *str = name;
+ }
+ }
+ }
+ return(ent);
+}
+
+/**
+ * htmlParseAttValue:
+ * @ctxt: an HTML parser context
+ *
+ * parse a value for an attribute
+ * Note: the parser won't do substitution of entities here, this
+ * will be handled later in xmlStringGetNodeList, unless it was
+ * asked for ctxt->replaceEntities != 0
+ *
+ * Returns the AttValue parsed or NULL.
+ */
+
+static xmlChar *
+htmlParseAttValue(htmlParserCtxtPtr ctxt) {
+ xmlChar *ret = NULL;
+
+ if (CUR == '"') {
+ NEXT;
+ ret = htmlParseHTMLAttribute(ctxt, '"');
+ if (CUR != '"') {
+ htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
+ "AttValue: \" expected\n", NULL, NULL);
+ } else
+ NEXT;
+ } else if (CUR == '\'') {
+ NEXT;
+ ret = htmlParseHTMLAttribute(ctxt, '\'');
+ if (CUR != '\'') {
+ htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
+ "AttValue: ' expected\n", NULL, NULL);
+ } else
+ NEXT;
+ } else {
+ /*
+ * That's an HTMLism, the attribute value may not be quoted
+ */
+ ret = htmlParseHTMLAttribute(ctxt, 0);
+ if (ret == NULL) {
+ htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE,
+ "AttValue: no value found\n", NULL, NULL);
+ }
+ }
+ return(ret);
+}
+
+/**
+ * htmlParseSystemLiteral:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML Literal
+ *
+ * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
+ *
+ * Returns the SystemLiteral parsed or NULL
+ */
+
+static xmlChar *
+htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
+ size_t len = 0, startPosition = 0;
+ xmlChar *ret = NULL;
+
+ if (CUR == '"') {
+ NEXT;
+
+ if (CUR_PTR < BASE_PTR)
+ return(ret);
+ startPosition = CUR_PTR - BASE_PTR;
+
+ while ((IS_CHAR_CH(CUR)) && (CUR != '"')) {
+ NEXT;
+ len++;
+ }
+ if (!IS_CHAR_CH(CUR)) {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
+ "Unfinished SystemLiteral\n", NULL, NULL);
+ } else {
+ ret = xmlStrndup((BASE_PTR+startPosition), len);
+ NEXT;
+ }
+ } else if (CUR == '\'') {
+ NEXT;
+
+ if (CUR_PTR < BASE_PTR)
+ return(ret);
+ startPosition = CUR_PTR - BASE_PTR;
+
+ while ((IS_CHAR_CH(CUR)) && (CUR != '\'')) {
+ NEXT;
+ len++;
+ }
+ if (!IS_CHAR_CH(CUR)) {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
+ "Unfinished SystemLiteral\n", NULL, NULL);
+ } else {
+ ret = xmlStrndup((BASE_PTR+startPosition), len);
+ NEXT;
+ }
+ } else {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
+ " or ' expected\n", NULL, NULL);
+ }
+
+ return(ret);
+}
+
+/**
+ * htmlParsePubidLiteral:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML public literal
+ *
+ * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
+ *
+ * Returns the PubidLiteral parsed or NULL.
+ */
+
+static xmlChar *
+htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
+ size_t len = 0, startPosition = 0;
+ xmlChar *ret = NULL;
+ /*
+ * Name ::= (Letter | '_') (NameChar)*
+ */
+ if (CUR == '"') {
+ NEXT;
+
+ if (CUR_PTR < BASE_PTR)
+ return(ret);
+ startPosition = CUR_PTR - BASE_PTR;
+
+ while (IS_PUBIDCHAR_CH(CUR)) {
+ len++;
+ NEXT;
+ }
+
+ if (CUR != '"') {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
+ "Unfinished PubidLiteral\n", NULL, NULL);
+ } else {
+ ret = xmlStrndup((BASE_PTR + startPosition), len);
+ NEXT;
+ }
+ } else if (CUR == '\'') {
+ NEXT;
+
+ if (CUR_PTR < BASE_PTR)
+ return(ret);
+ startPosition = CUR_PTR - BASE_PTR;
+
+ while ((IS_PUBIDCHAR_CH(CUR)) && (CUR != '\'')){
+ len++;
+ NEXT;
+ }
+
+ if (CUR != '\'') {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
+ "Unfinished PubidLiteral\n", NULL, NULL);
+ } else {
+ ret = xmlStrndup((BASE_PTR + startPosition), len);
+ NEXT;
+ }
+ } else {
+ htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
+ "PubidLiteral \" or ' expected\n", NULL, NULL);
+ }
+
+ return(ret);
+}
+
+/**
+ * htmlParseScript:
+ * @ctxt: an HTML parser context
+ *
+ * parse the content of an HTML SCRIPT or STYLE element
+ * http://www.w3.org/TR/html4/sgml/dtd.html#Script
+ * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
+ * http://www.w3.org/TR/html4/types.html#type-script
+ * http://www.w3.org/TR/html4/types.html#h-6.15
+ * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
+ *
+ * Script data ( %Script; in the DTD) can be the content of the SCRIPT
+ * element and the value of intrinsic event attributes. User agents must
+ * not evaluate script data as HTML markup but instead must pass it on as
+ * data to a script engine.
+ * NOTES:
+ * - The content is passed like CDATA
+ * - the attributes for style and scripting "onXXX" are also described
+ * as CDATA but SGML allows entities references in attributes so their
+ * processing is identical as other attributes
+ */
+static void
+htmlParseScript(htmlParserCtxtPtr ctxt) {
+ xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
+ int nbchar = 0;
+ int cur,l;
+
+ SHRINK;
+ cur = CUR_CHAR(l);
+ while (IS_CHAR_CH(cur)) {
+ if ((cur == '<') && (NXT(1) == '/')) {
+ /*
+ * One should break here, the specification is clear:
+ * Authors should therefore escape "</" within the content.
+ * Escape mechanisms are specific to each scripting or
+ * style sheet language.
+ *
+ * In recovery mode, only break if end tag match the
+ * current tag, effectively ignoring all tags inside the
+ * script/style block and treating the entire block as
+ * CDATA.
+ */
+ if (ctxt->recovery) {
+ if (xmlStrncasecmp(ctxt->name, ctxt->input->cur+2,
+ xmlStrlen(ctxt->name)) == 0)
+ {
+ break; /* while */
+ } else {
+ htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
+ "Element %s embeds close tag\n",
+ ctxt->name, NULL);
+ }
+ } else {
+ if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
+ ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
+ {
+ break; /* while */
+ }
+ }
+ }
+ COPY_BUF(l,buf,nbchar,cur);
+ if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
+ if (ctxt->sax->cdataBlock!= NULL) {
+ /*
+ * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
+ */
+ ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
+ } else if (ctxt->sax->characters != NULL) {
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ }
+ nbchar = 0;
+ }
+ GROW;
+ NEXTL(l);
+ cur = CUR_CHAR(l);
+ }
+
+ if ((!(IS_CHAR_CH(cur))) && (!((cur == 0) && (ctxt->progressive)))) {
+ htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
+ "Invalid char in CDATA 0x%X\n", cur);
+ if (ctxt->input->cur < ctxt->input->end) {
+ NEXT;
+ }
+ }
+
+ if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
+ if (ctxt->sax->cdataBlock!= NULL) {
+ /*
+ * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
+ */
+ ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
+ } else if (ctxt->sax->characters != NULL) {
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ }
+ }
+}
+
+
+/**
+ * htmlParseCharDataInternal:
+ * @ctxt: an HTML parser context
+ * @readahead: optional read ahead character in ascii range
+ *
+ * parse a CharData section.
+ * if we are within a CDATA section ']]>' marks an end of section.
+ *
+ * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
+ */
+
+static void
+htmlParseCharDataInternal(htmlParserCtxtPtr ctxt, int readahead) {
+ xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 6];
+ int nbchar = 0;
+ int cur, l;
+ int chunk = 0;
+
+ if (readahead)
+ buf[nbchar++] = readahead;
+
+ SHRINK;
+ cur = CUR_CHAR(l);
+ while (((cur != '<') || (ctxt->token == '<')) &&
+ ((cur != '&') || (ctxt->token == '&')) &&
+ (cur != 0)) {
+ if (!(IS_CHAR(cur))) {
+ htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
+ "Invalid char in CDATA 0x%X\n", cur);
+ } else {
+ COPY_BUF(l,buf,nbchar,cur);
+ }
+ if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
+ /*
+ * Ok the segment is to be consumed as chars.
+ */
+ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
+ if (areBlanks(ctxt, buf, nbchar)) {
+ if (ctxt->keepBlanks) {
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ } else {
+ if (ctxt->sax->ignorableWhitespace != NULL)
+ ctxt->sax->ignorableWhitespace(ctxt->userData,
+ buf, nbchar);
+ }
+ } else {
+ htmlCheckParagraph(ctxt);
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ }
+ }
+ nbchar = 0;
+ }
+ NEXTL(l);
+ chunk++;
+ if (chunk > HTML_PARSER_BUFFER_SIZE) {
+ chunk = 0;
+ SHRINK;
+ GROW;
+ }
+ cur = CUR_CHAR(l);
+ if (cur == 0) {
+ SHRINK;
+ GROW;
+ cur = CUR_CHAR(l);
+ }
+ }
+ if (nbchar != 0) {
+ buf[nbchar] = 0;
+
+ /*
+ * Ok the segment is to be consumed as chars.
+ */
+ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
+ if (areBlanks(ctxt, buf, nbchar)) {
+ if (ctxt->keepBlanks) {
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ } else {
+ if (ctxt->sax->ignorableWhitespace != NULL)
+ ctxt->sax->ignorableWhitespace(ctxt->userData,
+ buf, nbchar);
+ }
+ } else {
+ htmlCheckParagraph(ctxt);
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(ctxt->userData, buf, nbchar);
+ }
+ }
+ } else {
+ /*
+ * Loop detection
+ */
+ if (cur == 0)
+ ctxt->instate = XML_PARSER_EOF;
+ }
+}
+
+/**
+ * htmlParseCharData:
+ * @ctxt: an HTML parser context
+ *
+ * parse a CharData section.
+ * if we are within a CDATA section ']]>' marks an end of section.
+ *
+ * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
+ */
+
+static void
+htmlParseCharData(htmlParserCtxtPtr ctxt) {
+ htmlParseCharDataInternal(ctxt, 0);
+}
+
+/**
+ * htmlParseExternalID:
+ * @ctxt: an HTML parser context
+ * @publicID: a xmlChar** receiving PubidLiteral
+ *
+ * Parse an External ID or a Public ID
+ *
+ * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
+ * | 'PUBLIC' S PubidLiteral S SystemLiteral
+ *
+ * [83] PublicID ::= 'PUBLIC' S PubidLiteral
+ *
+ * Returns the function returns SystemLiteral and in the second
+ * case publicID receives PubidLiteral, is strict is off
+ * it is possible to return NULL and have publicID set.
+ */
+
+static xmlChar *
+htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) {
+ xmlChar *URI = NULL;
+
+ if ((UPPER == 'S') && (UPP(1) == 'Y') &&
+ (UPP(2) == 'S') && (UPP(3) == 'T') &&
+ (UPP(4) == 'E') && (UPP(5) == 'M')) {
+ SKIP(6);
+ if (!IS_BLANK_CH(CUR)) {
+ htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
+ "Space required after 'SYSTEM'\n", NULL, NULL);
+ }
+ SKIP_BLANKS;
+ URI = htmlParseSystemLiteral(ctxt);
+ if (URI == NULL) {
+ htmlParseErr(ctxt, XML_ERR_URI_REQUIRED,
+ "htmlParseExternalID: SYSTEM, no URI\n", NULL, NULL);
+ }
+ } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
+ (UPP(2) == 'B') && (UPP(3) == 'L') &&
+ (UPP(4) == 'I') && (UPP(5) == 'C')) {
+ SKIP(6);
+ if (!IS_BLANK_CH(CUR)) {
+ htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
+ "Space required after 'PUBLIC'\n", NULL, NULL);
+ }
+ SKIP_BLANKS;
+ *publicID = htmlParsePubidLiteral(ctxt);
+ if (*publicID == NULL) {
+ htmlParseErr(ctxt, XML_ERR_PUBID_REQUIRED,
+ "htmlParseExternalID: PUBLIC, no Public Identifier\n",
+ NULL, NULL);
+ }
+ SKIP_BLANKS;
+ if ((CUR == '"') || (CUR == '\'')) {
+ URI = htmlParseSystemLiteral(ctxt);
+ }
+ }
+ return(URI);
+}
+
+/**
+ * xmlParsePI:
+ * @ctxt: an XML parser context
+ *
+ * parse an XML Processing Instruction.
+ *
+ * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
+ */
+static void
+htmlParsePI(htmlParserCtxtPtr ctxt) {
+ xmlChar *buf = NULL;
+ int len = 0;
+ int size = HTML_PARSER_BUFFER_SIZE;
+ int cur, l;
+ const xmlChar *target;
+ xmlParserInputState state;
+ int count = 0;
+
+ if ((RAW == '<') && (NXT(1) == '?')) {
+ state = ctxt->instate;
+ ctxt->instate = XML_PARSER_PI;
+ /*
+ * this is a Processing Instruction.
+ */
+ SKIP(2);
+ SHRINK;
+
+ /*
+ * Parse the target name and check for special support like
+ * namespace.
+ */
+ target = htmlParseName(ctxt);
+ if (target != NULL) {
+ if (RAW == '>') {
+ SKIP(1);
+
+ /*
+ * SAX: PI detected.
+ */
+ if ((ctxt->sax) && (!ctxt->disableSAX) &&
+ (ctxt->sax->processingInstruction != NULL))
+ ctxt->sax->processingInstruction(ctxt->userData,
+ target, NULL);
+ ctxt->instate = state;
+ return;
+ }
+ buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
+ if (buf == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ ctxt->instate = state;
+ return;
+ }
+ cur = CUR;
+ if (!IS_BLANK(cur)) {
+ htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
+ "ParsePI: PI %s space expected\n", target, NULL);
+ }
+ SKIP_BLANKS;
+ cur = CUR_CHAR(l);
+ while (IS_CHAR(cur) && (cur != '>')) {
+ if (len + 5 >= size) {
+ xmlChar *tmp;
+
+ size *= 2;
+ tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
+ if (tmp == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ xmlFree(buf);
+ ctxt->instate = state;
+ return;
+ }
+ buf = tmp;
+ }
+ count++;
+ if (count > 50) {
+ GROW;
+ count = 0;
+ }
+ COPY_BUF(l,buf,len,cur);
+ NEXTL(l);
+ cur = CUR_CHAR(l);
+ if (cur == 0) {
+ SHRINK;
+ GROW;
+ cur = CUR_CHAR(l);
+ }
+ }
+ buf[len] = 0;
+ if (cur != '>') {
+ htmlParseErr(ctxt, XML_ERR_PI_NOT_FINISHED,
+ "ParsePI: PI %s never end ...\n", target, NULL);
+ } else {
+ SKIP(1);
+
+ /*
+ * SAX: PI detected.
+ */
+ if ((ctxt->sax) && (!ctxt->disableSAX) &&
+ (ctxt->sax->processingInstruction != NULL))
+ ctxt->sax->processingInstruction(ctxt->userData,
+ target, buf);
+ }
+ xmlFree(buf);
+ } else {
+ htmlParseErr(ctxt, XML_ERR_PI_NOT_STARTED,
+ "PI is not started correctly", NULL, NULL);
+ }
+ ctxt->instate = state;
+ }
+}
+
+/**
+ * htmlParseComment:
+ * @ctxt: an HTML parser context
+ *
+ * Parse an XML (SGML) comment <!-- .... -->
+ *
+ * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
+ */
+static void
+htmlParseComment(htmlParserCtxtPtr ctxt) {
+ xmlChar *buf = NULL;
+ int len;
+ int size = HTML_PARSER_BUFFER_SIZE;
+ int q, ql;
+ int r, rl;
+ int cur, l;
+ xmlParserInputState state;
+
+ /*
+ * Check that there is a comment right here.
+ */
+ if ((RAW != '<') || (NXT(1) != '!') ||
+ (NXT(2) != '-') || (NXT(3) != '-')) return;
+
+ state = ctxt->instate;
+ ctxt->instate = XML_PARSER_COMMENT;
+ SHRINK;
+ SKIP(4);
+ buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
+ if (buf == NULL) {
+ htmlErrMemory(ctxt, "buffer allocation failed\n");
+ ctxt->instate = state;
+ return;
+ }
+ len = 0;
+ buf[len] = 0;
+ q = CUR_CHAR(ql);
+ if (!IS_CHAR(q))
+ goto unfinished;
+ NEXTL(ql);
+ r = CUR_CHAR(rl);
+ if (!IS_CHAR(r))
+ goto unfinished;
+ NEXTL(rl);
+ cur = CUR_CHAR(l);
+ while (IS_CHAR(cur) &&
+ ((cur != '>') ||
+ (r != '-') || (q != '-'))) {
+ if (len + 5 >= size) {
+ xmlChar *tmp;
+
+ size *= 2;
+ tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
+ if (tmp == NULL) {
+ xmlFree(buf);
+ htmlErrMemory(ctxt, "growing buffer failed\n");
+ ctxt->instate = state;
+ return;
+ }
+ buf = tmp;
+ }
+ COPY_BUF(ql,buf,len,q);
+ q = r;
+ ql = rl;
+ r = cur;
+ rl = l;
+ NEXTL(l);
+ cur = CUR_CHAR(l);
+ if (cur == 0) {
+ SHRINK;
+ GROW;
+ cur = CUR_CHAR(l);
+ }
+ }
+ buf[len] = 0;
+ if (IS_CHAR(cur)) {
+ NEXT;
+ if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
+ (!ctxt->disableSAX))
+ ctxt->sax->comment(ctxt->userData, buf);
+ xmlFree(buf);
+ ctxt->instate = state;
+ return;
+ }
+
+unfinished:
+ htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
+ "Comment not terminated \n<!--%.50s\n", buf, NULL);
+ xmlFree(buf);
+}
+
+/**
+ * htmlParseCharRef:
+ * @ctxt: an HTML parser context
+ *
+ * parse Reference declarations
+ *
+ * [66] CharRef ::= '&#' [0-9]+ ';' |
+ * '&#x' [0-9a-fA-F]+ ';'
+ *
+ * Returns the value parsed (as an int)
+ */
+int
+htmlParseCharRef(htmlParserCtxtPtr ctxt) {
+ int val = 0;
+
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseCharRef: context error\n",
+ NULL, NULL);
+ return(0);
+ }
+ if ((CUR == '&') && (NXT(1) == '#') &&
+ ((NXT(2) == 'x') || NXT(2) == 'X')) {
+ SKIP(3);
+ while (CUR != ';') {
+ if ((CUR >= '0') && (CUR <= '9'))
+ val = val * 16 + (CUR - '0');
+ else if ((CUR >= 'a') && (CUR <= 'f'))
+ val = val * 16 + (CUR - 'a') + 10;
+ else if ((CUR >= 'A') && (CUR <= 'F'))
+ val = val * 16 + (CUR - 'A') + 10;
+ else {
+ htmlParseErr(ctxt, XML_ERR_INVALID_HEX_CHARREF,
+ "htmlParseCharRef: missing semicolon\n",
+ NULL, NULL);
+ break;
+ }
+ NEXT;
+ }
+ if (CUR == ';')
+ NEXT;
+ } else if ((CUR == '&') && (NXT(1) == '#')) {
+ SKIP(2);
+ while (CUR != ';') {
+ if ((CUR >= '0') && (CUR <= '9'))
+ val = val * 10 + (CUR - '0');
+ else {
+ htmlParseErr(ctxt, XML_ERR_INVALID_DEC_CHARREF,
+ "htmlParseCharRef: missing semicolon\n",
+ NULL, NULL);
+ break;
+ }
+ NEXT;
+ }
+ if (CUR == ';')
+ NEXT;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_INVALID_CHARREF,
+ "htmlParseCharRef: invalid value\n", NULL, NULL);
+ }
+ /*
+ * Check the value IS_CHAR ...
+ */
+ if (IS_CHAR(val)) {
+ return(val);
+ } else {
+ htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
+ "htmlParseCharRef: invalid xmlChar value %d\n",
+ val);
+ }
+ return(0);
+}
+
+
+/**
+ * htmlParseDocTypeDecl:
+ * @ctxt: an HTML parser context
+ *
+ * parse a DOCTYPE declaration
+ *
+ * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
+ * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
+ */
+
+static void
+htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
+ const xmlChar *name;
+ xmlChar *ExternalID = NULL;
+ xmlChar *URI = NULL;
+
+ /*
+ * We know that '<!DOCTYPE' has been detected.
+ */
+ SKIP(9);
+
+ SKIP_BLANKS;
+
+ /*
+ * Parse the DOCTYPE name.
+ */
+ name = htmlParseName(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "htmlParseDocTypeDecl : no DOCTYPE name !\n",
+ NULL, NULL);
+ }
+ /*
+ * Check that upper(name) == "HTML" !!!!!!!!!!!!!
+ */
+
+ SKIP_BLANKS;
+
+ /*
+ * Check for SystemID and ExternalID
+ */
+ URI = htmlParseExternalID(ctxt, &ExternalID);
+ SKIP_BLANKS;
+
+ /*
+ * We should be at the end of the DOCTYPE declaration.
+ */
+ if (CUR != '>') {
+ htmlParseErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED,
+ "DOCTYPE improperly terminated\n", NULL, NULL);
+ /* We shouldn't try to resynchronize ... */
+ }
+ NEXT;
+
+ /*
+ * Create or update the document accordingly to the DOCTYPE
+ */
+ if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
+ (!ctxt->disableSAX))
+ ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
+
+ /*
+ * Cleanup, since we don't use all those identifiers
+ */
+ if (URI != NULL) xmlFree(URI);
+ if (ExternalID != NULL) xmlFree(ExternalID);
+}
+
+/**
+ * htmlParseAttribute:
+ * @ctxt: an HTML parser context
+ * @value: a xmlChar ** used to store the value of the attribute
+ *
+ * parse an attribute
+ *
+ * [41] Attribute ::= Name Eq AttValue
+ *
+ * [25] Eq ::= S? '=' S?
+ *
+ * With namespace:
+ *
+ * [NS 11] Attribute ::= QName Eq AttValue
+ *
+ * Also the case QName == xmlns:??? is handled independently as a namespace
+ * definition.
+ *
+ * Returns the attribute name, and the value in *value.
+ */
+
+static const xmlChar *
+htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
+ const xmlChar *name;
+ xmlChar *val = NULL;
+
+ *value = NULL;
+ name = htmlParseHTMLName(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "error parsing attribute name\n", NULL, NULL);
+ return(NULL);
+ }
+
+ /*
+ * read the value
+ */
+ SKIP_BLANKS;
+ if (CUR == '=') {
+ NEXT;
+ SKIP_BLANKS;
+ val = htmlParseAttValue(ctxt);
+ }
+
+ *value = val;
+ return(name);
+}
+
+/**
+ * htmlCheckEncodingDirect:
+ * @ctxt: an HTML parser context
+ * @attvalue: the attribute value
+ *
+ * Checks an attribute value to detect
+ * the encoding
+ * If a new encoding is detected the parser is switched to decode
+ * it and pass UTF8
+ */
+static void
+htmlCheckEncodingDirect(htmlParserCtxtPtr ctxt, const xmlChar *encoding) {
+
+ if ((ctxt == NULL) || (encoding == NULL) ||
+ (ctxt->options & HTML_PARSE_IGNORE_ENC))
+ return;
+
+ /* do not change encoding */
+ if (ctxt->input->encoding != NULL)
+ return;
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+ xmlCharEncodingHandlerPtr handler;
+
+ while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
+
+ if (ctxt->input->encoding != NULL)
+ xmlFree((xmlChar *) ctxt->input->encoding);
+ ctxt->input->encoding = xmlStrdup(encoding);
+
+ enc = xmlParseCharEncoding((const char *) encoding);
+ /*
+ * registered set of known encodings
+ */
+ if (enc != XML_CHAR_ENCODING_ERROR) {
+ if (((enc == XML_CHAR_ENCODING_UTF16LE) ||
+ (enc == XML_CHAR_ENCODING_UTF16BE) ||
+ (enc == XML_CHAR_ENCODING_UCS4LE) ||
+ (enc == XML_CHAR_ENCODING_UCS4BE)) &&
+ (ctxt->input->buf != NULL) &&
+ (ctxt->input->buf->encoder == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
+ "htmlCheckEncoding: wrong encoding meta\n",
+ NULL, NULL);
+ } else {
+ xmlSwitchEncoding(ctxt, enc);
+ }
+ ctxt->charset = XML_CHAR_ENCODING_UTF8;
+ } else {
+ /*
+ * fallback for unknown encodings
+ */
+ handler = xmlFindCharEncodingHandler((const char *) encoding);
+ if (handler != NULL) {
+ xmlSwitchToEncoding(ctxt, handler);
+ ctxt->charset = XML_CHAR_ENCODING_UTF8;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
+ "htmlCheckEncoding: unknown encoding %s\n",
+ encoding, NULL);
+ }
+ }
+
+ if ((ctxt->input->buf != NULL) &&
+ (ctxt->input->buf->encoder != NULL) &&
+ (ctxt->input->buf->raw != NULL) &&
+ (ctxt->input->buf->buffer != NULL)) {
+ int nbchars;
+ int processed;
+
+ /*
+ * convert as much as possible to the parser reading buffer.
+ */
+ processed = ctxt->input->cur - ctxt->input->base;
+ xmlBufShrink(ctxt->input->buf->buffer, processed);
+ nbchars = xmlCharEncInput(ctxt->input->buf, 1);
+ xmlBufResetInput(ctxt->input->buf->buffer, ctxt->input);
+ if (nbchars < 0) {
+ htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
+ "htmlCheckEncoding: encoder error\n",
+ NULL, NULL);
+ }
+ }
+ }
+}
+
+/**
+ * htmlCheckEncoding:
+ * @ctxt: an HTML parser context
+ * @attvalue: the attribute value
+ *
+ * Checks an http-equiv attribute from a Meta tag to detect
+ * the encoding
+ * If a new encoding is detected the parser is switched to decode
+ * it and pass UTF8
+ */
+static void
+htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
+ const xmlChar *encoding;
+
+ if (!attvalue)
+ return;
+
+ encoding = xmlStrcasestr(attvalue, BAD_CAST"charset");
+ if (encoding != NULL) {
+ encoding += 7;
+ }
+ /*
+ * skip blank
+ */
+ if (encoding && IS_BLANK_CH(*encoding))
+ encoding = xmlStrcasestr(attvalue, BAD_CAST"=");
+ if (encoding && *encoding == '=') {
+ encoding ++;
+ htmlCheckEncodingDirect(ctxt, encoding);
+ }
+}
+
+/**
+ * htmlCheckMeta:
+ * @ctxt: an HTML parser context
+ * @atts: the attributes values
+ *
+ * Checks an attributes from a Meta tag
+ */
+static void
+htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
+ int i;
+ const xmlChar *att, *value;
+ int http = 0;
+ const xmlChar *content = NULL;
+
+ if ((ctxt == NULL) || (atts == NULL))
+ return;
+
+ i = 0;
+ att = atts[i++];
+ while (att != NULL) {
+ value = atts[i++];
+ if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
+ && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
+ http = 1;
+ else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"charset")))
+ htmlCheckEncodingDirect(ctxt, value);
+ else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
+ content = value;
+ att = atts[i++];
+ }
+ if ((http) && (content != NULL))
+ htmlCheckEncoding(ctxt, content);
+
+}
+
+/**
+ * htmlParseStartTag:
+ * @ctxt: an HTML parser context
+ *
+ * parse a start of tag either for rule element or
+ * EmptyElement. In both case we don't parse the tag closing chars.
+ *
+ * [40] STag ::= '<' Name (S Attribute)* S? '>'
+ *
+ * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
+ *
+ * With namespace:
+ *
+ * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
+ *
+ * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
+ *
+ * Returns 0 in case of success, -1 in case of error and 1 if discarded
+ */
+
+static int
+htmlParseStartTag(htmlParserCtxtPtr ctxt) {
+ const xmlChar *name;
+ const xmlChar *attname;
+ xmlChar *attvalue;
+ const xmlChar **atts;
+ int nbatts = 0;
+ int maxatts;
+ int meta = 0;
+ int i;
+ int discardtag = 0;
+
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseStartTag: context error\n", NULL, NULL);
+ return -1;
+ }
+ if (ctxt->instate == XML_PARSER_EOF)
+ return(-1);
+ if (CUR != '<') return -1;
+ NEXT;
+
+ atts = ctxt->atts;
+ maxatts = ctxt->maxatts;
+
+ GROW;
+ name = htmlParseHTMLName(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "htmlParseStartTag: invalid element name\n",
+ NULL, NULL);
+ /* if recover preserve text on classic misconstructs */
+ if ((ctxt->recovery) && ((IS_BLANK_CH(CUR)) || (CUR == '<') ||
+ (CUR == '=') || (CUR == '>') || (((CUR >= '0') && (CUR <= '9'))))) {
+ htmlParseCharDataInternal(ctxt, '<');
+ return(-1);
+ }
+
+
+ /* Dump the bogus tag like browsers do */
+ while ((IS_CHAR_CH(CUR)) && (CUR != '>') &&
+ (ctxt->instate != XML_PARSER_EOF))
+ NEXT;
+ return -1;
+ }
+ if (xmlStrEqual(name, BAD_CAST"meta"))
+ meta = 1;
+
+ /*
+ * Check for auto-closure of HTML elements.
+ */
+ htmlAutoClose(ctxt, name);
+
+ /*
+ * Check for implied HTML elements.
+ */
+ htmlCheckImplied(ctxt, name);
+
+ /*
+ * Avoid html at any level > 0, head at any level != 1
+ * or any attempt to recurse body
+ */
+ if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) {
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "htmlParseStartTag: misplaced <html> tag\n",
+ name, NULL);
+ discardtag = 1;
+ ctxt->depth++;
+ }
+ if ((ctxt->nameNr != 1) &&
+ (xmlStrEqual(name, BAD_CAST"head"))) {
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "htmlParseStartTag: misplaced <head> tag\n",
+ name, NULL);
+ discardtag = 1;
+ ctxt->depth++;
+ }
+ if (xmlStrEqual(name, BAD_CAST"body")) {
+ int indx;
+ for (indx = 0;indx < ctxt->nameNr;indx++) {
+ if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) {
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "htmlParseStartTag: misplaced <body> tag\n",
+ name, NULL);
+ discardtag = 1;
+ ctxt->depth++;
+ }
+ }
+ }
+
+ /*
+ * Now parse the attributes, it ends up with the ending
+ *
+ * (S Attribute)* S?
+ */
+ SKIP_BLANKS;
+ while ((IS_CHAR_CH(CUR)) &&
+ (CUR != '>') &&
+ ((CUR != '/') || (NXT(1) != '>'))) {
+ long cons = ctxt->nbChars;
+
+ GROW;
+ attname = htmlParseAttribute(ctxt, &attvalue);
+ if (attname != NULL) {
+
+ /*
+ * Well formedness requires at most one declaration of an attribute
+ */
+ for (i = 0; i < nbatts;i += 2) {
+ if (xmlStrEqual(atts[i], attname)) {
+ htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_REDEFINED,
+ "Attribute %s redefined\n", attname, NULL);
+ if (attvalue != NULL)
+ xmlFree(attvalue);
+ goto failed;
+ }
+ }
+
+ /*
+ * Add the pair to atts
+ */
+ if (atts == NULL) {
+ maxatts = 22; /* allow for 10 attrs by default */
+ atts = (const xmlChar **)
+ xmlMalloc(maxatts * sizeof(xmlChar *));
+ if (atts == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ if (attvalue != NULL)
+ xmlFree(attvalue);
+ goto failed;
+ }
+ ctxt->atts = atts;
+ ctxt->maxatts = maxatts;
+ } else if (nbatts + 4 > maxatts) {
+ const xmlChar **n;
+
+ maxatts *= 2;
+ n = (const xmlChar **) xmlRealloc((void *) atts,
+ maxatts * sizeof(const xmlChar *));
+ if (n == NULL) {
+ htmlErrMemory(ctxt, NULL);
+ if (attvalue != NULL)
+ xmlFree(attvalue);
+ goto failed;
+ }
+ atts = n;
+ ctxt->atts = atts;
+ ctxt->maxatts = maxatts;
+ }
+ atts[nbatts++] = attname;
+ atts[nbatts++] = attvalue;
+ atts[nbatts] = NULL;
+ atts[nbatts + 1] = NULL;
+ }
+ else {
+ if (attvalue != NULL)
+ xmlFree(attvalue);
+ /* Dump the bogus attribute string up to the next blank or
+ * the end of the tag. */
+ while ((IS_CHAR_CH(CUR)) &&
+ !(IS_BLANK_CH(CUR)) && (CUR != '>') &&
+ ((CUR != '/') || (NXT(1) != '>')))
+ NEXT;
+ }
+
+failed:
+ SKIP_BLANKS;
+ if (cons == ctxt->nbChars) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseStartTag: problem parsing attributes\n",
+ NULL, NULL);
+ break;
+ }
+ }
+
+ /*
+ * Handle specific association to the META tag
+ */
+ if (meta && (nbatts != 0))
+ htmlCheckMeta(ctxt, atts);
+
+ /*
+ * SAX: Start of Element !
+ */
+ if (!discardtag) {
+ htmlnamePush(ctxt, name);
+ if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) {
+ if (nbatts != 0)
+ ctxt->sax->startElement(ctxt->userData, name, atts);
+ else
+ ctxt->sax->startElement(ctxt->userData, name, NULL);
+ }
+ }
+
+ if (atts != NULL) {
+ for (i = 1;i < nbatts;i += 2) {
+ if (atts[i] != NULL)
+ xmlFree((xmlChar *) atts[i]);
+ }
+ }
+
+ return(discardtag);
+}
+
+/**
+ * htmlParseEndTag:
+ * @ctxt: an HTML parser context
+ *
+ * parse an end of tag
+ *
+ * [42] ETag ::= '</' Name S? '>'
+ *
+ * With namespace
+ *
+ * [NS 9] ETag ::= '</' QName S? '>'
+ *
+ * Returns 1 if the current level should be closed.
+ */
+
+static int
+htmlParseEndTag(htmlParserCtxtPtr ctxt)
+{
+ const xmlChar *name;
+ const xmlChar *oldname;
+ int i, ret;
+
+ if ((CUR != '<') || (NXT(1) != '/')) {
+ htmlParseErr(ctxt, XML_ERR_LTSLASH_REQUIRED,
+ "htmlParseEndTag: '</' not found\n", NULL, NULL);
+ return (0);
+ }
+ SKIP(2);
+
+ name = htmlParseHTMLName(ctxt);
+ if (name == NULL)
+ return (0);
+ /*
+ * We should definitely be at the ending "S? '>'" part
+ */
+ SKIP_BLANKS;
+ if ((!IS_CHAR_CH(CUR)) || (CUR != '>')) {
+ htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
+ "End tag : expected '>'\n", NULL, NULL);
+ if (ctxt->recovery) {
+ /*
+ * We're not at the ending > !!
+ * Error, unless in recover mode where we search forwards
+ * until we find a >
+ */
+ while (CUR != '\0' && CUR != '>') NEXT;
+ NEXT;
+ }
+ } else
+ NEXT;
+
+ /*
+ * if we ignored misplaced tags in htmlParseStartTag don't pop them
+ * out now.
+ */
+ if ((ctxt->depth > 0) &&
+ (xmlStrEqual(name, BAD_CAST "html") ||
+ xmlStrEqual(name, BAD_CAST "body") ||
+ xmlStrEqual(name, BAD_CAST "head"))) {
+ ctxt->depth--;
+ return (0);
+ }
+
+ /*
+ * If the name read is not one of the element in the parsing stack
+ * then return, it's just an error.
+ */
+ for (i = (ctxt->nameNr - 1); i >= 0; i--) {
+ if (xmlStrEqual(name, ctxt->nameTab[i]))
+ break;
+ }
+ if (i < 0) {
+ htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
+ "Unexpected end tag : %s\n", name, NULL);
+ return (0);
+ }
+
+
+ /*
+ * Check for auto-closure of HTML elements.
+ */
+
+ htmlAutoCloseOnClose(ctxt, name);
+
+ /*
+ * Well formedness constraints, opening and closing must match.
+ * With the exception that the autoclose may have popped stuff out
+ * of the stack.
+ */
+ if (!xmlStrEqual(name, ctxt->name)) {
+ if ((ctxt->name != NULL) && (!xmlStrEqual(ctxt->name, name))) {
+ htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
+ "Opening and ending tag mismatch: %s and %s\n",
+ name, ctxt->name);
+ }
+ }
+
+ /*
+ * SAX: End of Tag
+ */
+ oldname = ctxt->name;
+ if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlNodeInfoPop(ctxt);
+ htmlnamePop(ctxt);
+ ret = 1;
+ } else {
+ ret = 0;
+ }
+
+ return (ret);
+}
+
+
+/**
+ * htmlParseReference:
+ * @ctxt: an HTML parser context
+ *
+ * parse and handle entity references in content,
+ * this will end-up in a call to character() since this is either a
+ * CharRef, or a predefined entity.
+ */
+static void
+htmlParseReference(htmlParserCtxtPtr ctxt) {
+ const htmlEntityDesc * ent;
+ xmlChar out[6];
+ const xmlChar *name;
+ if (CUR != '&') return;
+
+ if (NXT(1) == '#') {
+ unsigned int c;
+ int bits, i = 0;
+
+ c = htmlParseCharRef(ctxt);
+ if (c == 0)
+ return;
+
+ if (c < 0x80) { out[i++]= c; bits= -6; }
+ else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
+ else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
+ else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
+
+ for ( ; bits >= 0; bits-= 6) {
+ out[i++]= ((c >> bits) & 0x3F) | 0x80;
+ }
+ out[i] = 0;
+
+ htmlCheckParagraph(ctxt);
+ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
+ ctxt->sax->characters(ctxt->userData, out, i);
+ } else {
+ ent = htmlParseEntityRef(ctxt, &name);
+ if (name == NULL) {
+ htmlCheckParagraph(ctxt);
+ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
+ ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
+ return;
+ }
+ if ((ent == NULL) || !(ent->value > 0)) {
+ htmlCheckParagraph(ctxt);
+ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
+ ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
+ ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
+ /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
+ }
+ } else {
+ unsigned int c;
+ int bits, i = 0;
+
+ c = ent->value;
+ if (c < 0x80)
+ { out[i++]= c; bits= -6; }
+ else if (c < 0x800)
+ { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
+ else if (c < 0x10000)
+ { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
+ else
+ { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
+
+ for ( ; bits >= 0; bits-= 6) {
+ out[i++]= ((c >> bits) & 0x3F) | 0x80;
+ }
+ out[i] = 0;
+
+ htmlCheckParagraph(ctxt);
+ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
+ ctxt->sax->characters(ctxt->userData, out, i);
+ }
+ }
+}
+
+/**
+ * htmlParseContent:
+ * @ctxt: an HTML parser context
+ *
+ * Parse a content: comment, sub-element, reference or text.
+ * Kept for compatibility with old code
+ */
+
+static void
+htmlParseContent(htmlParserCtxtPtr ctxt) {
+ xmlChar *currentNode;
+ int depth;
+ const xmlChar *name;
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ while (1) {
+ long cons = ctxt->nbChars;
+
+ GROW;
+
+ if (ctxt->instate == XML_PARSER_EOF)
+ break;
+
+ /*
+ * Our tag or one of it's parent or children is ending.
+ */
+ if ((CUR == '<') && (NXT(1) == '/')) {
+ if (htmlParseEndTag(ctxt) &&
+ ((currentNode != NULL) || (ctxt->nameNr == 0))) {
+ if (currentNode != NULL)
+ xmlFree(currentNode);
+ return;
+ }
+ continue; /* while */
+ }
+
+ else if ((CUR == '<') &&
+ ((IS_ASCII_LETTER(NXT(1))) ||
+ (NXT(1) == '_') || (NXT(1) == ':'))) {
+ name = htmlParseHTMLName_nonInvasive(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "htmlParseStartTag: invalid element name\n",
+ NULL, NULL);
+ /* Dump the bogus tag like browsers do */
+ while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
+ NEXT;
+
+ if (currentNode != NULL)
+ xmlFree(currentNode);
+ return;
+ }
+
+ if (ctxt->name != NULL) {
+ if (htmlCheckAutoClose(name, ctxt->name) == 1) {
+ htmlAutoClose(ctxt, name);
+ continue;
+ }
+ }
+ }
+
+ /*
+ * Has this node been popped out during parsing of
+ * the next element
+ */
+ if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
+ (!xmlStrEqual(currentNode, ctxt->name)))
+ {
+ if (currentNode != NULL) xmlFree(currentNode);
+ return;
+ }
+
+ if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
+ (xmlStrEqual(currentNode, BAD_CAST"style")))) {
+ /*
+ * Handle SCRIPT/STYLE separately
+ */
+ htmlParseScript(ctxt);
+ } else {
+ /*
+ * Sometimes DOCTYPE arrives in the middle of the document
+ */
+ if ((CUR == '<') && (NXT(1) == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "Misplaced DOCTYPE declaration\n",
+ BAD_CAST "DOCTYPE" , NULL);
+ htmlParseDocTypeDecl(ctxt);
+ }
+
+ /*
+ * First case : a comment
+ */
+ if ((CUR == '<') && (NXT(1) == '!') &&
+ (NXT(2) == '-') && (NXT(3) == '-')) {
+ htmlParseComment(ctxt);
+ }
+
+ /*
+ * Second case : a Processing Instruction.
+ */
+ else if ((CUR == '<') && (NXT(1) == '?')) {
+ htmlParsePI(ctxt);
+ }
+
+ /*
+ * Third case : a sub-element.
+ */
+ else if (CUR == '<') {
+ htmlParseElement(ctxt);
+ }
+
+ /*
+ * Fourth case : a reference. If if has not been resolved,
+ * parsing returns it's Name, create the node
+ */
+ else if (CUR == '&') {
+ htmlParseReference(ctxt);
+ }
+
+ /*
+ * Fifth case : end of the resource
+ */
+ else if (CUR == 0) {
+ htmlAutoCloseOnEnd(ctxt);
+ break;
+ }
+
+ /*
+ * Last case, text. Note that References are handled directly.
+ */
+ else {
+ htmlParseCharData(ctxt);
+ }
+
+ if (cons == ctxt->nbChars) {
+ if (ctxt->node != NULL) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "detected an error in element content\n",
+ NULL, NULL);
+ }
+ break;
+ }
+ }
+ GROW;
+ }
+ if (currentNode != NULL) xmlFree(currentNode);
+}
+
+/**
+ * htmlParseElement:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML element, this is highly recursive
+ * this is kept for compatibility with previous code versions
+ *
+ * [39] element ::= EmptyElemTag | STag content ETag
+ *
+ * [41] Attribute ::= Name Eq AttValue
+ */
+
+void
+htmlParseElement(htmlParserCtxtPtr ctxt) {
+ const xmlChar *name;
+ xmlChar *currentNode = NULL;
+ const htmlElemDesc * info;
+ htmlParserNodeInfo node_info;
+ int failed;
+ int depth;
+ const xmlChar *oldptr;
+
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseElement: context error\n", NULL, NULL);
+ return;
+ }
+
+ if (ctxt->instate == XML_PARSER_EOF)
+ return;
+
+ /* Capture start position */
+ if (ctxt->record_info) {
+ node_info.begin_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ node_info.begin_line = ctxt->input->line;
+ }
+
+ failed = htmlParseStartTag(ctxt);
+ name = ctxt->name;
+ if ((failed == -1) || (name == NULL)) {
+ if (CUR == '>')
+ NEXT;
+ return;
+ }
+
+ /*
+ * Lookup the info for that element.
+ */
+ info = htmlTagLookup(name);
+ if (info == NULL) {
+ htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
+ "Tag %s invalid\n", name, NULL);
+ }
+
+ /*
+ * Check for an Empty Element labeled the XML/SGML way
+ */
+ if ((CUR == '/') && (NXT(1) == '>')) {
+ SKIP(2);
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ return;
+ }
+
+ if (CUR == '>') {
+ NEXT;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
+ "Couldn't find end of Start Tag %s\n", name, NULL);
+
+ /*
+ * end of parsing of this node.
+ */
+ if (xmlStrEqual(name, ctxt->name)) {
+ nodePop(ctxt);
+ htmlnamePop(ctxt);
+ }
+
+ /*
+ * Capture end position and add node
+ */
+ if (ctxt->record_info) {
+ node_info.end_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ node_info.end_line = ctxt->input->line;
+ node_info.node = ctxt->node;
+ xmlParserAddNodeInfo(ctxt, &node_info);
+ }
+ return;
+ }
+
+ /*
+ * Check for an Empty Element from DTD definition
+ */
+ if ((info != NULL) && (info->empty)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ return;
+ }
+
+ /*
+ * Parse the content of the element:
+ */
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ while (IS_CHAR_CH(CUR)) {
+ oldptr = ctxt->input->cur;
+ htmlParseContent(ctxt);
+ if (oldptr==ctxt->input->cur) break;
+ if (ctxt->nameNr < depth) break;
+ }
+
+ /*
+ * Capture end position and add node
+ */
+ if ( currentNode != NULL && ctxt->record_info ) {
+ node_info.end_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ node_info.end_line = ctxt->input->line;
+ node_info.node = ctxt->node;
+ xmlParserAddNodeInfo(ctxt, &node_info);
+ }
+ if (!IS_CHAR_CH(CUR)) {
+ htmlAutoCloseOnEnd(ctxt);
+ }
+
+ if (currentNode != NULL)
+ xmlFree(currentNode);
+}
+
+static void
+htmlParserFinishElementParsing(htmlParserCtxtPtr ctxt) {
+ /*
+ * Capture end position and add node
+ */
+ if ( ctxt->node != NULL && ctxt->record_info ) {
+ ctxt->nodeInfo->end_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ ctxt->nodeInfo->end_line = ctxt->input->line;
+ ctxt->nodeInfo->node = ctxt->node;
+ xmlParserAddNodeInfo(ctxt, ctxt->nodeInfo);
+ htmlNodeInfoPop(ctxt);
+ }
+ if (!IS_CHAR_CH(CUR)) {
+ htmlAutoCloseOnEnd(ctxt);
+ }
+}
+
+/**
+ * htmlParseElementInternal:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML element, new version, non recursive
+ *
+ * [39] element ::= EmptyElemTag | STag content ETag
+ *
+ * [41] Attribute ::= Name Eq AttValue
+ */
+
+static void
+htmlParseElementInternal(htmlParserCtxtPtr ctxt) {
+ const xmlChar *name;
+ const htmlElemDesc * info;
+ htmlParserNodeInfo node_info = { NULL, 0, 0, 0, 0 };
+ int failed;
+
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseElementInternal: context error\n", NULL, NULL);
+ return;
+ }
+
+ if (ctxt->instate == XML_PARSER_EOF)
+ return;
+
+ /* Capture start position */
+ if (ctxt->record_info) {
+ node_info.begin_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ node_info.begin_line = ctxt->input->line;
+ }
+
+ failed = htmlParseStartTag(ctxt);
+ name = ctxt->name;
+ if ((failed == -1) || (name == NULL)) {
+ if (CUR == '>')
+ NEXT;
+ return;
+ }
+
+ /*
+ * Lookup the info for that element.
+ */
+ info = htmlTagLookup(name);
+ if (info == NULL) {
+ htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
+ "Tag %s invalid\n", name, NULL);
+ }
+
+ /*
+ * Check for an Empty Element labeled the XML/SGML way
+ */
+ if ((CUR == '/') && (NXT(1) == '>')) {
+ SKIP(2);
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ return;
+ }
+
+ if (CUR == '>') {
+ NEXT;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
+ "Couldn't find end of Start Tag %s\n", name, NULL);
+
+ /*
+ * end of parsing of this node.
+ */
+ if (xmlStrEqual(name, ctxt->name)) {
+ nodePop(ctxt);
+ htmlnamePop(ctxt);
+ }
+
+ if (ctxt->record_info)
+ htmlNodeInfoPush(ctxt, &node_info);
+ htmlParserFinishElementParsing(ctxt);
+ return;
+ }
+
+ /*
+ * Check for an Empty Element from DTD definition
+ */
+ if ((info != NULL) && (info->empty)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ return;
+ }
+
+ if (ctxt->record_info)
+ htmlNodeInfoPush(ctxt, &node_info);
+}
+
+/**
+ * htmlParseContentInternal:
+ * @ctxt: an HTML parser context
+ *
+ * Parse a content: comment, sub-element, reference or text.
+ * New version for non recursive htmlParseElementInternal
+ */
+
+static void
+htmlParseContentInternal(htmlParserCtxtPtr ctxt) {
+ xmlChar *currentNode;
+ int depth;
+ const xmlChar *name;
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ while (1) {
+ long cons = ctxt->nbChars;
+
+ GROW;
+
+ if (ctxt->instate == XML_PARSER_EOF)
+ break;
+
+ /*
+ * Our tag or one of it's parent or children is ending.
+ */
+ if ((CUR == '<') && (NXT(1) == '/')) {
+ if (htmlParseEndTag(ctxt) &&
+ ((currentNode != NULL) || (ctxt->nameNr == 0))) {
+ if (currentNode != NULL)
+ xmlFree(currentNode);
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ }
+ continue; /* while */
+ }
+
+ else if ((CUR == '<') &&
+ ((IS_ASCII_LETTER(NXT(1))) ||
+ (NXT(1) == '_') || (NXT(1) == ':'))) {
+ name = htmlParseHTMLName_nonInvasive(ctxt);
+ if (name == NULL) {
+ htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
+ "htmlParseStartTag: invalid element name\n",
+ NULL, NULL);
+ /* Dump the bogus tag like browsers do */
+ while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
+ NEXT;
+
+ htmlParserFinishElementParsing(ctxt);
+ if (currentNode != NULL)
+ xmlFree(currentNode);
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ continue;
+ }
+
+ if (ctxt->name != NULL) {
+ if (htmlCheckAutoClose(name, ctxt->name) == 1) {
+ htmlAutoClose(ctxt, name);
+ continue;
+ }
+ }
+ }
+
+ /*
+ * Has this node been popped out during parsing of
+ * the next element
+ */
+ if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
+ (!xmlStrEqual(currentNode, ctxt->name)))
+ {
+ htmlParserFinishElementParsing(ctxt);
+ if (currentNode != NULL) xmlFree(currentNode);
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ continue;
+ }
+
+ if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
+ (xmlStrEqual(currentNode, BAD_CAST"style")))) {
+ /*
+ * Handle SCRIPT/STYLE separately
+ */
+ htmlParseScript(ctxt);
+ } else {
+ /*
+ * Sometimes DOCTYPE arrives in the middle of the document
+ */
+ if ((CUR == '<') && (NXT(1) == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "Misplaced DOCTYPE declaration\n",
+ BAD_CAST "DOCTYPE" , NULL);
+ htmlParseDocTypeDecl(ctxt);
+ }
+
+ /*
+ * First case : a comment
+ */
+ if ((CUR == '<') && (NXT(1) == '!') &&
+ (NXT(2) == '-') && (NXT(3) == '-')) {
+ htmlParseComment(ctxt);
+ }
+
+ /*
+ * Second case : a Processing Instruction.
+ */
+ else if ((CUR == '<') && (NXT(1) == '?')) {
+ htmlParsePI(ctxt);
+ }
+
+ /*
+ * Third case : a sub-element.
+ */
+ else if (CUR == '<') {
+ htmlParseElementInternal(ctxt);
+ if (currentNode != NULL) xmlFree(currentNode);
+
+ currentNode = xmlStrdup(ctxt->name);
+ depth = ctxt->nameNr;
+ }
+
+ /*
+ * Fourth case : a reference. If if has not been resolved,
+ * parsing returns it's Name, create the node
+ */
+ else if (CUR == '&') {
+ htmlParseReference(ctxt);
+ }
+
+ /*
+ * Fifth case : end of the resource
+ */
+ else if (CUR == 0) {
+ htmlAutoCloseOnEnd(ctxt);
+ break;
+ }
+
+ /*
+ * Last case, text. Note that References are handled directly.
+ */
+ else {
+ htmlParseCharData(ctxt);
+ }
+
+ if (cons == ctxt->nbChars) {
+ if (ctxt->node != NULL) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "detected an error in element content\n",
+ NULL, NULL);
+ }
+ break;
+ }
+ }
+ GROW;
+ }
+ if (currentNode != NULL) xmlFree(currentNode);
+}
+
+/**
+ * htmlParseContent:
+ * @ctxt: an HTML parser context
+ *
+ * Parse a content: comment, sub-element, reference or text.
+ * This is the entry point when called from parser.c
+ */
+
+void
+__htmlParseContent(void *ctxt) {
+ if (ctxt != NULL)
+ htmlParseContentInternal((htmlParserCtxtPtr) ctxt);
+}
+
+/**
+ * htmlParseDocument:
+ * @ctxt: an HTML parser context
+ *
+ * parse an HTML document (and build a tree if using the standard SAX
+ * interface).
+ *
+ * Returns 0, -1 in case of error. the parser context is augmented
+ * as a result of the parsing.
+ */
+
+int
+htmlParseDocument(htmlParserCtxtPtr ctxt) {
+ xmlChar start[4];
+ xmlCharEncoding enc;
+ xmlDtdPtr dtd;
+
+ xmlInitParser();
+
+ htmlDefaultSAXHandlerInit();
+
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseDocument: context error\n", NULL, NULL);
+ return(XML_ERR_INTERNAL_ERROR);
+ }
+ ctxt->html = 1;
+ ctxt->linenumbers = 1;
+ GROW;
+ /*
+ * SAX: beginning of the document processing.
+ */
+ if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
+ ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
+
+ if ((ctxt->encoding == (const xmlChar *)XML_CHAR_ENCODING_NONE) &&
+ ((ctxt->input->end - ctxt->input->cur) >= 4)) {
+ /*
+ * Get the 4 first bytes and decode the charset
+ * if enc != XML_CHAR_ENCODING_NONE
+ * plug some encoding conversion routines.
+ */
+ start[0] = RAW;
+ start[1] = NXT(1);
+ start[2] = NXT(2);
+ start[3] = NXT(3);
+ enc = xmlDetectCharEncoding(&start[0], 4);
+ if (enc != XML_CHAR_ENCODING_NONE) {
+ xmlSwitchEncoding(ctxt, enc);
+ }
+ }
+
+ /*
+ * Wipe out everything which is before the first '<'
+ */
+ SKIP_BLANKS;
+ if (CUR == 0) {
+ htmlParseErr(ctxt, XML_ERR_DOCUMENT_EMPTY,
+ "Document is empty\n", NULL, NULL);
+ }
+
+ if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
+ ctxt->sax->startDocument(ctxt->userData);
+
+
+ /*
+ * Parse possible comments and PIs before any content
+ */
+ while (((CUR == '<') && (NXT(1) == '!') &&
+ (NXT(2) == '-') && (NXT(3) == '-')) ||
+ ((CUR == '<') && (NXT(1) == '?'))) {
+ htmlParseComment(ctxt);
+ htmlParsePI(ctxt);
+ SKIP_BLANKS;
+ }
+
+
+ /*
+ * Then possibly doc type declaration(s) and more Misc
+ * (doctypedecl Misc*)?
+ */
+ if ((CUR == '<') && (NXT(1) == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ htmlParseDocTypeDecl(ctxt);
+ }
+ SKIP_BLANKS;
+
+ /*
+ * Parse possible comments and PIs before any content
+ */
+ while (((CUR == '<') && (NXT(1) == '!') &&
+ (NXT(2) == '-') && (NXT(3) == '-')) ||
+ ((CUR == '<') && (NXT(1) == '?'))) {
+ htmlParseComment(ctxt);
+ htmlParsePI(ctxt);
+ SKIP_BLANKS;
+ }
+
+ /*
+ * Time to start parsing the tree itself
+ */
+ htmlParseContentInternal(ctxt);
+
+ /*
+ * autoclose
+ */
+ if (CUR == 0)
+ htmlAutoCloseOnEnd(ctxt);
+
+
+ /*
+ * SAX: end of the document processing.
+ */
+ if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
+ ctxt->sax->endDocument(ctxt->userData);
+
+ if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL)) {
+ dtd = xmlGetIntSubset(ctxt->myDoc);
+ if (dtd == NULL)
+ ctxt->myDoc->intSubset =
+ xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
+ BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
+ BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
+ }
+ if (! ctxt->wellFormed) return(-1);
+ return(0);
+}
+
+
+/************************************************************************
+ * *
+ * Parser contexts handling *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlInitParserCtxt:
+ * @ctxt: an HTML parser context
+ *
+ * Initialize a parser context
+ *
+ * Returns 0 in case of success and -1 in case of error
+ */
+
+static int
+htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
+{
+ htmlSAXHandler *sax;
+
+ if (ctxt == NULL) return(-1);
+ memset(ctxt, 0, sizeof(htmlParserCtxt));
+
+ ctxt->dict = xmlDictCreate();
+ if (ctxt->dict == NULL) {
+ htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
+ return(-1);
+ }
+ sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
+ if (sax == NULL) {
+ htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
+ return(-1);
+ }
+ else
+ memset(sax, 0, sizeof(htmlSAXHandler));
+
+ /* Allocate the Input stack */
+ ctxt->inputTab = (htmlParserInputPtr *)
+ xmlMalloc(5 * sizeof(htmlParserInputPtr));
+ if (ctxt->inputTab == NULL) {
+ htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
+ ctxt->inputNr = 0;
+ ctxt->inputMax = 0;
+ ctxt->input = NULL;
+ return(-1);
+ }
+ ctxt->inputNr = 0;
+ ctxt->inputMax = 5;
+ ctxt->input = NULL;
+ ctxt->version = NULL;
+ ctxt->encoding = NULL;
+ ctxt->standalone = -1;
+ ctxt->instate = XML_PARSER_START;
+
+ /* Allocate the Node stack */
+ ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
+ if (ctxt->nodeTab == NULL) {
+ htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
+ ctxt->nodeNr = 0;
+ ctxt->nodeMax = 0;
+ ctxt->node = NULL;
+ ctxt->inputNr = 0;
+ ctxt->inputMax = 0;
+ ctxt->input = NULL;
+ return(-1);
+ }
+ ctxt->nodeNr = 0;
+ ctxt->nodeMax = 10;
+ ctxt->node = NULL;
+
+ /* Allocate the Name stack */
+ ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
+ if (ctxt->nameTab == NULL) {
+ htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
+ ctxt->nameNr = 0;
+ ctxt->nameMax = 0;
+ ctxt->name = NULL;
+ ctxt->nodeNr = 0;
+ ctxt->nodeMax = 0;
+ ctxt->node = NULL;
+ ctxt->inputNr = 0;
+ ctxt->inputMax = 0;
+ ctxt->input = NULL;
+ return(-1);
+ }
+ ctxt->nameNr = 0;
+ ctxt->nameMax = 10;
+ ctxt->name = NULL;
+
+ ctxt->nodeInfoTab = NULL;
+ ctxt->nodeInfoNr = 0;
+ ctxt->nodeInfoMax = 0;
+
+ if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
+ else {
+ ctxt->sax = sax;
+ memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
+ }
+ ctxt->userData = ctxt;
+ ctxt->myDoc = NULL;
+ ctxt->wellFormed = 1;
+ ctxt->replaceEntities = 0;
+ ctxt->linenumbers = xmlLineNumbersDefaultValue;
+ ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
+ ctxt->html = 1;
+ ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
+ ctxt->vctxt.userData = ctxt;
+ ctxt->vctxt.error = xmlParserValidityError;
+ ctxt->vctxt.warning = xmlParserValidityWarning;
+ ctxt->record_info = 0;
+ ctxt->validate = 0;
+ ctxt->nbChars = 0;
+ ctxt->checkIndex = 0;
+ ctxt->catalogs = NULL;
+ xmlInitNodeInfoSeq(&ctxt->node_seq);
+ return(0);
+}
+
+/**
+ * htmlFreeParserCtxt:
+ * @ctxt: an HTML parser context
+ *
+ * Free all the memory used by a parser context. However the parsed
+ * document in ctxt->myDoc is not freed.
+ */
+
+void
+htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
+{
+ xmlFreeParserCtxt(ctxt);
+}
+
+/**
+ * htmlNewParserCtxt:
+ *
+ * Allocate and initialize a new parser context.
+ *
+ * Returns the htmlParserCtxtPtr or NULL in case of allocation error
+ */
+
+htmlParserCtxtPtr
+htmlNewParserCtxt(void)
+{
+ xmlParserCtxtPtr ctxt;
+
+ ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
+ if (ctxt == NULL) {
+ htmlErrMemory(NULL, "NewParserCtxt: out of memory\n");
+ return(NULL);
+ }
+ memset(ctxt, 0, sizeof(xmlParserCtxt));
+ if (htmlInitParserCtxt(ctxt) < 0) {
+ htmlFreeParserCtxt(ctxt);
+ return(NULL);
+ }
+ return(ctxt);
+}
+
+/**
+ * htmlCreateMemoryParserCtxt:
+ * @buffer: a pointer to a char array
+ * @size: the size of the array
+ *
+ * Create a parser context for an HTML in-memory document.
+ *
+ * Returns the new parser context or NULL
+ */
+htmlParserCtxtPtr
+htmlCreateMemoryParserCtxt(const char *buffer, int size) {
+ xmlParserCtxtPtr ctxt;
+ xmlParserInputPtr input;
+ xmlParserInputBufferPtr buf;
+
+ if (buffer == NULL)
+ return(NULL);
+ if (size <= 0)
+ return(NULL);
+
+ ctxt = htmlNewParserCtxt();
+ if (ctxt == NULL)
+ return(NULL);
+
+ buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
+ if (buf == NULL) return(NULL);
+
+ input = xmlNewInputStream(ctxt);
+ if (input == NULL) {
+ xmlFreeParserCtxt(ctxt);
+ return(NULL);
+ }
+
+ input->filename = NULL;
+ input->buf = buf;
+ xmlBufResetInput(buf->buffer, input);
+
+ inputPush(ctxt, input);
+ return(ctxt);
+}
+
+/**
+ * htmlCreateDocParserCtxt:
+ * @cur: a pointer to an array of xmlChar
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ *
+ * Create a parser context for an HTML document.
+ *
+ * TODO: check the need to add encoding handling there
+ *
+ * Returns the new parser context or NULL
+ */
+static htmlParserCtxtPtr
+htmlCreateDocParserCtxt(const xmlChar *cur, const char *encoding) {
+ int len;
+ htmlParserCtxtPtr ctxt;
+
+ if (cur == NULL)
+ return(NULL);
+ len = xmlStrlen(cur);
+ ctxt = htmlCreateMemoryParserCtxt((char *)cur, len);
+ if (ctxt == NULL)
+ return(NULL);
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+ xmlCharEncodingHandlerPtr handler;
+
+ if (ctxt->input->encoding != NULL)
+ xmlFree((xmlChar *) ctxt->input->encoding);
+ ctxt->input->encoding = xmlStrdup((const xmlChar *) encoding);
+
+ enc = xmlParseCharEncoding(encoding);
+ /*
+ * registered set of known encodings
+ */
+ if (enc != XML_CHAR_ENCODING_ERROR) {
+ xmlSwitchEncoding(ctxt, enc);
+ if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
+ htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
+ "Unsupported encoding %s\n",
+ (const xmlChar *) encoding, NULL);
+ }
+ } else {
+ /*
+ * fallback for unknown encodings
+ */
+ handler = xmlFindCharEncodingHandler((const char *) encoding);
+ if (handler != NULL) {
+ xmlSwitchToEncoding(ctxt, handler);
+ } else {
+ htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
+ "Unsupported encoding %s\n",
+ (const xmlChar *) encoding, NULL);
+ }
+ }
+ }
+ return(ctxt);
+}
+
+#ifdef LIBXML_PUSH_ENABLED
+/************************************************************************
+ * *
+ * Progressive parsing interfaces *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlParseLookupSequence:
+ * @ctxt: an HTML parser context
+ * @first: the first char to lookup
+ * @next: the next char to lookup or zero
+ * @third: the next char to lookup or zero
+ * @comment: flag to force checking inside comments
+ *
+ * Try to find if a sequence (first, next, third) or just (first next) or
+ * (first) is available in the input stream.
+ * This function has a side effect of (possibly) incrementing ctxt->checkIndex
+ * to avoid rescanning sequences of bytes, it DOES change the state of the
+ * parser, do not use liberally.
+ * This is basically similar to xmlParseLookupSequence()
+ *
+ * Returns the index to the current parsing point if the full sequence
+ * is available, -1 otherwise.
+ */
+static int
+htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
+ xmlChar next, xmlChar third, int iscomment,
+ int ignoreattrval)
+{
+ int base, len;
+ htmlParserInputPtr in;
+ const xmlChar *buf;
+ int incomment = 0;
+ int invalue = 0;
+ char valdellim = 0x0;
+
+ in = ctxt->input;
+ if (in == NULL)
+ return (-1);
+
+ base = in->cur - in->base;
+ if (base < 0)
+ return (-1);
+
+ if (ctxt->checkIndex > base)
+ base = ctxt->checkIndex;
+
+ if (in->buf == NULL) {
+ buf = in->base;
+ len = in->length;
+ } else {
+ buf = xmlBufContent(in->buf->buffer);
+ len = xmlBufUse(in->buf->buffer);
+ }
+
+ /* take into account the sequence length */
+ if (third)
+ len -= 2;
+ else if (next)
+ len--;
+ for (; base < len; base++) {
+ if ((!incomment) && (base + 4 < len) && (!iscomment)) {
+ if ((buf[base] == '<') && (buf[base + 1] == '!') &&
+ (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
+ incomment = 1;
+ /* do not increment past <! - some people use <!--> */
+ base += 2;
+ }
+ }
+ if (ignoreattrval) {
+ if (buf[base] == '"' || buf[base] == '\'') {
+ if (invalue) {
+ if (buf[base] == valdellim) {
+ invalue = 0;
+ continue;
+ }
+ } else {
+ valdellim = buf[base];
+ invalue = 1;
+ continue;
+ }
+ } else if (invalue) {
+ continue;
+ }
+ }
+ if (incomment) {
+ if (base + 3 > len)
+ return (-1);
+ if ((buf[base] == '-') && (buf[base + 1] == '-') &&
+ (buf[base + 2] == '>')) {
+ incomment = 0;
+ base += 2;
+ }
+ continue;
+ }
+ if (buf[base] == first) {
+ if (third != 0) {
+ if ((buf[base + 1] != next) || (buf[base + 2] != third))
+ continue;
+ } else if (next != 0) {
+ if (buf[base + 1] != next)
+ continue;
+ }
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ if (next == 0)
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c' found at %d\n",
+ first, base);
+ else if (third == 0)
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c%c' found at %d\n",
+ first, next, base);
+ else
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c%c%c' found at %d\n",
+ first, next, third, base);
+#endif
+ return (base - (in->cur - in->base));
+ }
+ }
+ if ((!incomment) && (!invalue))
+ ctxt->checkIndex = base;
+#ifdef DEBUG_PUSH
+ if (next == 0)
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c' failed\n", first);
+ else if (third == 0)
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c%c' failed\n", first, next);
+ else
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: lookup '%c%c%c' failed\n", first, next,
+ third);
+#endif
+ return (-1);
+}
+
+/**
+ * htmlParseLookupChars:
+ * @ctxt: an HTML parser context
+ * @stop: Array of chars, which stop the lookup.
+ * @stopLen: Length of stop-Array
+ *
+ * Try to find if any char of the stop-Array is available in the input
+ * stream.
+ * This function has a side effect of (possibly) incrementing ctxt->checkIndex
+ * to avoid rescanning sequences of bytes, it DOES change the state of the
+ * parser, do not use liberally.
+ *
+ * Returns the index to the current parsing point if a stopChar
+ * is available, -1 otherwise.
+ */
+static int
+htmlParseLookupChars(htmlParserCtxtPtr ctxt, const xmlChar * stop,
+ int stopLen)
+{
+ int base, len;
+ htmlParserInputPtr in;
+ const xmlChar *buf;
+ int incomment = 0;
+ int i;
+
+ in = ctxt->input;
+ if (in == NULL)
+ return (-1);
+
+ base = in->cur - in->base;
+ if (base < 0)
+ return (-1);
+
+ if (ctxt->checkIndex > base)
+ base = ctxt->checkIndex;
+
+ if (in->buf == NULL) {
+ buf = in->base;
+ len = in->length;
+ } else {
+ buf = xmlBufContent(in->buf->buffer);
+ len = xmlBufUse(in->buf->buffer);
+ }
+
+ for (; base < len; base++) {
+ if (!incomment && (base + 4 < len)) {
+ if ((buf[base] == '<') && (buf[base + 1] == '!') &&
+ (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
+ incomment = 1;
+ /* do not increment past <! - some people use <!--> */
+ base += 2;
+ }
+ }
+ if (incomment) {
+ if (base + 3 > len)
+ return (-1);
+ if ((buf[base] == '-') && (buf[base + 1] == '-') &&
+ (buf[base + 2] == '>')) {
+ incomment = 0;
+ base += 2;
+ }
+ continue;
+ }
+ for (i = 0; i < stopLen; ++i) {
+ if (buf[base] == stop[i]) {
+ ctxt->checkIndex = 0;
+ return (base - (in->cur - in->base));
+ }
+ }
+ }
+ ctxt->checkIndex = base;
+ return (-1);
+}
+
+/**
+ * htmlParseTryOrFinish:
+ * @ctxt: an HTML parser context
+ * @terminate: last chunk indicator
+ *
+ * Try to progress on parsing
+ *
+ * Returns zero if no parsing was possible
+ */
+static int
+htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
+ int ret = 0;
+ htmlParserInputPtr in;
+ int avail = 0;
+ xmlChar cur, next;
+
+ htmlParserNodeInfo node_info;
+
+#ifdef DEBUG_PUSH
+ switch (ctxt->instate) {
+ case XML_PARSER_EOF:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try EOF\n"); break;
+ case XML_PARSER_START:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try START\n"); break;
+ case XML_PARSER_MISC:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try MISC\n");break;
+ case XML_PARSER_COMMENT:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try COMMENT\n");break;
+ case XML_PARSER_PROLOG:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try PROLOG\n");break;
+ case XML_PARSER_START_TAG:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try START_TAG\n");break;
+ case XML_PARSER_CONTENT:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try CONTENT\n");break;
+ case XML_PARSER_CDATA_SECTION:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try CDATA_SECTION\n");break;
+ case XML_PARSER_END_TAG:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try END_TAG\n");break;
+ case XML_PARSER_ENTITY_DECL:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try ENTITY_DECL\n");break;
+ case XML_PARSER_ENTITY_VALUE:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try ENTITY_VALUE\n");break;
+ case XML_PARSER_ATTRIBUTE_VALUE:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try ATTRIBUTE_VALUE\n");break;
+ case XML_PARSER_DTD:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try DTD\n");break;
+ case XML_PARSER_EPILOG:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try EPILOG\n");break;
+ case XML_PARSER_PI:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try PI\n");break;
+ case XML_PARSER_SYSTEM_LITERAL:
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: try SYSTEM_LITERAL\n");break;
+ }
+#endif
+
+ while (1) {
+
+ in = ctxt->input;
+ if (in == NULL) break;
+ if (in->buf == NULL)
+ avail = in->length - (in->cur - in->base);
+ else
+ avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
+ if ((avail == 0) && (terminate)) {
+ htmlAutoCloseOnEnd(ctxt);
+ if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
+ /*
+ * SAX: end of the document processing.
+ */
+ ctxt->instate = XML_PARSER_EOF;
+ if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
+ ctxt->sax->endDocument(ctxt->userData);
+ }
+ }
+ if (avail < 1)
+ goto done;
+ cur = in->cur[0];
+ if (cur == 0) {
+ SKIP(1);
+ continue;
+ }
+
+ switch (ctxt->instate) {
+ case XML_PARSER_EOF:
+ /*
+ * Document parsing is done !
+ */
+ goto done;
+ case XML_PARSER_START:
+ /*
+ * Very first chars read from the document flow.
+ */
+ cur = in->cur[0];
+ if (IS_BLANK_CH(cur)) {
+ SKIP_BLANKS;
+ if (in->buf == NULL)
+ avail = in->length - (in->cur - in->base);
+ else
+ avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
+ }
+ if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
+ ctxt->sax->setDocumentLocator(ctxt->userData,
+ &xmlDefaultSAXLocator);
+ if ((ctxt->sax) && (ctxt->sax->startDocument) &&
+ (!ctxt->disableSAX))
+ ctxt->sax->startDocument(ctxt->userData);
+
+ cur = in->cur[0];
+ next = in->cur[1];
+ if ((cur == '<') && (next == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing internal subset\n");
+#endif
+ htmlParseDocTypeDecl(ctxt);
+ ctxt->instate = XML_PARSER_PROLOG;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering PROLOG\n");
+#endif
+ } else {
+ ctxt->instate = XML_PARSER_MISC;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering MISC\n");
+#endif
+ }
+ break;
+ case XML_PARSER_MISC:
+ SKIP_BLANKS;
+ if (in->buf == NULL)
+ avail = in->length - (in->cur - in->base);
+ else
+ avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
+ /*
+ * no chars in buffer
+ */
+ if (avail < 1)
+ goto done;
+ /*
+ * not enouth chars in buffer
+ */
+ if (avail < 2) {
+ if (!terminate)
+ goto done;
+ else
+ next = ' ';
+ } else {
+ next = in->cur[1];
+ }
+ cur = in->cur[0];
+ if ((cur == '<') && (next == '!') &&
+ (in->cur[2] == '-') && (in->cur[3] == '-')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing Comment\n");
+#endif
+ htmlParseComment(ctxt);
+ ctxt->instate = XML_PARSER_MISC;
+ } else if ((cur == '<') && (next == '?')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing PI\n");
+#endif
+ htmlParsePI(ctxt);
+ ctxt->instate = XML_PARSER_MISC;
+ } else if ((cur == '<') && (next == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing internal subset\n");
+#endif
+ htmlParseDocTypeDecl(ctxt);
+ ctxt->instate = XML_PARSER_PROLOG;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering PROLOG\n");
+#endif
+ } else if ((cur == '<') && (next == '!') &&
+ (avail < 9)) {
+ goto done;
+ } else {
+ ctxt->instate = XML_PARSER_START_TAG;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering START_TAG\n");
+#endif
+ }
+ break;
+ case XML_PARSER_PROLOG:
+ SKIP_BLANKS;
+ if (in->buf == NULL)
+ avail = in->length - (in->cur - in->base);
+ else
+ avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
+ if (avail < 2)
+ goto done;
+ cur = in->cur[0];
+ next = in->cur[1];
+ if ((cur == '<') && (next == '!') &&
+ (in->cur[2] == '-') && (in->cur[3] == '-')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing Comment\n");
+#endif
+ htmlParseComment(ctxt);
+ ctxt->instate = XML_PARSER_PROLOG;
+ } else if ((cur == '<') && (next == '?')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing PI\n");
+#endif
+ htmlParsePI(ctxt);
+ ctxt->instate = XML_PARSER_PROLOG;
+ } else if ((cur == '<') && (next == '!') &&
+ (avail < 4)) {
+ goto done;
+ } else {
+ ctxt->instate = XML_PARSER_START_TAG;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering START_TAG\n");
+#endif
+ }
+ break;
+ case XML_PARSER_EPILOG:
+ if (in->buf == NULL)
+ avail = in->length - (in->cur - in->base);
+ else
+ avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
+ if (avail < 1)
+ goto done;
+ cur = in->cur[0];
+ if (IS_BLANK_CH(cur)) {
+ htmlParseCharData(ctxt);
+ goto done;
+ }
+ if (avail < 2)
+ goto done;
+ next = in->cur[1];
+ if ((cur == '<') && (next == '!') &&
+ (in->cur[2] == '-') && (in->cur[3] == '-')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing Comment\n");
+#endif
+ htmlParseComment(ctxt);
+ ctxt->instate = XML_PARSER_EPILOG;
+ } else if ((cur == '<') && (next == '?')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing PI\n");
+#endif
+ htmlParsePI(ctxt);
+ ctxt->instate = XML_PARSER_EPILOG;
+ } else if ((cur == '<') && (next == '!') &&
+ (avail < 4)) {
+ goto done;
+ } else {
+ ctxt->errNo = XML_ERR_DOCUMENT_END;
+ ctxt->wellFormed = 0;
+ ctxt->instate = XML_PARSER_EOF;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering EOF\n");
+#endif
+ if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
+ ctxt->sax->endDocument(ctxt->userData);
+ goto done;
+ }
+ break;
+ case XML_PARSER_START_TAG: {
+ const xmlChar *name;
+ int failed;
+ const htmlElemDesc * info;
+
+ /*
+ * no chars in buffer
+ */
+ if (avail < 1)
+ goto done;
+ /*
+ * not enouth chars in buffer
+ */
+ if (avail < 2) {
+ if (!terminate)
+ goto done;
+ else
+ next = ' ';
+ } else {
+ next = in->cur[1];
+ }
+ cur = in->cur[0];
+ if (cur != '<') {
+ ctxt->instate = XML_PARSER_CONTENT;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ }
+ if (next == '/') {
+ ctxt->instate = XML_PARSER_END_TAG;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering END_TAG\n");
+#endif
+ break;
+ }
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+
+ /* Capture start position */
+ if (ctxt->record_info) {
+ node_info.begin_pos = ctxt->input->consumed +
+ (CUR_PTR - ctxt->input->base);
+ node_info.begin_line = ctxt->input->line;
+ }
+
+
+ failed = htmlParseStartTag(ctxt);
+ name = ctxt->name;
+ if ((failed == -1) ||
+ (name == NULL)) {
+ if (CUR == '>')
+ NEXT;
+ break;
+ }
+
+ /*
+ * Lookup the info for that element.
+ */
+ info = htmlTagLookup(name);
+ if (info == NULL) {
+ htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
+ "Tag %s invalid\n", name, NULL);
+ }
+
+ /*
+ * Check for an Empty Element labeled the XML/SGML way
+ */
+ if ((CUR == '/') && (NXT(1) == '>')) {
+ SKIP(2);
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ ctxt->instate = XML_PARSER_CONTENT;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ }
+
+ if (CUR == '>') {
+ NEXT;
+ } else {
+ htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
+ "Couldn't find end of Start Tag %s\n",
+ name, NULL);
+
+ /*
+ * end of parsing of this node.
+ */
+ if (xmlStrEqual(name, ctxt->name)) {
+ nodePop(ctxt);
+ htmlnamePop(ctxt);
+ }
+
+ if (ctxt->record_info)
+ htmlNodeInfoPush(ctxt, &node_info);
+
+ ctxt->instate = XML_PARSER_CONTENT;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ }
+
+ /*
+ * Check for an Empty Element from DTD definition
+ */
+ if ((info != NULL) && (info->empty)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
+ ctxt->sax->endElement(ctxt->userData, name);
+ htmlnamePop(ctxt);
+ }
+
+ if (ctxt->record_info)
+ htmlNodeInfoPush(ctxt, &node_info);
+
+ ctxt->instate = XML_PARSER_CONTENT;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ }
+ case XML_PARSER_CONTENT: {
+ long cons;
+ /*
+ * Handle preparsed entities and charRef
+ */
+ if (ctxt->token != 0) {
+ xmlChar chr[2] = { 0 , 0 } ;
+
+ chr[0] = (xmlChar) ctxt->token;
+ htmlCheckParagraph(ctxt);
+ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
+ ctxt->sax->characters(ctxt->userData, chr, 1);
+ ctxt->token = 0;
+ ctxt->checkIndex = 0;
+ }
+ if ((avail == 1) && (terminate)) {
+ cur = in->cur[0];
+ if ((cur != '<') && (cur != '&')) {
+ if (ctxt->sax != NULL) {
+ if (IS_BLANK_CH(cur)) {
+ if (ctxt->keepBlanks) {
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(
+ ctxt->userData, &in->cur[0], 1);
+ } else {
+ if (ctxt->sax->ignorableWhitespace != NULL)
+ ctxt->sax->ignorableWhitespace(
+ ctxt->userData, &in->cur[0], 1);
+ }
+ } else {
+ htmlCheckParagraph(ctxt);
+ if (ctxt->sax->characters != NULL)
+ ctxt->sax->characters(
+ ctxt->userData, &in->cur[0], 1);
+ }
+ }
+ ctxt->token = 0;
+ ctxt->checkIndex = 0;
+ in->cur++;
+ break;
+ }
+ }
+ if (avail < 2)
+ goto done;
+ cur = in->cur[0];
+ next = in->cur[1];
+ cons = ctxt->nbChars;
+ if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
+ (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
+ /*
+ * Handle SCRIPT/STYLE separately
+ */
+ if (!terminate) {
+ int idx;
+ xmlChar val;
+
+ idx = htmlParseLookupSequence(ctxt, '<', '/', 0, 0, 0);
+ if (idx < 0)
+ goto done;
+ val = in->cur[idx + 2];
+ if (val == 0) /* bad cut of input */
+ goto done;
+ }
+ htmlParseScript(ctxt);
+ if ((cur == '<') && (next == '/')) {
+ ctxt->instate = XML_PARSER_END_TAG;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering END_TAG\n");
+#endif
+ break;
+ }
+ } else {
+ /*
+ * Sometimes DOCTYPE arrives in the middle of the document
+ */
+ if ((cur == '<') && (next == '!') &&
+ (UPP(2) == 'D') && (UPP(3) == 'O') &&
+ (UPP(4) == 'C') && (UPP(5) == 'T') &&
+ (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+ (UPP(8) == 'E')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+ htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
+ "Misplaced DOCTYPE declaration\n",
+ BAD_CAST "DOCTYPE" , NULL);
+ htmlParseDocTypeDecl(ctxt);
+ } else if ((cur == '<') && (next == '!') &&
+ (in->cur[2] == '-') && (in->cur[3] == '-')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(
+ ctxt, '-', '-', '>', 1, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing Comment\n");
+#endif
+ htmlParseComment(ctxt);
+ ctxt->instate = XML_PARSER_CONTENT;
+ } else if ((cur == '<') && (next == '?')) {
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing PI\n");
+#endif
+ htmlParsePI(ctxt);
+ ctxt->instate = XML_PARSER_CONTENT;
+ } else if ((cur == '<') && (next == '!') && (avail < 4)) {
+ goto done;
+ } else if ((cur == '<') && (next == '/')) {
+ ctxt->instate = XML_PARSER_END_TAG;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering END_TAG\n");
+#endif
+ break;
+ } else if (cur == '<') {
+ ctxt->instate = XML_PARSER_START_TAG;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering START_TAG\n");
+#endif
+ break;
+ } else if (cur == '&') {
+ if ((!terminate) &&
+ (htmlParseLookupChars(ctxt,
+ BAD_CAST "; >/", 4) < 0))
+ goto done;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing Reference\n");
+#endif
+ /* TODO: check generation of subtrees if noent !!! */
+ htmlParseReference(ctxt);
+ } else {
+ /*
+ * check that the text sequence is complete
+ * before handing out the data to the parser
+ * to avoid problems with erroneous end of
+ * data detection.
+ */
+ if ((!terminate) &&
+ (htmlParseLookupChars(ctxt, BAD_CAST "<&", 2) < 0))
+ goto done;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: Parsing char data\n");
+#endif
+ htmlParseCharData(ctxt);
+ }
+ }
+ if (cons == ctxt->nbChars) {
+ if (ctxt->node != NULL) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "detected an error in element content\n",
+ NULL, NULL);
+ }
+ NEXT;
+ break;
+ }
+
+ break;
+ }
+ case XML_PARSER_END_TAG:
+ if (avail < 2)
+ goto done;
+ if ((!terminate) &&
+ (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
+ goto done;
+ htmlParseEndTag(ctxt);
+ if (ctxt->nameNr == 0) {
+ ctxt->instate = XML_PARSER_EPILOG;
+ } else {
+ ctxt->instate = XML_PARSER_CONTENT;
+ }
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_CDATA_SECTION:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == CDATA\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_DTD:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == DTD\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_COMMENT:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == COMMENT\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_PI:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == PI\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_ENTITY_DECL:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == ENTITY_DECL\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_ENTITY_VALUE:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == ENTITY_VALUE\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering DTD\n");
+#endif
+ break;
+ case XML_PARSER_ATTRIBUTE_VALUE:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == ATTRIBUTE_VALUE\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_START_TAG;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering START_TAG\n");
+#endif
+ break;
+ case XML_PARSER_SYSTEM_LITERAL:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_IGNORE:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == XML_PARSER_IGNORE\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+ case XML_PARSER_PUBLIC_LITERAL:
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "HPP: internal error, state == XML_PARSER_LITERAL\n",
+ NULL, NULL);
+ ctxt->instate = XML_PARSER_CONTENT;
+ ctxt->checkIndex = 0;
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext,
+ "HPP: entering CONTENT\n");
+#endif
+ break;
+
+ }
+ }
+done:
+ if ((avail == 0) && (terminate)) {
+ htmlAutoCloseOnEnd(ctxt);
+ if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
+ /*
+ * SAX: end of the document processing.
+ */
+ ctxt->instate = XML_PARSER_EOF;
+ if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
+ ctxt->sax->endDocument(ctxt->userData);
+ }
+ }
+ if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL) &&
+ ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
+ (ctxt->instate == XML_PARSER_EPILOG))) {
+ xmlDtdPtr dtd;
+ dtd = xmlGetIntSubset(ctxt->myDoc);
+ if (dtd == NULL)
+ ctxt->myDoc->intSubset =
+ xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
+ BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
+ BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
+ }
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
+#endif
+ return(ret);
+}
+
+/**
+ * htmlParseChunk:
+ * @ctxt: an HTML parser context
+ * @chunk: an char array
+ * @size: the size in byte of the chunk
+ * @terminate: last chunk indicator
+ *
+ * Parse a Chunk of memory
+ *
+ * Returns zero if no error, the xmlParserErrors otherwise.
+ */
+int
+htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
+ int terminate) {
+ if ((ctxt == NULL) || (ctxt->input == NULL)) {
+ htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
+ "htmlParseChunk: context error\n", NULL, NULL);
+ return(XML_ERR_INTERNAL_ERROR);
+ }
+ if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
+ (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
+ size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
+ size_t cur = ctxt->input->cur - ctxt->input->base;
+ int res;
+
+ res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
+ if (res < 0) {
+ ctxt->errNo = XML_PARSER_EOF;
+ ctxt->disableSAX = 1;
+ return (XML_PARSER_EOF);
+ }
+ xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
+#endif
+
+#if 0
+ if ((terminate) || (ctxt->input->buf->buffer->use > 80))
+ htmlParseTryOrFinish(ctxt, terminate);
+#endif
+ } else if (ctxt->instate != XML_PARSER_EOF) {
+ if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
+ xmlParserInputBufferPtr in = ctxt->input->buf;
+ if ((in->encoder != NULL) && (in->buffer != NULL) &&
+ (in->raw != NULL)) {
+ int nbchars;
+ size_t base = xmlBufGetInputBase(in->buffer, ctxt->input);
+ size_t current = ctxt->input->cur - ctxt->input->base;
+
+ nbchars = xmlCharEncInput(in, terminate);
+ if (nbchars < 0) {
+ htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
+ "encoder error\n", NULL, NULL);
+ return(XML_ERR_INVALID_ENCODING);
+ }
+ xmlBufSetInputBaseCur(in->buffer, ctxt->input, base, current);
+ }
+ }
+ }
+ htmlParseTryOrFinish(ctxt, terminate);
+ if (terminate) {
+ if ((ctxt->instate != XML_PARSER_EOF) &&
+ (ctxt->instate != XML_PARSER_EPILOG) &&
+ (ctxt->instate != XML_PARSER_MISC)) {
+ ctxt->errNo = XML_ERR_DOCUMENT_END;
+ ctxt->wellFormed = 0;
+ }
+ if (ctxt->instate != XML_PARSER_EOF) {
+ if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
+ ctxt->sax->endDocument(ctxt->userData);
+ }
+ ctxt->instate = XML_PARSER_EOF;
+ }
+ return((xmlParserErrors) ctxt->errNo);
+}
+
+/************************************************************************
+ * *
+ * User entry points *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlCreatePushParserCtxt:
+ * @sax: a SAX handler
+ * @user_data: The user data returned on SAX callbacks
+ * @chunk: a pointer to an array of chars
+ * @size: number of chars in the array
+ * @filename: an optional file name or URI
+ * @enc: an optional encoding
+ *
+ * Create a parser context for using the HTML parser in push mode
+ * The value of @filename is used for fetching external entities
+ * and error/warning reports.
+ *
+ * Returns the new parser context or NULL
+ */
+htmlParserCtxtPtr
+htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
+ const char *chunk, int size, const char *filename,
+ xmlCharEncoding enc) {
+ htmlParserCtxtPtr ctxt;
+ htmlParserInputPtr inputStream;
+ xmlParserInputBufferPtr buf;
+
+ xmlInitParser();
+
+ buf = xmlAllocParserInputBuffer(enc);
+ if (buf == NULL) return(NULL);
+
+ ctxt = htmlNewParserCtxt();
+ if (ctxt == NULL) {
+ xmlFreeParserInputBuffer(buf);
+ return(NULL);
+ }
+ if(enc==XML_CHAR_ENCODING_UTF8 || buf->encoder)
+ ctxt->charset=XML_CHAR_ENCODING_UTF8;
+ if (sax != NULL) {
+ if (ctxt->sax != (xmlSAXHandlerPtr) &htmlDefaultSAXHandler)
+ xmlFree(ctxt->sax);
+ ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
+ if (ctxt->sax == NULL) {
+ xmlFree(buf);
+ xmlFree(ctxt);
+ return(NULL);
+ }
+ memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
+ if (user_data != NULL)
+ ctxt->userData = user_data;
+ }
+ if (filename == NULL) {
+ ctxt->directory = NULL;
+ } else {
+ ctxt->directory = xmlParserGetDirectory(filename);
+ }
+
+ inputStream = htmlNewInputStream(ctxt);
+ if (inputStream == NULL) {
+ xmlFreeParserCtxt(ctxt);
+ xmlFree(buf);
+ return(NULL);
+ }
+
+ if (filename == NULL)
+ inputStream->filename = NULL;
+ else
+ inputStream->filename = (char *)
+ xmlCanonicPath((const xmlChar *) filename);
+ inputStream->buf = buf;
+ xmlBufResetInput(buf->buffer, inputStream);
+
+ inputPush(ctxt, inputStream);
+
+ if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
+ (ctxt->input->buf != NULL)) {
+ size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
+ size_t cur = ctxt->input->cur - ctxt->input->base;
+
+ xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
+
+ xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
+#ifdef DEBUG_PUSH
+ xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
+#endif
+ }
+ ctxt->progressive = 1;
+
+ return(ctxt);
+}
+#endif /* LIBXML_PUSH_ENABLED */
+
+/**
+ * htmlSAXParseDoc:
+ * @cur: a pointer to an array of xmlChar
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ * @sax: the SAX handler block
+ * @userData: if using SAX, this pointer will be provided on callbacks.
+ *
+ * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
+ * to handle parse events. If sax is NULL, fallback to the default DOM
+ * behavior and return a tree.
+ *
+ * Returns the resulting document tree unless SAX is NULL or the document is
+ * not well formed.
+ */
+
+htmlDocPtr
+htmlSAXParseDoc(const xmlChar *cur, const char *encoding,
+ htmlSAXHandlerPtr sax, void *userData) {
+ htmlDocPtr ret;
+ htmlParserCtxtPtr ctxt;
+
+ xmlInitParser();
+
+ if (cur == NULL) return(NULL);
+
+
+ ctxt = htmlCreateDocParserCtxt(cur, encoding);
+ if (ctxt == NULL) return(NULL);
+ if (sax != NULL) {
+ if (ctxt->sax != NULL) xmlFree (ctxt->sax);
+ ctxt->sax = sax;
+ ctxt->userData = userData;
+ }
+
+ htmlParseDocument(ctxt);
+ ret = ctxt->myDoc;
+ if (sax != NULL) {
+ ctxt->sax = NULL;
+ ctxt->userData = NULL;
+ }
+ htmlFreeParserCtxt(ctxt);
+
+ return(ret);
+}
+
+/**
+ * htmlParseDoc:
+ * @cur: a pointer to an array of xmlChar
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ *
+ * parse an HTML in-memory document and build a tree.
+ *
+ * Returns the resulting document tree
+ */
+
+htmlDocPtr
+htmlParseDoc(const xmlChar *cur, const char *encoding) {
+ return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
+}
+
+
+/**
+ * htmlCreateFileParserCtxt:
+ * @filename: the filename
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ *
+ * Create a parser context for a file content.
+ * Automatic support for ZLIB/Compress compressed document is provided
+ * by default if found at compile-time.
+ *
+ * Returns the new parser context or NULL
+ */
+htmlParserCtxtPtr
+htmlCreateFileParserCtxt(const char *filename, const char *encoding)
+{
+ htmlParserCtxtPtr ctxt;
+ htmlParserInputPtr inputStream;
+ char *canonicFilename;
+ /* htmlCharEncoding enc; */
+ xmlChar *content, *content_line = (xmlChar *) "charset=";
+
+ if (filename == NULL)
+ return(NULL);
+
+ ctxt = htmlNewParserCtxt();
+ if (ctxt == NULL) {
+ return(NULL);
+ }
+ canonicFilename = (char *) xmlCanonicPath((const xmlChar *) filename);
+ if (canonicFilename == NULL) {
+#ifdef LIBXML_SAX1_ENABLED
+ if (xmlDefaultSAXHandler.error != NULL) {
+ xmlDefaultSAXHandler.error(NULL, "out of memory\n");
+ }
+#endif
+ xmlFreeParserCtxt(ctxt);
+ return(NULL);
+ }
+
+ inputStream = xmlLoadExternalEntity(canonicFilename, NULL, ctxt);
+ xmlFree(canonicFilename);
+ if (inputStream == NULL) {
+ xmlFreeParserCtxt(ctxt);
+ return(NULL);
+ }
+
+ inputPush(ctxt, inputStream);
+
+ /* set encoding */
+ if (encoding) {
+ size_t l = strlen(encoding);
+
+ if (l < 1000) {
+ content = xmlMallocAtomic (xmlStrlen(content_line) + l + 1);
+ if (content) {
+ strcpy ((char *)content, (char *)content_line);
+ strcat ((char *)content, (char *)encoding);
+ htmlCheckEncoding (ctxt, content);
+ xmlFree (content);
+ }
+ }
+ }
+
+ return(ctxt);
+}
+
+/**
+ * htmlSAXParseFile:
+ * @filename: the filename
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ * @sax: the SAX handler block
+ * @userData: if using SAX, this pointer will be provided on callbacks.
+ *
+ * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
+ * compressed document is provided by default if found at compile-time.
+ * It use the given SAX function block to handle the parsing callback.
+ * If sax is NULL, fallback to the default DOM tree building routines.
+ *
+ * Returns the resulting document tree unless SAX is NULL or the document is
+ * not well formed.
+ */
+
+htmlDocPtr
+htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
+ void *userData) {
+ htmlDocPtr ret;
+ htmlParserCtxtPtr ctxt;
+ htmlSAXHandlerPtr oldsax = NULL;
+
+ xmlInitParser();
+
+ ctxt = htmlCreateFileParserCtxt(filename, encoding);
+ if (ctxt == NULL) return(NULL);
+ if (sax != NULL) {
+ oldsax = ctxt->sax;
+ ctxt->sax = sax;
+ ctxt->userData = userData;
+ }
+
+ htmlParseDocument(ctxt);
+
+ ret = ctxt->myDoc;
+ if (sax != NULL) {
+ ctxt->sax = oldsax;
+ ctxt->userData = NULL;
+ }
+ htmlFreeParserCtxt(ctxt);
+
+ return(ret);
+}
+
+/**
+ * htmlParseFile:
+ * @filename: the filename
+ * @encoding: a free form C string describing the HTML document encoding, or NULL
+ *
+ * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
+ * compressed document is provided by default if found at compile-time.
+ *
+ * Returns the resulting document tree
+ */
+
+htmlDocPtr
+htmlParseFile(const char *filename, const char *encoding) {
+ return(htmlSAXParseFile(filename, encoding, NULL, NULL));
+}
+
+/**
+ * htmlHandleOmittedElem:
+ * @val: int 0 or 1
+ *
+ * Set and return the previous value for handling HTML omitted tags.
+ *
+ * Returns the last value for 0 for no handling, 1 for auto insertion.
+ */
+
+int
+htmlHandleOmittedElem(int val) {
+ int old = htmlOmittedDefaultValue;
+
+ htmlOmittedDefaultValue = val;
+ return(old);
+}
+
+/**
+ * htmlElementAllowedHere:
+ * @parent: HTML parent element
+ * @elt: HTML element
+ *
+ * Checks whether an HTML element may be a direct child of a parent element.
+ * Note - doesn't check for deprecated elements
+ *
+ * Returns 1 if allowed; 0 otherwise.
+ */
+int
+htmlElementAllowedHere(const htmlElemDesc* parent, const xmlChar* elt) {
+ const char** p ;
+
+ if ( ! elt || ! parent || ! parent->subelts )
+ return 0 ;
+
+ for ( p = parent->subelts; *p; ++p )
+ if ( !xmlStrcmp((const xmlChar *)*p, elt) )
+ return 1 ;
+
+ return 0 ;
+}
+/**
+ * htmlElementStatusHere:
+ * @parent: HTML parent element
+ * @elt: HTML element
+ *
+ * Checks whether an HTML element may be a direct child of a parent element.
+ * and if so whether it is valid or deprecated.
+ *
+ * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
+ */
+htmlStatus
+htmlElementStatusHere(const htmlElemDesc* parent, const htmlElemDesc* elt) {
+ if ( ! parent || ! elt )
+ return HTML_INVALID ;
+ if ( ! htmlElementAllowedHere(parent, (const xmlChar*) elt->name ) )
+ return HTML_INVALID ;
+
+ return ( elt->dtd == 0 ) ? HTML_VALID : HTML_DEPRECATED ;
+}
+/**
+ * htmlAttrAllowed:
+ * @elt: HTML element
+ * @attr: HTML attribute
+ * @legacy: whether to allow deprecated attributes
+ *
+ * Checks whether an attribute is valid for an element
+ * Has full knowledge of Required and Deprecated attributes
+ *
+ * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
+ */
+htmlStatus
+htmlAttrAllowed(const htmlElemDesc* elt, const xmlChar* attr, int legacy) {
+ const char** p ;
+
+ if ( !elt || ! attr )
+ return HTML_INVALID ;
+
+ if ( elt->attrs_req )
+ for ( p = elt->attrs_req; *p; ++p)
+ if ( !xmlStrcmp((const xmlChar*)*p, attr) )
+ return HTML_REQUIRED ;
+
+ if ( elt->attrs_opt )
+ for ( p = elt->attrs_opt; *p; ++p)
+ if ( !xmlStrcmp((const xmlChar*)*p, attr) )
+ return HTML_VALID ;
+
+ if ( legacy && elt->attrs_depr )
+ for ( p = elt->attrs_depr; *p; ++p)
+ if ( !xmlStrcmp((const xmlChar*)*p, attr) )
+ return HTML_DEPRECATED ;
+
+ return HTML_INVALID ;
+}
+/**
+ * htmlNodeStatus:
+ * @node: an htmlNodePtr in a tree
+ * @legacy: whether to allow deprecated elements (YES is faster here
+ * for Element nodes)
+ *
+ * Checks whether the tree node is valid. Experimental (the author
+ * only uses the HTML enhancements in a SAX parser)
+ *
+ * Return: for Element nodes, a return from htmlElementAllowedHere (if
+ * legacy allowed) or htmlElementStatusHere (otherwise).
+ * for Attribute nodes, a return from htmlAttrAllowed
+ * for other nodes, HTML_NA (no checks performed)
+ */
+htmlStatus
+htmlNodeStatus(const htmlNodePtr node, int legacy) {
+ if ( ! node )
+ return HTML_INVALID ;
+
+ switch ( node->type ) {
+ case XML_ELEMENT_NODE:
+ return legacy
+ ? ( htmlElementAllowedHere (
+ htmlTagLookup(node->parent->name) , node->name
+ ) ? HTML_VALID : HTML_INVALID )
+ : htmlElementStatusHere(
+ htmlTagLookup(node->parent->name) ,
+ htmlTagLookup(node->name) )
+ ;
+ case XML_ATTRIBUTE_NODE:
+ return htmlAttrAllowed(
+ htmlTagLookup(node->parent->name) , node->name, legacy) ;
+ default: return HTML_NA ;
+ }
+}
+/************************************************************************
+ * *
+ * New set (2.6.0) of simpler and more flexible APIs *
+ * *
+ ************************************************************************/
+/**
+ * DICT_FREE:
+ * @str: a string
+ *
+ * Free a string if it is not owned by the "dict" dictionary in the
+ * current scope
+ */
+#define DICT_FREE(str) \
+ if ((str) && ((!dict) || \
+ (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
+ xmlFree((char *)(str));
+
+/**
+ * htmlCtxtReset:
+ * @ctxt: an HTML parser context
+ *
+ * Reset a parser context
+ */
+void
+htmlCtxtReset(htmlParserCtxtPtr ctxt)
+{
+ xmlParserInputPtr input;
+ xmlDictPtr dict;
+
+ if (ctxt == NULL)
+ return;
+
+ xmlInitParser();
+ dict = ctxt->dict;
+
+ while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
+ xmlFreeInputStream(input);
+ }
+ ctxt->inputNr = 0;
+ ctxt->input = NULL;
+
+ ctxt->spaceNr = 0;
+ if (ctxt->spaceTab != NULL) {
+ ctxt->spaceTab[0] = -1;
+ ctxt->space = &ctxt->spaceTab[0];
+ } else {
+ ctxt->space = NULL;
+ }
+
+
+ ctxt->nodeNr = 0;
+ ctxt->node = NULL;
+
+ ctxt->nameNr = 0;
+ ctxt->name = NULL;
+
+ DICT_FREE(ctxt->version);
+ ctxt->version = NULL;
+ DICT_FREE(ctxt->encoding);
+ ctxt->encoding = NULL;
+ DICT_FREE(ctxt->directory);
+ ctxt->directory = NULL;
+ DICT_FREE(ctxt->extSubURI);
+ ctxt->extSubURI = NULL;
+ DICT_FREE(ctxt->extSubSystem);
+ ctxt->extSubSystem = NULL;
+ if (ctxt->myDoc != NULL)
+ xmlFreeDoc(ctxt->myDoc);
+ ctxt->myDoc = NULL;
+
+ ctxt->standalone = -1;
+ ctxt->hasExternalSubset = 0;
+ ctxt->hasPErefs = 0;
+ ctxt->html = 1;
+ ctxt->external = 0;
+ ctxt->instate = XML_PARSER_START;
+ ctxt->token = 0;
+
+ ctxt->wellFormed = 1;
+ ctxt->nsWellFormed = 1;
+ ctxt->disableSAX = 0;
+ ctxt->valid = 1;
+ ctxt->vctxt.userData = ctxt;
+ ctxt->vctxt.error = xmlParserValidityError;
+ ctxt->vctxt.warning = xmlParserValidityWarning;
+ ctxt->record_info = 0;
+ ctxt->nbChars = 0;
+ ctxt->checkIndex = 0;
+ ctxt->inSubset = 0;
+ ctxt->errNo = XML_ERR_OK;
+ ctxt->depth = 0;
+ ctxt->charset = XML_CHAR_ENCODING_NONE;
+ ctxt->catalogs = NULL;
+ xmlInitNodeInfoSeq(&ctxt->node_seq);
+
+ if (ctxt->attsDefault != NULL) {
+ xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator);
+ ctxt->attsDefault = NULL;
+ }
+ if (ctxt->attsSpecial != NULL) {
+ xmlHashFree(ctxt->attsSpecial, NULL);
+ ctxt->attsSpecial = NULL;
+ }
+}
+
+/**
+ * htmlCtxtUseOptions:
+ * @ctxt: an HTML parser context
+ * @options: a combination of htmlParserOption(s)
+ *
+ * Applies the options to the parser context
+ *
+ * Returns 0 in case of success, the set of unknown or unimplemented options
+ * in case of error.
+ */
+int
+htmlCtxtUseOptions(htmlParserCtxtPtr ctxt, int options)
+{
+ if (ctxt == NULL)
+ return(-1);
+
+ if (options & HTML_PARSE_NOWARNING) {
+ ctxt->sax->warning = NULL;
+ ctxt->vctxt.warning = NULL;
+ options -= XML_PARSE_NOWARNING;
+ ctxt->options |= XML_PARSE_NOWARNING;
+ }
+ if (options & HTML_PARSE_NOERROR) {
+ ctxt->sax->error = NULL;
+ ctxt->vctxt.error = NULL;
+ ctxt->sax->fatalError = NULL;
+ options -= XML_PARSE_NOERROR;
+ ctxt->options |= XML_PARSE_NOERROR;
+ }
+ if (options & HTML_PARSE_PEDANTIC) {
+ ctxt->pedantic = 1;
+ options -= XML_PARSE_PEDANTIC;
+ ctxt->options |= XML_PARSE_PEDANTIC;
+ } else
+ ctxt->pedantic = 0;
+ if (options & XML_PARSE_NOBLANKS) {
+ ctxt->keepBlanks = 0;
+ ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
+ options -= XML_PARSE_NOBLANKS;
+ ctxt->options |= XML_PARSE_NOBLANKS;
+ } else
+ ctxt->keepBlanks = 1;
+ if (options & HTML_PARSE_RECOVER) {
+ ctxt->recovery = 1;
+ options -= HTML_PARSE_RECOVER;
+ } else
+ ctxt->recovery = 0;
+ if (options & HTML_PARSE_COMPACT) {
+ ctxt->options |= HTML_PARSE_COMPACT;
+ options -= HTML_PARSE_COMPACT;
+ }
+ if (options & XML_PARSE_HUGE) {
+ ctxt->options |= XML_PARSE_HUGE;
+ options -= XML_PARSE_HUGE;
+ }
+ if (options & HTML_PARSE_NODEFDTD) {
+ ctxt->options |= HTML_PARSE_NODEFDTD;
+ options -= HTML_PARSE_NODEFDTD;
+ }
+ if (options & HTML_PARSE_IGNORE_ENC) {
+ ctxt->options |= HTML_PARSE_IGNORE_ENC;
+ options -= HTML_PARSE_IGNORE_ENC;
+ }
+ if (options & HTML_PARSE_NOIMPLIED) {
+ ctxt->options |= HTML_PARSE_NOIMPLIED;
+ options -= HTML_PARSE_NOIMPLIED;
+ }
+ ctxt->dictNames = 0;
+ return (options);
+}
+
+/**
+ * htmlDoRead:
+ * @ctxt: an HTML parser context
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ * @reuse: keep the context for reuse
+ *
+ * Common front-end for the htmlRead functions
+ *
+ * Returns the resulting document tree or NULL
+ */
+static htmlDocPtr
+htmlDoRead(htmlParserCtxtPtr ctxt, const char *URL, const char *encoding,
+ int options, int reuse)
+{
+ htmlDocPtr ret;
+
+ htmlCtxtUseOptions(ctxt, options);
+ ctxt->html = 1;
+ if (encoding != NULL) {
+ xmlCharEncodingHandlerPtr hdlr;
+
+ hdlr = xmlFindCharEncodingHandler(encoding);
+ if (hdlr != NULL) {
+ xmlSwitchToEncoding(ctxt, hdlr);
+ if (ctxt->input->encoding != NULL)
+ xmlFree((xmlChar *) ctxt->input->encoding);
+ ctxt->input->encoding = xmlStrdup((xmlChar *)encoding);
+ }
+ }
+ if ((URL != NULL) && (ctxt->input != NULL) &&
+ (ctxt->input->filename == NULL))
+ ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL);
+ htmlParseDocument(ctxt);
+ ret = ctxt->myDoc;
+ ctxt->myDoc = NULL;
+ if (!reuse) {
+ if ((ctxt->dictNames) &&
+ (ret != NULL) &&
+ (ret->dict == ctxt->dict))
+ ctxt->dict = NULL;
+ xmlFreeParserCtxt(ctxt);
+ }
+ return (ret);
+}
+
+/**
+ * htmlReadDoc:
+ * @cur: a pointer to a zero terminated string
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML in-memory document and build a tree.
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options)
+{
+ htmlParserCtxtPtr ctxt;
+
+ if (cur == NULL)
+ return (NULL);
+
+ xmlInitParser();
+ ctxt = htmlCreateDocParserCtxt(cur, NULL);
+ if (ctxt == NULL)
+ return (NULL);
+ return (htmlDoRead(ctxt, URL, encoding, options, 0));
+}
+
+/**
+ * htmlReadFile:
+ * @filename: a file or URL
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML file from the filesystem or the network.
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlReadFile(const char *filename, const char *encoding, int options)
+{
+ htmlParserCtxtPtr ctxt;
+
+ xmlInitParser();
+ ctxt = htmlCreateFileParserCtxt(filename, encoding);
+ if (ctxt == NULL)
+ return (NULL);
+ return (htmlDoRead(ctxt, NULL, NULL, options, 0));
+}
+
+/**
+ * htmlReadMemory:
+ * @buffer: a pointer to a char array
+ * @size: the size of the array
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML in-memory document and build a tree.
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options)
+{
+ htmlParserCtxtPtr ctxt;
+
+ xmlInitParser();
+ ctxt = xmlCreateMemoryParserCtxt(buffer, size);
+ if (ctxt == NULL)
+ return (NULL);
+ htmlDefaultSAXHandlerInit();
+ if (ctxt->sax != NULL)
+ memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
+ return (htmlDoRead(ctxt, URL, encoding, options, 0));
+}
+
+/**
+ * htmlReadFd:
+ * @fd: an open file descriptor
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML from a file descriptor and build a tree.
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlReadFd(int fd, const char *URL, const char *encoding, int options)
+{
+ htmlParserCtxtPtr ctxt;
+ xmlParserInputBufferPtr input;
+ xmlParserInputPtr stream;
+
+ if (fd < 0)
+ return (NULL);
+ xmlInitParser();
+
+ xmlInitParser();
+ input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
+ if (input == NULL)
+ return (NULL);
+ ctxt = xmlNewParserCtxt();
+ if (ctxt == NULL) {
+ xmlFreeParserInputBuffer(input);
+ return (NULL);
+ }
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+ if (stream == NULL) {
+ xmlFreeParserInputBuffer(input);
+ xmlFreeParserCtxt(ctxt);
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 0));
+}
+
+/**
+ * htmlReadIO:
+ * @ioread: an I/O read function
+ * @ioclose: an I/O close function
+ * @ioctx: an I/O handler
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an HTML document from I/O functions and source and build a tree.
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose,
+ void *ioctx, const char *URL, const char *encoding, int options)
+{
+ htmlParserCtxtPtr ctxt;
+ xmlParserInputBufferPtr input;
+ xmlParserInputPtr stream;
+
+ if (ioread == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
+ XML_CHAR_ENCODING_NONE);
+ if (input == NULL) {
+ if (ioclose != NULL)
+ ioclose(ioctx);
+ return (NULL);
+ }
+ ctxt = htmlNewParserCtxt();
+ if (ctxt == NULL) {
+ xmlFreeParserInputBuffer(input);
+ return (NULL);
+ }
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+ if (stream == NULL) {
+ xmlFreeParserInputBuffer(input);
+ xmlFreeParserCtxt(ctxt);
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 0));
+}
+
+/**
+ * htmlCtxtReadDoc:
+ * @ctxt: an HTML parser context
+ * @cur: a pointer to a zero terminated string
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML in-memory document and build a tree.
+ * This reuses the existing @ctxt parser context
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlCtxtReadDoc(htmlParserCtxtPtr ctxt, const xmlChar * cur,
+ const char *URL, const char *encoding, int options)
+{
+ xmlParserInputPtr stream;
+
+ if (cur == NULL)
+ return (NULL);
+ if (ctxt == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ htmlCtxtReset(ctxt);
+
+ stream = xmlNewStringInputStream(ctxt, cur);
+ if (stream == NULL) {
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 1));
+}
+
+/**
+ * htmlCtxtReadFile:
+ * @ctxt: an HTML parser context
+ * @filename: a file or URL
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML file from the filesystem or the network.
+ * This reuses the existing @ctxt parser context
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlCtxtReadFile(htmlParserCtxtPtr ctxt, const char *filename,
+ const char *encoding, int options)
+{
+ xmlParserInputPtr stream;
+
+ if (filename == NULL)
+ return (NULL);
+ if (ctxt == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ htmlCtxtReset(ctxt);
+
+ stream = xmlLoadExternalEntity(filename, NULL, ctxt);
+ if (stream == NULL) {
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, NULL, encoding, options, 1));
+}
+
+/**
+ * htmlCtxtReadMemory:
+ * @ctxt: an HTML parser context
+ * @buffer: a pointer to a char array
+ * @size: the size of the array
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML in-memory document and build a tree.
+ * This reuses the existing @ctxt parser context
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlCtxtReadMemory(htmlParserCtxtPtr ctxt, const char *buffer, int size,
+ const char *URL, const char *encoding, int options)
+{
+ xmlParserInputBufferPtr input;
+ xmlParserInputPtr stream;
+
+ if (ctxt == NULL)
+ return (NULL);
+ if (buffer == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ htmlCtxtReset(ctxt);
+
+ input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
+ if (input == NULL) {
+ return(NULL);
+ }
+
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+ if (stream == NULL) {
+ xmlFreeParserInputBuffer(input);
+ return(NULL);
+ }
+
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 1));
+}
+
+/**
+ * htmlCtxtReadFd:
+ * @ctxt: an HTML parser context
+ * @fd: an open file descriptor
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an XML from a file descriptor and build a tree.
+ * This reuses the existing @ctxt parser context
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlCtxtReadFd(htmlParserCtxtPtr ctxt, int fd,
+ const char *URL, const char *encoding, int options)
+{
+ xmlParserInputBufferPtr input;
+ xmlParserInputPtr stream;
+
+ if (fd < 0)
+ return (NULL);
+ if (ctxt == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ htmlCtxtReset(ctxt);
+
+
+ input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
+ if (input == NULL)
+ return (NULL);
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+ if (stream == NULL) {
+ xmlFreeParserInputBuffer(input);
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 1));
+}
+
+/**
+ * htmlCtxtReadIO:
+ * @ctxt: an HTML parser context
+ * @ioread: an I/O read function
+ * @ioclose: an I/O close function
+ * @ioctx: an I/O handler
+ * @URL: the base URL to use for the document
+ * @encoding: the document encoding, or NULL
+ * @options: a combination of htmlParserOption(s)
+ *
+ * parse an HTML document from I/O functions and source and build a tree.
+ * This reuses the existing @ctxt parser context
+ *
+ * Returns the resulting document tree
+ */
+htmlDocPtr
+htmlCtxtReadIO(htmlParserCtxtPtr ctxt, xmlInputReadCallback ioread,
+ xmlInputCloseCallback ioclose, void *ioctx,
+ const char *URL,
+ const char *encoding, int options)
+{
+ xmlParserInputBufferPtr input;
+ xmlParserInputPtr stream;
+
+ if (ioread == NULL)
+ return (NULL);
+ if (ctxt == NULL)
+ return (NULL);
+ xmlInitParser();
+
+ htmlCtxtReset(ctxt);
+
+ input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
+ XML_CHAR_ENCODING_NONE);
+ if (input == NULL) {
+ if (ioclose != NULL)
+ ioclose(ioctx);
+ return (NULL);
+ }
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+ if (stream == NULL) {
+ xmlFreeParserInputBuffer(input);
+ return (NULL);
+ }
+ inputPush(ctxt, stream);
+ return (htmlDoRead(ctxt, URL, encoding, options, 1));
+}
+
+#define bottom_HTMLparser
+#include "elfgcchack.h"
+#endif /* LIBXML_HTML_ENABLED */
diff --git a/HTMLtree.c b/HTMLtree.c
new file mode 100644
index 0000000..21cfcfe
--- /dev/null
+++ b/HTMLtree.c
@@ -0,0 +1,1247 @@
+/*
+ * HTMLtree.c : implementation of access function for an HTML tree.
+ *
+ * See Copyright for the status of this software.
+ *
+ * daniel@veillard.com
+ */
+
+
+#define IN_LIBXML
+#include "libxml.h"
+#ifdef LIBXML_HTML_ENABLED
+
+#include <string.h> /* for memset() only ! */
+
+#ifdef HAVE_CTYPE_H
+#include <ctype.h>
+#endif
+#ifdef HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
+
+#include <libxml/xmlmemory.h>
+#include <libxml/HTMLparser.h>
+#include <libxml/HTMLtree.h>
+#include <libxml/entities.h>
+#include <libxml/valid.h>
+#include <libxml/xmlerror.h>
+#include <libxml/parserInternals.h>
+#include <libxml/globals.h>
+#include <libxml/uri.h>
+
+#include "buf.h"
+
+/************************************************************************
+ * *
+ * Getting/Setting encoding meta tags *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlGetMetaEncoding:
+ * @doc: the document
+ *
+ * Encoding definition lookup in the Meta tags
+ *
+ * Returns the current encoding as flagged in the HTML source
+ */
+const xmlChar *
+htmlGetMetaEncoding(htmlDocPtr doc) {
+ htmlNodePtr cur;
+ const xmlChar *content;
+ const xmlChar *encoding;
+
+ if (doc == NULL)
+ return(NULL);
+ cur = doc->children;
+
+ /*
+ * Search the html
+ */
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrEqual(cur->name, BAD_CAST"html"))
+ break;
+ if (xmlStrEqual(cur->name, BAD_CAST"head"))
+ goto found_head;
+ if (xmlStrEqual(cur->name, BAD_CAST"meta"))
+ goto found_meta;
+ }
+ cur = cur->next;
+ }
+ if (cur == NULL)
+ return(NULL);
+ cur = cur->children;
+
+ /*
+ * Search the head
+ */
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrEqual(cur->name, BAD_CAST"head"))
+ break;
+ if (xmlStrEqual(cur->name, BAD_CAST"meta"))
+ goto found_meta;
+ }
+ cur = cur->next;
+ }
+ if (cur == NULL)
+ return(NULL);
+found_head:
+ cur = cur->children;
+
+ /*
+ * Search the meta elements
+ */
+found_meta:
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrEqual(cur->name, BAD_CAST"meta")) {
+ xmlAttrPtr attr = cur->properties;
+ int http;
+ const xmlChar *value;
+
+ content = NULL;
+ http = 0;
+ while (attr != NULL) {
+ if ((attr->children != NULL) &&
+ (attr->children->type == XML_TEXT_NODE) &&
+ (attr->children->next == NULL)) {
+ value = attr->children->content;
+ if ((!xmlStrcasecmp(attr->name, BAD_CAST"http-equiv"))
+ && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
+ http = 1;
+ else if ((value != NULL)
+ && (!xmlStrcasecmp(attr->name, BAD_CAST"content")))
+ content = value;
+ if ((http != 0) && (content != NULL))
+ goto found_content;
+ }
+ attr = attr->next;
+ }
+ }
+ }
+ cur = cur->next;
+ }
+ return(NULL);
+
+found_content:
+ encoding = xmlStrstr(content, BAD_CAST"charset=");
+ if (encoding == NULL)
+ encoding = xmlStrstr(content, BAD_CAST"Charset=");
+ if (encoding == NULL)
+ encoding = xmlStrstr(content, BAD_CAST"CHARSET=");
+ if (encoding != NULL) {
+ encoding += 8;
+ } else {
+ encoding = xmlStrstr(content, BAD_CAST"charset =");
+ if (encoding == NULL)
+ encoding = xmlStrstr(content, BAD_CAST"Charset =");
+ if (encoding == NULL)
+ encoding = xmlStrstr(content, BAD_CAST"CHARSET =");
+ if (encoding != NULL)
+ encoding += 9;
+ }
+ if (encoding != NULL) {
+ while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
+ }
+ return(encoding);
+}
+
+/**
+ * htmlSetMetaEncoding:
+ * @doc: the document
+ * @encoding: the encoding string
+ *
+ * Sets the current encoding in the Meta tags
+ * NOTE: this will not change the document content encoding, just
+ * the META flag associated.
+ *
+ * Returns 0 in case of success and -1 in case of error
+ */
+int
+htmlSetMetaEncoding(htmlDocPtr doc, const xmlChar *encoding) {
+ htmlNodePtr cur, meta = NULL, head = NULL;
+ const xmlChar *content = NULL;
+ char newcontent[100];
+
+ newcontent[0] = 0;
+
+ if (doc == NULL)
+ return(-1);
+
+ /* html isn't a real encoding it's just libxml2 way to get entities */
+ if (!xmlStrcasecmp(encoding, BAD_CAST "html"))
+ return(-1);
+
+ if (encoding != NULL) {
+ snprintf(newcontent, sizeof(newcontent), "text/html; charset=%s",
+ (char *)encoding);
+ newcontent[sizeof(newcontent) - 1] = 0;
+ }
+
+ cur = doc->children;
+
+ /*
+ * Search the html
+ */
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrcasecmp(cur->name, BAD_CAST"html") == 0)
+ break;
+ if (xmlStrcasecmp(cur->name, BAD_CAST"head") == 0)
+ goto found_head;
+ if (xmlStrcasecmp(cur->name, BAD_CAST"meta") == 0)
+ goto found_meta;
+ }
+ cur = cur->next;
+ }
+ if (cur == NULL)
+ return(-1);
+ cur = cur->children;
+
+ /*
+ * Search the head
+ */
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrcasecmp(cur->name, BAD_CAST"head") == 0)
+ break;
+ if (xmlStrcasecmp(cur->name, BAD_CAST"meta") == 0) {
+ head = cur->parent;
+ goto found_meta;
+ }
+ }
+ cur = cur->next;
+ }
+ if (cur == NULL)
+ return(-1);
+found_head:
+ head = cur;
+ if (cur->children == NULL)
+ goto create;
+ cur = cur->children;
+
+found_meta:
+ /*
+ * Search and update all the remaining the meta elements carrying
+ * encoding informations
+ */
+ while (cur != NULL) {
+ if ((cur->type == XML_ELEMENT_NODE) && (cur->name != NULL)) {
+ if (xmlStrcasecmp(cur->name, BAD_CAST"meta") == 0) {
+ xmlAttrPtr attr = cur->properties;
+ int http;
+ const xmlChar *value;
+
+ content = NULL;
+ http = 0;
+ while (attr != NULL) {
+ if ((attr->children != NULL) &&
+ (attr->children->type == XML_TEXT_NODE) &&
+ (attr->children->next == NULL)) {
+ value = attr->children->content;
+ if ((!xmlStrcasecmp(attr->name, BAD_CAST"http-equiv"))
+ && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
+ http = 1;
+ else
+ {
+ if ((value != NULL) &&
+ (!xmlStrcasecmp(attr->name, BAD_CAST"content")))
+ content = value;
+ }
+ if ((http != 0) && (content != NULL))
+ break;
+ }
+ attr = attr->next;
+ }
+ if ((http != 0) && (content != NULL)) {
+ meta = cur;
+ break;
+ }
+
+ }
+ }
+ cur = cur->next;
+ }
+create:
+ if (meta == NULL) {
+ if ((encoding != NULL) && (head != NULL)) {
+ /*
+ * Create a new Meta element with the right attributes
+ */
+
+ meta = xmlNewDocNode(doc, NULL, BAD_CAST"meta", NULL);
+ if (head->children == NULL)
+ xmlAddChild(head, meta);
+ else
+ xmlAddPrevSibling(head->children, meta);
+ xmlNewProp(meta, BAD_CAST"http-equiv", BAD_CAST"Content-Type");
+ xmlNewProp(meta, BAD_CAST"content", BAD_CAST newcontent);
+ }
+ } else {
+ /* remove the meta tag if NULL is passed */
+ if (encoding == NULL) {
+ xmlUnlinkNode(meta);
+ xmlFreeNode(meta);
+ }
+ /* change the document only if there is a real encoding change */
+ else if (xmlStrcasestr(content, encoding) == NULL) {
+ xmlSetProp(meta, BAD_CAST"content", BAD_CAST newcontent);
+ }
+ }
+
+
+ return(0);
+}
+
+/**
+ * booleanHTMLAttrs:
+ *
+ * These are the HTML attributes which will be output
+ * in minimized form, i.e. <option selected="selected"> will be
+ * output as <option selected>, as per XSLT 1.0 16.2 "HTML Output Method"
+ *
+ */
+static const char* htmlBooleanAttrs[] = {
+ "checked", "compact", "declare", "defer", "disabled", "ismap",
+ "multiple", "nohref", "noresize", "noshade", "nowrap", "readonly",
+ "selected", NULL
+};
+
+
+/**
+ * htmlIsBooleanAttr:
+ * @name: the name of the attribute to check
+ *
+ * Determine if a given attribute is a boolean attribute.
+ *
+ * returns: false if the attribute is not boolean, true otherwise.
+ */
+int
+htmlIsBooleanAttr(const xmlChar *name)
+{
+ int i = 0;
+
+ while (htmlBooleanAttrs[i] != NULL) {
+ if (xmlStrcasecmp((const xmlChar *)htmlBooleanAttrs[i], name) == 0)
+ return 1;
+ i++;
+ }
+ return 0;
+}
+
+#ifdef LIBXML_OUTPUT_ENABLED
+/*
+ * private routine exported from xmlIO.c
+ */
+xmlOutputBufferPtr
+xmlAllocOutputBufferInternal(xmlCharEncodingHandlerPtr encoder);
+/************************************************************************
+ * *
+ * Output error handlers *
+ * *
+ ************************************************************************/
+/**
+ * htmlSaveErrMemory:
+ * @extra: extra informations
+ *
+ * Handle an out of memory condition
+ */
+static void
+htmlSaveErrMemory(const char *extra)
+{
+ __xmlSimpleError(XML_FROM_OUTPUT, XML_ERR_NO_MEMORY, NULL, NULL, extra);
+}
+
+/**
+ * htmlSaveErr:
+ * @code: the error number
+ * @node: the location of the error.
+ * @extra: extra informations
+ *
+ * Handle an out of memory condition
+ */
+static void
+htmlSaveErr(int code, xmlNodePtr node, const char *extra)
+{
+ const char *msg = NULL;
+
+ switch(code) {
+ case XML_SAVE_NOT_UTF8:
+ msg = "string is not in UTF-8\n";
+ break;
+ case XML_SAVE_CHAR_INVALID:
+ msg = "invalid character value\n";
+ break;
+ case XML_SAVE_UNKNOWN_ENCODING:
+ msg = "unknown encoding %s\n";
+ break;
+ case XML_SAVE_NO_DOCTYPE:
+ msg = "HTML has no DOCTYPE\n";
+ break;
+ default:
+ msg = "unexpected error number\n";
+ }
+ __xmlSimpleError(XML_FROM_OUTPUT, code, node, msg, extra);
+}
+
+/************************************************************************
+ * *
+ * Dumping HTML tree content to a simple buffer *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlBufNodeDumpFormat:
+ * @buf: the xmlBufPtr output
+ * @doc: the document
+ * @cur: the current node
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too.
+ *
+ * Returns the number of byte written or -1 in case of error
+ */
+static size_t
+htmlBufNodeDumpFormat(xmlBufPtr buf, xmlDocPtr doc, xmlNodePtr cur,
+ int format) {
+ size_t use;
+ int ret;
+ xmlOutputBufferPtr outbuf;
+
+ if (cur == NULL) {
+ return (-1);
+ }
+ if (buf == NULL) {
+ return (-1);
+ }
+ outbuf = (xmlOutputBufferPtr) xmlMalloc(sizeof(xmlOutputBuffer));
+ if (outbuf == NULL) {
+ htmlSaveErrMemory("allocating HTML output buffer");
+ return (-1);
+ }
+ memset(outbuf, 0, (size_t) sizeof(xmlOutputBuffer));
+ outbuf->buffer = buf;
+ outbuf->encoder = NULL;
+ outbuf->writecallback = NULL;
+ outbuf->closecallback = NULL;
+ outbuf->context = NULL;
+ outbuf->written = 0;
+
+ use = xmlBufUse(buf);
+ htmlNodeDumpFormatOutput(outbuf, doc, cur, NULL, format);
+ xmlFree(outbuf);
+ ret = xmlBufUse(buf) - use;
+ return (ret);
+}
+
+/**
+ * htmlNodeDump:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the current node
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too,
+ * and formatting returns are added.
+ *
+ * Returns the number of byte written or -1 in case of error
+ */
+int
+htmlNodeDump(xmlBufferPtr buf, xmlDocPtr doc, xmlNodePtr cur) {
+ xmlBufPtr buffer;
+ size_t ret;
+
+ if ((buf == NULL) || (cur == NULL))
+ return(-1);
+
+ xmlInitParser();
+ buffer = xmlBufFromBuffer(buf);
+ if (buffer == NULL)
+ return(-1);
+
+ ret = htmlBufNodeDumpFormat(buffer, doc, cur, 1);
+
+ xmlBufBackToBuffer(buffer);
+
+ if (ret > INT_MAX)
+ return(-1);
+ return((int) ret);
+}
+
+/**
+ * htmlNodeDumpFileFormat:
+ * @out: the FILE pointer
+ * @doc: the document
+ * @cur: the current node
+ * @encoding: the document encoding
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too.
+ *
+ * TODO: if encoding == NULL try to save in the doc encoding
+ *
+ * returns: the number of byte written or -1 in case of failure.
+ */
+int
+htmlNodeDumpFileFormat(FILE *out, xmlDocPtr doc,
+ xmlNodePtr cur, const char *encoding, int format) {
+ xmlOutputBufferPtr buf;
+ xmlCharEncodingHandlerPtr handler = NULL;
+ int ret;
+
+ xmlInitParser();
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+
+ enc = xmlParseCharEncoding(encoding);
+ if (enc != XML_CHAR_ENCODING_UTF8) {
+ handler = xmlFindCharEncodingHandler(encoding);
+ if (handler == NULL)
+ htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
+ }
+ } else {
+ /*
+ * Fallback to HTML or ASCII when the encoding is unspecified
+ */
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("HTML");
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("ascii");
+ }
+
+ /*
+ * save the content to a temp buffer.
+ */
+ buf = xmlOutputBufferCreateFile(out, handler);
+ if (buf == NULL) return(0);
+
+ htmlNodeDumpFormatOutput(buf, doc, cur, encoding, format);
+
+ ret = xmlOutputBufferClose(buf);
+ return(ret);
+}
+
+/**
+ * htmlNodeDumpFile:
+ * @out: the FILE pointer
+ * @doc: the document
+ * @cur: the current node
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too,
+ * and formatting returns are added.
+ */
+void
+htmlNodeDumpFile(FILE *out, xmlDocPtr doc, xmlNodePtr cur) {
+ htmlNodeDumpFileFormat(out, doc, cur, NULL, 1);
+}
+
+/**
+ * htmlDocDumpMemoryFormat:
+ * @cur: the document
+ * @mem: OUT: the memory pointer
+ * @size: OUT: the memory length
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML document in memory and return the xmlChar * and it's size.
+ * It's up to the caller to free the memory.
+ */
+void
+htmlDocDumpMemoryFormat(xmlDocPtr cur, xmlChar**mem, int *size, int format) {
+ xmlOutputBufferPtr buf;
+ xmlCharEncodingHandlerPtr handler = NULL;
+ const char *encoding;
+
+ xmlInitParser();
+
+ if ((mem == NULL) || (size == NULL))
+ return;
+ if (cur == NULL) {
+ *mem = NULL;
+ *size = 0;
+ return;
+ }
+
+ encoding = (const char *) htmlGetMetaEncoding(cur);
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+
+ enc = xmlParseCharEncoding(encoding);
+ if (enc != XML_CHAR_ENCODING_UTF8) {
+ handler = xmlFindCharEncodingHandler(encoding);
+ if (handler == NULL)
+ htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
+
+ }
+ } else {
+ /*
+ * Fallback to HTML or ASCII when the encoding is unspecified
+ */
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("HTML");
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("ascii");
+ }
+
+ buf = xmlAllocOutputBufferInternal(handler);
+ if (buf == NULL) {
+ *mem = NULL;
+ *size = 0;
+ return;
+ }
+
+ htmlDocContentDumpFormatOutput(buf, cur, NULL, format);
+
+ xmlOutputBufferFlush(buf);
+ if (buf->conv != NULL) {
+ *size = xmlBufUse(buf->conv);
+ *mem = xmlStrndup(xmlBufContent(buf->conv), *size);
+ } else {
+ *size = xmlBufUse(buf->buffer);
+ *mem = xmlStrndup(xmlBufContent(buf->buffer), *size);
+ }
+ (void)xmlOutputBufferClose(buf);
+}
+
+/**
+ * htmlDocDumpMemory:
+ * @cur: the document
+ * @mem: OUT: the memory pointer
+ * @size: OUT: the memory length
+ *
+ * Dump an HTML document in memory and return the xmlChar * and it's size.
+ * It's up to the caller to free the memory.
+ */
+void
+htmlDocDumpMemory(xmlDocPtr cur, xmlChar**mem, int *size) {
+ htmlDocDumpMemoryFormat(cur, mem, size, 1);
+}
+
+
+/************************************************************************
+ * *
+ * Dumping HTML tree content to an I/O output buffer *
+ * *
+ ************************************************************************/
+
+void xmlNsListDumpOutput(xmlOutputBufferPtr buf, xmlNsPtr cur);
+
+/**
+ * htmlDtdDumpOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @encoding: the encoding string
+ *
+ * TODO: check whether encoding is needed
+ *
+ * Dump the HTML document DTD, if any.
+ */
+static void
+htmlDtdDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
+ const char *encoding ATTRIBUTE_UNUSED) {
+ xmlDtdPtr cur = doc->intSubset;
+
+ if (cur == NULL) {
+ htmlSaveErr(XML_SAVE_NO_DOCTYPE, (xmlNodePtr) doc, NULL);
+ return;
+ }
+ xmlOutputBufferWriteString(buf, "<!DOCTYPE ");
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ if (cur->ExternalID != NULL) {
+ xmlOutputBufferWriteString(buf, " PUBLIC ");
+ xmlBufWriteQuotedString(buf->buffer, cur->ExternalID);
+ if (cur->SystemID != NULL) {
+ xmlOutputBufferWriteString(buf, " ");
+ xmlBufWriteQuotedString(buf->buffer, cur->SystemID);
+ }
+ } else if (cur->SystemID != NULL &&
+ xmlStrcmp(cur->SystemID, BAD_CAST "about:legacy-compat")) {
+ xmlOutputBufferWriteString(buf, " SYSTEM ");
+ xmlBufWriteQuotedString(buf->buffer, cur->SystemID);
+ }
+ xmlOutputBufferWriteString(buf, ">\n");
+}
+
+/**
+ * htmlAttrDumpOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the attribute pointer
+ * @encoding: the encoding string
+ *
+ * Dump an HTML attribute
+ */
+static void
+htmlAttrDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur,
+ const char *encoding ATTRIBUTE_UNUSED) {
+ xmlChar *value;
+
+ /*
+ * The html output method should not escape a & character
+ * occurring in an attribute value immediately followed by
+ * a { character (see Section B.7.1 of the HTML 4.0 Recommendation).
+ * This is implemented in xmlEncodeEntitiesReentrant
+ */
+
+ if (cur == NULL) {
+ return;
+ }
+ xmlOutputBufferWriteString(buf, " ");
+ if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
+ xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
+ xmlOutputBufferWriteString(buf, ":");
+ }
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ if ((cur->children != NULL) && (!htmlIsBooleanAttr(cur->name))) {
+ value = xmlNodeListGetString(doc, cur->children, 0);
+ if (value) {
+ xmlOutputBufferWriteString(buf, "=");
+ if ((cur->ns == NULL) && (cur->parent != NULL) &&
+ (cur->parent->ns == NULL) &&
+ ((!xmlStrcasecmp(cur->name, BAD_CAST "href")) ||
+ (!xmlStrcasecmp(cur->name, BAD_CAST "action")) ||
+ (!xmlStrcasecmp(cur->name, BAD_CAST "src")) ||
+ ((!xmlStrcasecmp(cur->name, BAD_CAST "name")) &&
+ (!xmlStrcasecmp(cur->parent->name, BAD_CAST "a"))))) {
+ xmlChar *tmp = value;
+ /* xmlURIEscapeStr() escapes '"' so it can be safely used. */
+ xmlBufCCat(buf->buffer, "\"");
+
+ while (IS_BLANK_CH(*tmp)) tmp++;
+
+ /* URI Escape everything, except server side includes. */
+ for ( ; ; ) {
+ xmlChar *escaped;
+ xmlChar endChar;
+ xmlChar *end = NULL;
+ xmlChar *start = (xmlChar *)xmlStrstr(tmp, BAD_CAST "<!--");
+ if (start != NULL) {
+ end = (xmlChar *)xmlStrstr(tmp, BAD_CAST "-->");
+ if (end != NULL) {
+ *start = '\0';
+ }
+ }
+
+ /* Escape the whole string, or until start (set to '\0'). */
+ escaped = xmlURIEscapeStr(tmp, BAD_CAST"@/:=?;#%&,+");
+ if (escaped != NULL) {
+ xmlBufCat(buf->buffer, escaped);
+ xmlFree(escaped);
+ } else {
+ xmlBufCat(buf->buffer, tmp);
+ }
+
+ if (end == NULL) { /* Everything has been written. */
+ break;
+ }
+
+ /* Do not escape anything within server side includes. */
+ *start = '<'; /* Restore the first character of "<!--". */
+ end += 3; /* strlen("-->") */
+ endChar = *end;
+ *end = '\0';
+ xmlBufCat(buf->buffer, start);
+ *end = endChar;
+ tmp = end;
+ }
+
+ xmlBufCCat(buf->buffer, "\"");
+ } else {
+ xmlBufWriteQuotedString(buf->buffer, value);
+ }
+ xmlFree(value);
+ } else {
+ xmlOutputBufferWriteString(buf, "=\"\"");
+ }
+ }
+}
+
+/**
+ * htmlAttrListDumpOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the first attribute pointer
+ * @encoding: the encoding string
+ *
+ * Dump a list of HTML attributes
+ */
+static void
+htmlAttrListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur, const char *encoding) {
+ if (cur == NULL) {
+ return;
+ }
+ while (cur != NULL) {
+ htmlAttrDumpOutput(buf, doc, cur, encoding);
+ cur = cur->next;
+ }
+}
+
+
+
+/**
+ * htmlNodeListDumpOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the first node
+ * @encoding: the encoding string
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML node list, recursive behaviour,children are printed too.
+ */
+static void
+htmlNodeListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
+ xmlNodePtr cur, const char *encoding, int format) {
+ if (cur == NULL) {
+ return;
+ }
+ while (cur != NULL) {
+ htmlNodeDumpFormatOutput(buf, doc, cur, encoding, format);
+ cur = cur->next;
+ }
+}
+
+/**
+ * htmlNodeDumpFormatOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the current node
+ * @encoding: the encoding string
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too.
+ */
+void
+htmlNodeDumpFormatOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
+ xmlNodePtr cur, const char *encoding, int format) {
+ const htmlElemDesc * info;
+
+ xmlInitParser();
+
+ if ((cur == NULL) || (buf == NULL)) {
+ return;
+ }
+ /*
+ * Special cases.
+ */
+ if (cur->type == XML_DTD_NODE)
+ return;
+ if ((cur->type == XML_HTML_DOCUMENT_NODE) ||
+ (cur->type == XML_DOCUMENT_NODE)){
+ htmlDocContentDumpOutput(buf, (xmlDocPtr) cur, encoding);
+ return;
+ }
+ if (cur->type == XML_ATTRIBUTE_NODE) {
+ htmlAttrDumpOutput(buf, doc, (xmlAttrPtr) cur, encoding);
+ return;
+ }
+ if (cur->type == HTML_TEXT_NODE) {
+ if (cur->content != NULL) {
+ if (((cur->name == (const xmlChar *)xmlStringText) ||
+ (cur->name != (const xmlChar *)xmlStringTextNoenc)) &&
+ ((cur->parent == NULL) ||
+ ((xmlStrcasecmp(cur->parent->name, BAD_CAST "script")) &&
+ (xmlStrcasecmp(cur->parent->name, BAD_CAST "style"))))) {
+ xmlChar *buffer;
+
+ buffer = xmlEncodeEntitiesReentrant(doc, cur->content);
+ if (buffer != NULL) {
+ xmlOutputBufferWriteString(buf, (const char *)buffer);
+ xmlFree(buffer);
+ }
+ } else {
+ xmlOutputBufferWriteString(buf, (const char *)cur->content);
+ }
+ }
+ return;
+ }
+ if (cur->type == HTML_COMMENT_NODE) {
+ if (cur->content != NULL) {
+ xmlOutputBufferWriteString(buf, "<!--");
+ xmlOutputBufferWriteString(buf, (const char *)cur->content);
+ xmlOutputBufferWriteString(buf, "-->");
+ }
+ return;
+ }
+ if (cur->type == HTML_PI_NODE) {
+ if (cur->name == NULL)
+ return;
+ xmlOutputBufferWriteString(buf, "<?");
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ if (cur->content != NULL) {
+ xmlOutputBufferWriteString(buf, " ");
+ xmlOutputBufferWriteString(buf, (const char *)cur->content);
+ }
+ xmlOutputBufferWriteString(buf, ">");
+ return;
+ }
+ if (cur->type == HTML_ENTITY_REF_NODE) {
+ xmlOutputBufferWriteString(buf, "&");
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ xmlOutputBufferWriteString(buf, ";");
+ return;
+ }
+ if (cur->type == HTML_PRESERVE_NODE) {
+ if (cur->content != NULL) {
+ xmlOutputBufferWriteString(buf, (const char *)cur->content);
+ }
+ return;
+ }
+
+ /*
+ * Get specific HTML info for that node.
+ */
+ if (cur->ns == NULL)
+ info = htmlTagLookup(cur->name);
+ else
+ info = NULL;
+
+ xmlOutputBufferWriteString(buf, "<");
+ if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
+ xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
+ xmlOutputBufferWriteString(buf, ":");
+ }
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ if (cur->nsDef)
+ xmlNsListDumpOutput(buf, cur->nsDef);
+ if (cur->properties != NULL)
+ htmlAttrListDumpOutput(buf, doc, cur->properties, encoding);
+
+ if ((info != NULL) && (info->empty)) {
+ xmlOutputBufferWriteString(buf, ">");
+ if ((format) && (!info->isinline) && (cur->next != NULL)) {
+ if ((cur->next->type != HTML_TEXT_NODE) &&
+ (cur->next->type != HTML_ENTITY_REF_NODE) &&
+ (cur->parent != NULL) &&
+ (cur->parent->name != NULL) &&
+ (cur->parent->name[0] != 'p')) /* p, pre, param */
+ xmlOutputBufferWriteString(buf, "\n");
+ }
+ return;
+ }
+ if (((cur->type == XML_ELEMENT_NODE) || (cur->content == NULL)) &&
+ (cur->children == NULL)) {
+ if ((info != NULL) && (info->saveEndTag != 0) &&
+ (xmlStrcmp(BAD_CAST info->name, BAD_CAST "html")) &&
+ (xmlStrcmp(BAD_CAST info->name, BAD_CAST "body"))) {
+ xmlOutputBufferWriteString(buf, ">");
+ } else {
+ xmlOutputBufferWriteString(buf, "></");
+ if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
+ xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
+ xmlOutputBufferWriteString(buf, ":");
+ }
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ xmlOutputBufferWriteString(buf, ">");
+ }
+ if ((format) && (cur->next != NULL) &&
+ (info != NULL) && (!info->isinline)) {
+ if ((cur->next->type != HTML_TEXT_NODE) &&
+ (cur->next->type != HTML_ENTITY_REF_NODE) &&
+ (cur->parent != NULL) &&
+ (cur->parent->name != NULL) &&
+ (cur->parent->name[0] != 'p')) /* p, pre, param */
+ xmlOutputBufferWriteString(buf, "\n");
+ }
+ return;
+ }
+ xmlOutputBufferWriteString(buf, ">");
+ if ((cur->type != XML_ELEMENT_NODE) &&
+ (cur->content != NULL)) {
+ /*
+ * Uses the OutputBuffer property to automatically convert
+ * invalids to charrefs
+ */
+
+ xmlOutputBufferWriteString(buf, (const char *) cur->content);
+ }
+ if (cur->children != NULL) {
+ if ((format) && (info != NULL) && (!info->isinline) &&
+ (cur->children->type != HTML_TEXT_NODE) &&
+ (cur->children->type != HTML_ENTITY_REF_NODE) &&
+ (cur->children != cur->last) &&
+ (cur->name != NULL) &&
+ (cur->name[0] != 'p')) /* p, pre, param */
+ xmlOutputBufferWriteString(buf, "\n");
+ htmlNodeListDumpOutput(buf, doc, cur->children, encoding, format);
+ if ((format) && (info != NULL) && (!info->isinline) &&
+ (cur->last->type != HTML_TEXT_NODE) &&
+ (cur->last->type != HTML_ENTITY_REF_NODE) &&
+ (cur->children != cur->last) &&
+ (cur->name != NULL) &&
+ (cur->name[0] != 'p')) /* p, pre, param */
+ xmlOutputBufferWriteString(buf, "\n");
+ }
+ xmlOutputBufferWriteString(buf, "</");
+ if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
+ xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
+ xmlOutputBufferWriteString(buf, ":");
+ }
+ xmlOutputBufferWriteString(buf, (const char *)cur->name);
+ xmlOutputBufferWriteString(buf, ">");
+ if ((format) && (info != NULL) && (!info->isinline) &&
+ (cur->next != NULL)) {
+ if ((cur->next->type != HTML_TEXT_NODE) &&
+ (cur->next->type != HTML_ENTITY_REF_NODE) &&
+ (cur->parent != NULL) &&
+ (cur->parent->name != NULL) &&
+ (cur->parent->name[0] != 'p')) /* p, pre, param */
+ xmlOutputBufferWriteString(buf, "\n");
+ }
+}
+
+/**
+ * htmlNodeDumpOutput:
+ * @buf: the HTML buffer output
+ * @doc: the document
+ * @cur: the current node
+ * @encoding: the encoding string
+ *
+ * Dump an HTML node, recursive behaviour,children are printed too,
+ * and formatting returns/spaces are added.
+ */
+void
+htmlNodeDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
+ xmlNodePtr cur, const char *encoding) {
+ htmlNodeDumpFormatOutput(buf, doc, cur, encoding, 1);
+}
+
+/**
+ * htmlDocContentDumpFormatOutput:
+ * @buf: the HTML buffer output
+ * @cur: the document
+ * @encoding: the encoding string
+ * @format: should formatting spaces been added
+ *
+ * Dump an HTML document.
+ */
+void
+htmlDocContentDumpFormatOutput(xmlOutputBufferPtr buf, xmlDocPtr cur,
+ const char *encoding, int format) {
+ int type;
+
+ xmlInitParser();
+
+ if ((buf == NULL) || (cur == NULL))
+ return;
+
+ /*
+ * force to output the stuff as HTML, especially for entities
+ */
+ type = cur->type;
+ cur->type = XML_HTML_DOCUMENT_NODE;
+ if (cur->intSubset != NULL) {
+ htmlDtdDumpOutput(buf, cur, NULL);
+ }
+ if (cur->children != NULL) {
+ htmlNodeListDumpOutput(buf, cur, cur->children, encoding, format);
+ }
+ xmlOutputBufferWriteString(buf, "\n");
+ cur->type = (xmlElementType) type;
+}
+
+/**
+ * htmlDocContentDumpOutput:
+ * @buf: the HTML buffer output
+ * @cur: the document
+ * @encoding: the encoding string
+ *
+ * Dump an HTML document. Formating return/spaces are added.
+ */
+void
+htmlDocContentDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr cur,
+ const char *encoding) {
+ htmlDocContentDumpFormatOutput(buf, cur, encoding, 1);
+}
+
+/************************************************************************
+ * *
+ * Saving functions front-ends *
+ * *
+ ************************************************************************/
+
+/**
+ * htmlDocDump:
+ * @f: the FILE*
+ * @cur: the document
+ *
+ * Dump an HTML document to an open FILE.
+ *
+ * returns: the number of byte written or -1 in case of failure.
+ */
+int
+htmlDocDump(FILE *f, xmlDocPtr cur) {
+ xmlOutputBufferPtr buf;
+ xmlCharEncodingHandlerPtr handler = NULL;
+ const char *encoding;
+ int ret;
+
+ xmlInitParser();
+
+ if ((cur == NULL) || (f == NULL)) {
+ return(-1);
+ }
+
+ encoding = (const char *) htmlGetMetaEncoding(cur);
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+
+ enc = xmlParseCharEncoding(encoding);
+ if (enc != XML_CHAR_ENCODING_UTF8) {
+ handler = xmlFindCharEncodingHandler(encoding);
+ if (handler == NULL)
+ htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
+ }
+ } else {
+ /*
+ * Fallback to HTML or ASCII when the encoding is unspecified
+ */
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("HTML");
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("ascii");
+ }
+
+ buf = xmlOutputBufferCreateFile(f, handler);
+ if (buf == NULL) return(-1);
+ htmlDocContentDumpOutput(buf, cur, NULL);
+
+ ret = xmlOutputBufferClose(buf);
+ return(ret);
+}
+
+/**
+ * htmlSaveFile:
+ * @filename: the filename (or URL)
+ * @cur: the document
+ *
+ * Dump an HTML document to a file. If @filename is "-" the stdout file is
+ * used.
+ * returns: the number of byte written or -1 in case of failure.
+ */
+int
+htmlSaveFile(const char *filename, xmlDocPtr cur) {
+ xmlOutputBufferPtr buf;
+ xmlCharEncodingHandlerPtr handler = NULL;
+ const char *encoding;
+ int ret;
+
+ if ((cur == NULL) || (filename == NULL))
+ return(-1);
+
+ xmlInitParser();
+
+ encoding = (const char *) htmlGetMetaEncoding(cur);
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+
+ enc = xmlParseCharEncoding(encoding);
+ if (enc != XML_CHAR_ENCODING_UTF8) {
+ handler = xmlFindCharEncodingHandler(encoding);
+ if (handler == NULL)
+ htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
+ }
+ } else {
+ /*
+ * Fallback to HTML or ASCII when the encoding is unspecified
+ */
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("HTML");
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("ascii");
+ }
+
+ /*
+ * save the content to a temp buffer.
+ */
+ buf = xmlOutputBufferCreateFilename(filename, handler, cur->compression);
+ if (buf == NULL) return(0);
+
+ htmlDocContentDumpOutput(buf, cur, NULL);
+
+ ret = xmlOutputBufferClose(buf);
+ return(ret);
+}
+
+/**
+ * htmlSaveFileFormat:
+ * @filename: the filename
+ * @cur: the document
+ * @format: should formatting spaces been added
+ * @encoding: the document encoding
+ *
+ * Dump an HTML document to a file using a given encoding.
+ *
+ * returns: the number of byte written or -1 in case of failure.
+ */
+int
+htmlSaveFileFormat(const char *filename, xmlDocPtr cur,
+ const char *encoding, int format) {
+ xmlOutputBufferPtr buf;
+ xmlCharEncodingHandlerPtr handler = NULL;
+ int ret;
+
+ if ((cur == NULL) || (filename == NULL))
+ return(-1);
+
+ xmlInitParser();
+
+ if (encoding != NULL) {
+ xmlCharEncoding enc;
+
+ enc = xmlParseCharEncoding(encoding);
+ if (enc != XML_CHAR_ENCODING_UTF8) {
+ handler = xmlFindCharEncodingHandler(encoding);
+ if (handler == NULL)
+ htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
+ }
+ htmlSetMetaEncoding(cur, (const xmlChar *) encoding);
+ } else {
+ htmlSetMetaEncoding(cur, (const xmlChar *) "UTF-8");
+
+ /*
+ * Fallback to HTML or ASCII when the encoding is unspecified
+ */
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("HTML");
+ if (handler == NULL)
+ handler = xmlFindCharEncodingHandler("ascii");
+ }
+
+ /*
+ * save the content to a temp buffer.
+ */
+ buf = xmlOutputBufferCreateFilename(filename, handler, 0);
+ if (buf == NULL) return(0);
+
+ htmlDocContentDumpFormatOutput(buf, cur, encoding, format);
+
+ ret = xmlOutputBufferClose(buf);
+ return(ret);
+}
+
+/**
+ * htmlSaveFileEnc:
+ * @filename: the filename
+ * @cur: the document
+ * @encoding: the document encoding
+ *
+ * Dump an HTML document to a file using a given encoding
+ * and formatting returns/spaces are added.
+ *
+ * returns: the number of byte written or -1 in case of failure.
+ */
+int
+htmlSaveFileEnc(const char *filename, xmlDocPtr cur, const char *encoding) {
+ return(htmlSaveFileFormat(filename, cur, encoding, 1));
+}
+
+#endif /* LIBXML_OUTPUT_ENABLED */
+
+#define bottom_HTMLtree
+#include "elfgcchack.h"
+#endif /* LIBXML_HTML_ENABLED */
diff --git a/METADATA b/METADATA
index f27a977..279a869 100644
--- a/METADATA
+++ b/METADATA
@@ -11,9 +11,9 @@
}
url {
type: ARCHIVE
- value: "ftp://xmlsoft.org/libxml2/libxml2-2.9.8.tar.gz"
+ value: "ftp://xmlsoft.org/libxml2/libxml2-2.9.9.tar.gz"
}
- version: "2.9.8"
- last_upgrade_date { year: 2018 month: 3 day: 5 }
+ version: "2.9.9"
+ last_upgrade_date { year: 2019 month: 1 day: 3 }
license_type: NOTICE
}
diff --git a/SAX2.c b/SAX2.c
index 0f261b7..7642501 100644
--- a/SAX2.c
+++ b/SAX2.c
@@ -1665,7 +1665,11 @@
#ifdef DEBUG_SAX_TREE
xmlGenericError(xmlGenericErrorContext, "pushing(%s)\n", name);
#endif
- nodePush(ctxt, ret);
+ if (nodePush(ctxt, ret) < 0) {
+ xmlUnlinkNode(ret);
+ xmlFreeNode(ret);
+ return;
+ }
/*
* Link the child element
@@ -2257,6 +2261,7 @@
ctxt->freeElems = ret->next;
ctxt->freeElemsNr--;
memset(ret, 0, sizeof(xmlNode));
+ ret->doc = ctxt->myDoc;
ret->type = XML_ELEMENT_NODE;
if (ctxt->dictNames)
@@ -2336,7 +2341,11 @@
/*
* We are parsing a new node.
*/
- nodePush(ctxt, ret);
+ if (nodePush(ctxt, ret) < 0) {
+ xmlUnlinkNode(ret);
+ xmlFreeNode(ret);
+ return;
+ }
/*
* Link the child element
diff --git a/autogen.sh b/autogen.sh
index 8b47ea5..d03a40c 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -67,6 +67,15 @@
# Replaced by autoreconf below
autoreconf -if -Wall
+if ! grep -q pkg.m4 aclocal.m4; then
+ cat <<EOF
+
+Couldn't find pkg.m4 from pkg-config. Install the appropriate package for
+your distribution or set ACLOCAL_PATH to the directory containing pkg.m4.
+EOF
+ exit 1
+fi
+
cd $THEDIR
if test x$OBJ_DIR != x; then
@@ -76,6 +85,11 @@
if test -z "$NOCONFIGURE"; then
$srcdir/configure $EXTRA_ARGS "$@"
- echo
- echo "Now type 'make' to compile libxml2."
+ if test "$?" -ne 0; then
+ echo
+ echo "Configure script failed, check config.log for more info."
+ else
+ echo
+ echo "Now type 'make' to compile libxml2."
+ fi
fi
diff --git a/c14n.c b/c14n.c
index c04ce66..d80ae8b 100644
--- a/c14n.c
+++ b/c14n.c
@@ -1798,15 +1798,6 @@
}
/*
- * Validate the XML document encoding value, if provided.
- */
- if (doc->charset != XML_CHAR_ENCODING_UTF8) {
- xmlC14NErr(ctx, (xmlNodePtr) doc, XML_C14N_REQUIRES_UTF8,
- "xmlC14NNewCtx: source document not in UTF8\n");
- return (NULL);
- }
-
- /*
* Allocate a new xmlC14NCtxPtr and fill the fields.
*/
ctx = (xmlC14NCtxPtr) xmlMalloc(sizeof(xmlC14NCtx));
diff --git a/compile b/compile
new file mode 100755
index 0000000..de0005d
--- /dev/null
+++ b/compile
@@ -0,0 +1,348 @@
+#!/bin/sh
+# Wrapper for compilers which do not understand '-c -o'.
+
+scriptversion=2016-01-11.22; # UTC
+
+# Copyright (C) 1999-2017 Free Software Foundation, Inc.
+# Written by Tom Tromey <tromey@cygnus.com>.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2, or (at your option)
+# any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+# As a special exception to the GNU General Public License, if you
+# distribute this file as part of a program that contains a
+# configuration script generated by Autoconf, you may include it under
+# the same distribution terms that you use for the rest of that program.
+
+# This file is maintained in Automake, please report
+# bugs to <bug-automake@gnu.org> or send patches to
+# <automake-patches@gnu.org>.
+
+nl='
+'
+
+# We need space, tab and new line, in precisely that order. Quoting is
+# there to prevent tools from complaining about whitespace usage.
+IFS=" "" $nl"
+
+file_conv=
+
+# func_file_conv build_file lazy
+# Convert a $build file to $host form and store it in $file
+# Currently only supports Windows hosts. If the determined conversion
+# type is listed in (the comma separated) LAZY, no conversion will
+# take place.
+func_file_conv ()
+{
+ file=$1
+ case $file in
+ / | /[!/]*) # absolute file, and not a UNC file
+ if test -z "$file_conv"; then
+ # lazily determine how to convert abs files
+ case `uname -s` in
+ MINGW*)
+ file_conv=mingw
+ ;;
+ CYGWIN*)
+ file_conv=cygwin
+ ;;
+ *)
+ file_conv=wine
+ ;;
+ esac
+ fi
+ case $file_conv/,$2, in
+ *,$file_conv,*)
+ ;;
+ mingw/*)
+ file=`cmd //C echo "$file " | sed -e 's/"\(.*\) " *$/\1/'`
+ ;;
+ cygwin/*)
+ file=`cygpath -m "$file" || echo "$file"`
+ ;;
+ wine/*)
+ file=`winepath -w "$file" || echo "$file"`
+ ;;
+ esac
+ ;;
+ esac
+}
+
+# func_cl_dashL linkdir
+# Make cl look for libraries in LINKDIR
+func_cl_dashL ()
+{
+ func_file_conv "$1"
+ if test -z "$lib_path"; then
+ lib_path=$file
+ else
+ lib_path="$lib_path;$file"
+ fi
+ linker_opts="$linker_opts -LIBPATH:$file"
+}
+
+# func_cl_dashl library
+# Do a library search-path lookup for cl
+func_cl_dashl ()
+{
+ lib=$1
+ found=no
+ save_IFS=$IFS
+ IFS=';'
+ for dir in $lib_path $LIB
+ do
+ IFS=$save_IFS
+ if $shared && test -f "$dir/$lib.dll.lib"; then
+ found=yes
+ lib=$dir/$lib.dll.lib
+ break
+ fi
+ if test -f "$dir/$lib.lib"; then
+ found=yes
+ lib=$dir/$lib.lib
+ break
+ fi
+ if test -f "$dir/lib$lib.a"; then
+ found=yes
+ lib=$dir/lib$lib.a
+ break
+ fi
+ done
+ IFS=$save_IFS
+
+ if test "$found" != yes; then
+ lib=$lib.lib
+ fi
+}
+
+# func_cl_wrapper cl arg...
+# Adjust compile command to suit cl
+func_cl_wrapper ()
+{
+ # Assume a capable shell
+ lib_path=
+ shared=:
+ linker_opts=
+ for arg
+ do
+ if test -n "$eat"; then
+ eat=
+ else
+ case $1 in
+ -o)
+ # configure might choose to run compile as 'compile cc -o foo foo.c'.
+ eat=1
+ case $2 in
+ *.o | *.[oO][bB][jJ])
+ func_file_conv "$2"
+ set x "$@" -Fo"$file"
+ shift
+ ;;
+ *)
+ func_file_conv "$2"
+ set x "$@" -Fe"$file"
+ shift
+ ;;
+ esac
+ ;;
+ -I)
+ eat=1
+ func_file_conv "$2" mingw
+ set x "$@" -I"$file"
+ shift
+ ;;
+ -I*)
+ func_file_conv "${1#-I}" mingw
+ set x "$@" -I"$file"
+ shift
+ ;;
+ -l)
+ eat=1
+ func_cl_dashl "$2"
+ set x "$@" "$lib"
+ shift
+ ;;
+ -l*)
+ func_cl_dashl "${1#-l}"
+ set x "$@" "$lib"
+ shift
+ ;;
+ -L)
+ eat=1
+ func_cl_dashL "$2"
+ ;;
+ -L*)
+ func_cl_dashL "${1#-L}"
+ ;;
+ -static)
+ shared=false
+ ;;
+ -Wl,*)
+ arg=${1#-Wl,}
+ save_ifs="$IFS"; IFS=','
+ for flag in $arg; do
+ IFS="$save_ifs"
+ linker_opts="$linker_opts $flag"
+ done
+ IFS="$save_ifs"
+ ;;
+ -Xlinker)
+ eat=1
+ linker_opts="$linker_opts $2"
+ ;;
+ -*)
+ set x "$@" "$1"
+ shift
+ ;;
+ *.cc | *.CC | *.cxx | *.CXX | *.[cC]++)
+ func_file_conv "$1"
+ set x "$@" -Tp"$file"
+ shift
+ ;;
+ *.c | *.cpp | *.CPP | *.lib | *.LIB | *.Lib | *.OBJ | *.obj | *.[oO])
+ func_file_conv "$1" mingw
+ set x "$@" "$file"
+ shift
+ ;;
+ *)
+ set x "$@" "$1"
+ shift
+ ;;
+ esac
+ fi
+ shift
+ done
+ if test -n "$linker_opts"; then
+ linker_opts="-link$linker_opts"
+ fi
+ exec "$@" $linker_opts
+ exit 1
+}
+
+eat=
+
+case $1 in
+ '')
+ echo "$0: No command. Try '$0 --help' for more information." 1>&2
+ exit 1;
+ ;;
+ -h | --h*)
+ cat <<\EOF
+Usage: compile [--help] [--version] PROGRAM [ARGS]
+
+Wrapper for compilers which do not understand '-c -o'.
+Remove '-o dest.o' from ARGS, run PROGRAM with the remaining
+arguments, and rename the output as expected.
+
+If you are trying to build a whole package this is not the
+right script to run: please start by reading the file 'INSTALL'.
+
+Report bugs to <bug-automake@gnu.org>.
+EOF
+ exit $?
+ ;;
+ -v | --v*)
+ echo "compile $scriptversion"
+ exit $?
+ ;;
+ cl | *[/\\]cl | cl.exe | *[/\\]cl.exe | \
+ icl | *[/\\]icl | icl.exe | *[/\\]icl.exe )
+ func_cl_wrapper "$@" # Doesn't return...
+ ;;
+esac
+
+ofile=
+cfile=
+
+for arg
+do
+ if test -n "$eat"; then
+ eat=
+ else
+ case $1 in
+ -o)
+ # configure might choose to run compile as 'compile cc -o foo foo.c'.
+ # So we strip '-o arg' only if arg is an object.
+ eat=1
+ case $2 in
+ *.o | *.obj)
+ ofile=$2
+ ;;
+ *)
+ set x "$@" -o "$2"
+ shift
+ ;;
+ esac
+ ;;
+ *.c)
+ cfile=$1
+ set x "$@" "$1"
+ shift
+ ;;
+ *)
+ set x "$@" "$1"
+ shift
+ ;;
+ esac
+ fi
+ shift
+done
+
+if test -z "$ofile" || test -z "$cfile"; then
+ # If no '-o' option was seen then we might have been invoked from a
+ # pattern rule where we don't need one. That is ok -- this is a
+ # normal compilation that the losing compiler can handle. If no
+ # '.c' file was seen then we are probably linking. That is also
+ # ok.
+ exec "$@"
+fi
+
+# Name of file we expect compiler to create.
+cofile=`echo "$cfile" | sed 's|^.*[\\/]||; s|^[a-zA-Z]:||; s/\.c$/.o/'`
+
+# Create the lock directory.
+# Note: use '[/\\:.-]' here to ensure that we don't use the same name
+# that we are using for the .o file. Also, base the name on the expected
+# object file name, since that is what matters with a parallel build.
+lockdir=`echo "$cofile" | sed -e 's|[/\\:.-]|_|g'`.d
+while true; do
+ if mkdir "$lockdir" >/dev/null 2>&1; then
+ break
+ fi
+ sleep 1
+done
+# FIXME: race condition here if user kills between mkdir and trap.
+trap "rmdir '$lockdir'; exit 1" 1 2 15
+
+# Run the compile.
+"$@"
+ret=$?
+
+if test -f "$cofile"; then
+ test "$cofile" = "$ofile" || mv "$cofile" "$ofile"
+elif test -f "${cofile}bj"; then
+ test "${cofile}bj" = "$ofile" || mv "${cofile}bj" "$ofile"
+fi
+
+rmdir "$lockdir"
+exit $ret
+
+# Local Variables:
+# mode: shell-script
+# sh-indentation: 2
+# eval: (add-hook 'write-file-hooks 'time-stamp)
+# time-stamp-start: "scriptversion="
+# time-stamp-format: "%:y-%02m-%02d.%02H"
+# time-stamp-time-zone: "UTC0"
+# time-stamp-end: "; # UTC"
+# End:
diff --git a/configure.ac b/configure.ac
index 370e2de..ca911f3 100644
--- a/configure.ac
+++ b/configure.ac
@@ -9,7 +9,7 @@
LIBXML_MAJOR_VERSION=2
LIBXML_MINOR_VERSION=9
-LIBXML_MICRO_VERSION=8
+LIBXML_MICRO_VERSION=9
LIBXML_MICRO_VERSION_SUFFIX=
LIBXML_VERSION=$LIBXML_MAJOR_VERSION.$LIBXML_MINOR_VERSION.$LIBXML_MICRO_VERSION$LIBXML_MICRO_VERSION_SUFFIX
LIBXML_VERSION_INFO=`expr $LIBXML_MAJOR_VERSION + $LIBXML_MINOR_VERSION`:$LIBXML_MICRO_VERSION:$LIBXML_MINOR_VERSION
diff --git a/doc/libxml2.xsa b/doc/libxml2.xsa
index 33f3848..7f4c048 100644
--- a/doc/libxml2.xsa
+++ b/doc/libxml2.xsa
@@ -8,45 +8,51 @@
</vendor>
<product id="libxml2">
<name>libxml2</name>
- <version>2.9.7</version>
- <last-release> Nov 02 2017</last-release>
+ <version>v2.9.8</version>
+ <last-release> Mar 05 2018</last-release>
<info-url>http://xmlsoft.org/</info-url>
- <changes> - Documentation:
- xmlcatalog: refresh man page wrt. quering system catalog easily (Jan Pokorný)
-
- - Portability:
- Fix deprecated Travis compiler flag (Nick Wellnhofer),
- Add declaration for DllMain (J. Peter Mugaas),
- Fix preprocessor conditional in threads.h (J. Peter Mugaas),
- Fix pointer comparison warnings on 64-bit Windows (J. Peter Mugaas),
- Fix macro redefinition warning (J. Peter Mugaas),
- Default to native threads on MinGW-w64 (Nick Wellnhofer),
- Simplify Windows IO functions (Nick Wellnhofer),
- Fix runtest on Windows (Nick Wellnhofer),
- socklen_t is always int on Windows (Nick Wellnhofer),
- Don't redefine socket error codes on Windows (Nick Wellnhofer),
- Fix pointer/int cast warnings on 64-bit Windows (Nick Wellnhofer),
- Fix Windows compiler warnings in xmlCanonicPath (Nick Wellnhofer)
+ <changes> - Portability:
+ python: remove single use of _PyVerify_fd (Patrick Welche),
+ Build more test executables on Windows/MSVC (Nick Wellnhofer),
+ Stop including ansidecl.h (Nick Wellnhofer),
+ Fix libz and liblzma detection (Nick Wellnhofer),
+ Revert "Compile testapi with -Wno-unused-function" (Nick Wellnhofer)
- Bug Fixes:
- xmlcatalog: restore ability to query system catalog easily (Jan Pokorný),
- Fix comparison of nodesets to strings (Nick Wellnhofer)
+ Fix xmlParserEntityCheck (Nick Wellnhofer),
+ Halt parser in case of encoding error (Nick Wellnhofer),
+ Clear entity content in case of errors (Nick Wellnhofer),
+ Change calls to xmlCharEncInput to set flush false when not final call. Having flush incorrectly set to true causes errors for ICU. (Joel Hockey),
+ Fix buffer over-read in xmlParseNCNameComplex (Nick Wellnhofer),
+ Fix ICU library filenames on Windows/MSVC (Nick Wellnhofer),
+ Fix xmlXPathIsNaN broken by recent commit (Nick Wellnhofer),
+ Fix -Wenum-compare warnings (Nick Wellnhofer),
+ Fix callback signature in testapi.c (Nick Wellnhofer),
+ Fix unused parameter warning without ICU (Nick Wellnhofer),
+ Fix IO callback signatures (Nick Wellnhofer),
+ Fix misc callback signatures (Nick Wellnhofer),
+ Fix list callback signatures (Nick Wellnhofer),
+ Fix hash callback signatures (Nick Wellnhofer),
+ Refactor name and type signature for xmlNop (Vlad Tsyrklevich),
+ Fixed ICU to set flush correctly and provide pivot buffer. (Joel Hockey),
+ Skip EBCDIC tests if EBCDIC isn't supported (Nick Wellnhofer)
- Improvements:
- Add Makefile rules to rebuild HTML man pages (Nick Wellnhofer),
- Fix mixed decls and code in timsort.h (Nick Wellnhofer),
- Rework handling of return values in thread tests (Nick Wellnhofer),
- Fix unused variable warnings in testrecurse (Nick Wellnhofer),
- Fix -Wimplicit-fallthrough warnings (J. Peter Mugaas),
- Upgrade timsort.h to latest revision (Nick Wellnhofer),
- Increase warning level to /W3 under MSVC (Nick Wellnhofer),
- Fix a couple of warnings in dict.c and threads.c (Nick Wellnhofer),
- Update .gitignore for Windows (Nick Wellnhofer),
- Fix unused variable warnings in nanohttp.c (Nick Wellnhofer),
- Fix the Windows header mess (Nick Wellnhofer),
- Don't include winsock2.h in xmllint.c (Nick Wellnhofer),
- Remove generated file python/setup.py from version control (Nick Wellnhofer),
- Use __linux__ macro in generated code (Nick Wellnhofer)
+ Disable pointer-overflow UBSan checks under Travis (Nick Wellnhofer),
+ Improve handling of context input_id (Daniel Veillard),
+ Add resource file to Windows DLL (ccpaging),
+ Run Travis tests with -Werror (Nick Wellnhofer),
+ Build with "-Wall -Wextra" (Nick Wellnhofer),
+ Fix -Wtautological-pointer-compare warnings (Nick Wellnhofer),
+ Remove unused AC_CHECKs (Nick Wellnhofer),
+ Update information about contributing (Nick Wellnhofer),
+ Fix -Wmisleading-indentation warnings (Nick Wellnhofer),
+ Don't touch CFLAGS in configure.ac (Nick Wellnhofer),
+ Ignore function pointer cast warnings (Nick Wellnhofer),
+ Simplify XPath NaN, inf and -0 handling (Nick Wellnhofer),
+ Introduce xmlPosixStrdup and update xmlMemStrdup (Nick Wellnhofer),
+ Add test for ICU flush and pivot buffer (Nick Wellnhofer),
+ Compile testapi with -Wno-unused-function (Nick Wellnhofer)
</changes>
diff --git a/doc/news.html b/doc/news.html
index 65839ad..db07ab6 100644
--- a/doc/news.html
+++ b/doc/news.html
@@ -8,7 +8,72 @@
H3 {font-family: Verdana,Arial,Helvetica}
A:link, A:visited, A:active { text-decoration: underline }
</style><title>Releases</title></head><body bgcolor="#8b7765" text="#000000" link="#a06060" vlink="#000000"><table border="0" width="100%" cellpadding="5" cellspacing="0" align="center"><tr><td width="120"><a href="http://swpat.ffii.org/"><img src="epatents.png" alt="Action against software patents" /></a></td><td width="180"><a href="http://www.gnome.org/"><img src="gnome2.png" alt="Gnome2 Logo" /></a><a href="http://www.w3.org/Status"><img src="w3c.png" alt="W3C Logo" /></a><a href="http://www.redhat.com/"><img src="redhat.gif" alt="Red Hat Logo" /></a><div align="left"><a href="http://xmlsoft.org/"><img src="Libxml2-Logo-180x168.gif" alt="Made with Libxml2 Logo" /></a></div></td><td><table border="0" width="90%" cellpadding="2" cellspacing="0" align="center" bgcolor="#000000"><tr><td><table width="100%" border="0" cellspacing="1" cellpadding="3" bgcolor="#fffacd"><tr><td align="center"><h1>The XML C parser and toolkit of Gnome</h1><h2>Releases</h2></td></tr></table></td></tr></table></td></tr></table><table border="0" cellpadding="4" cellspacing="0" width="100%" align="center"><tr><td bgcolor="#8b7765"><table border="0" cellspacing="0" cellpadding="2" width="100%"><tr><td valign="top" width="200" bgcolor="#8b7765"><table border="0" cellspacing="0" cellpadding="1" width="100%" bgcolor="#000000"><tr><td><table width="100%" border="0" cellspacing="1" cellpadding="3"><tr><td colspan="1" bgcolor="#eecfa1" align="center"><center><b>Main Menu</b></center></td></tr><tr><td bgcolor="#fffacd"><form action="search.php" enctype="application/x-www-form-urlencoded" method="get"><input name="query" type="text" size="20" value="" /><input name="submit" type="submit" value="Search ..." /></form><ul><li><a href="index.html">Home</a></li><li><a href="html/index.html">Reference Manual</a></li><li><a href="intro.html">Introduction</a></li><li><a href="FAQ.html">FAQ</a></li><li><a href="docs.html" style="font-weight:bold">Developer Menu</a></li><li><a href="bugs.html">Reporting bugs and getting help</a></li><li><a href="help.html">How to help</a></li><li><a href="downloads.html">Downloads</a></li><li><a href="news.html">Releases</a></li><li><a href="XMLinfo.html">XML</a></li><li><a href="XSLT.html">XSLT</a></li><li><a href="xmldtd.html">Validation & DTDs</a></li><li><a href="encoding.html">Encodings support</a></li><li><a href="catalog.html">Catalog support</a></li><li><a href="namespaces.html">Namespaces</a></li><li><a href="contribs.html">Contributions</a></li><li><a href="examples/index.html" style="font-weight:bold">Code Examples</a></li><li><a href="html/index.html" style="font-weight:bold">API Menu</a></li><li><a href="guidelines.html">XML Guidelines</a></li><li><a href="ChangeLog.html">Recent Changes</a></li></ul></td></tr></table><table width="100%" border="0" cellspacing="1" cellpadding="3"><tr><td colspan="1" bgcolor="#eecfa1" align="center"><center><b>Related links</b></center></td></tr><tr><td bgcolor="#fffacd"><ul><li><a href="http://mail.gnome.org/archives/xml/">Mail archive</a></li><li><a href="http://xmlsoft.org/XSLT/">XSLT libxslt</a></li><li><a href="http://phd.cs.unibo.it/gdome2/">DOM gdome2</a></li><li><a href="http://www.aleksey.com/xmlsec/">XML-DSig xmlsec</a></li><li><a href="ftp://xmlsoft.org/">FTP</a></li><li><a href="http://www.zlatkovic.com/projects/libxml/">Windows binaries</a></li><li><a href="http://opencsw.org/packages/libxml2">Solaris binaries</a></li><li><a href="http://www.explain.com.au/oss/libxml2xslt.html">MacOsX binaries</a></li><li><a href="http://lxml.de/">lxml Python bindings</a></li><li><a href="http://cpan.uwinnipeg.ca/dist/XML-LibXML">Perl bindings</a></li><li><a href="http://libxmlplusplus.sourceforge.net/">C++ bindings</a></li><li><a href="http://www.zend.com/php5/articles/php5-xmlphp.php#Heading4">PHP bindings</a></li><li><a href="http://sourceforge.net/projects/libxml2-pas/">Pascal bindings</a></li><li><a href="http://libxml.rubyforge.org/">Ruby bindings</a></li><li><a href="http://tclxml.sourceforge.net/">Tcl bindings</a></li><li><a href="http://bugzilla.gnome.org/buglist.cgi?product=libxml2">Bug Tracker</a></li></ul></td></tr></table></td></tr></table></td><td valign="top" bgcolor="#8b7765"><table border="0" cellspacing="0" cellpadding="1" width="100%"><tr><td><table border="0" cellspacing="0" cellpadding="1" width="100%" bgcolor="#000000"><tr><td><table border="0" cellpadding="3" cellspacing="1" width="100%"><tr><td bgcolor="#fffacd"><p>The <a href="ChangeLog.html">change log</a> describes the recents commits
-to the <a href="http://git.gnome.org/browse/libxml2/">GIT</a> code base.</p><p>Here is the list of public releases:</p><h3>v2.9.8: Mar 05 2018</h3><ul>
+to the <a href="http://git.gnome.org/browse/libxml2/">GIT</a> code base.</p><p>Here is the list of public releases:</p><h3>v2.9.9: Jan 03 2019</h3><ul>
+ <li>Security:<br />
+ CVE-2018-9251 CVE-2018-14567 Fix infinite loop in LZMA decompression (Nick Wellnhofer),<br />
+ CVE-2018-14404 Fix nullptr deref with XPath logic ops (Nick Wellnhofer),<br />
+ </li>
+
+ <li>Documentation:<br />
+ reader: Fix documentation comment (Mohammed Sadiq)<br />
+ </li>
+
+ <li>Portability:<br />
+ Fix MSVC build with lzma (Nick Wellnhofer),<br />
+ Variables need 'extern' in static lib on Cygwin (Michael Haubenwallner),<br />
+ Really declare dllexport/dllimport for Cygwin (Michael Haubenwallner),<br />
+ Merge branch 'patch-2' into 'master' (Nick Wellnhofer),<br />
+ Change dir to $THEDIR after ACLOCAL_PATH check autoreconf creates aclocal.m4 in $srcdir (Vitaly Buka),<br />
+ Improve error message if pkg.m4 couldn't be found (Nick Wellnhofer),<br />
+ NaN and Inf fixes for pre-C99 compilers (Nick Wellnhofer)<br />
+ </li>
+
+ <li>Bug Fixes:<br />
+ Revert "Support xmlTextReaderNextSibling w/o preparsed doc" (Nick Wellnhofer),<br />
+ Fix building relative URIs (Thomas Holder),<br />
+ Problem with data in interleave in RelaxNG validation (Nikolai Weibull),<br />
+ Fix memory leak in xmlSwitchInputEncodingInt error path (Nick Wellnhofer),<br />
+ Set doc on element obtained from freeElems (Nick Wellnhofer),<br />
+ Fix HTML serialization with UTF-8 encoding (Nick Wellnhofer),<br />
+ Use actual doc in xmlTextReaderRead*Xml (Nick Wellnhofer),<br />
+ Unlink node before freeing it in xmlSAX2StartElement (Nick Wellnhofer),<br />
+ Check return value of nodePush in xmlSAX2StartElement (Nick Wellnhofer),<br />
+ Free input buffer in xmlHaltParser (Nick Wellnhofer),<br />
+ Reset HTML parser input pointers on encoding failure (Nick Wellnhofer),<br />
+ Don't run icu_parse_test if EUC-JP is unsupported (Nick Wellnhofer),<br />
+ Fix xmlSchemaValidCtxtPtr reuse memory leak (Greg Hildstrom),<br />
+ Fix xmlTextReaderNext with preparsed document (Felix Bünemann),<br />
+ Remove stray character from comment (Nick Wellnhofer),<br />
+ Remove a misleading line from xmlCharEncOutput (Andrey Bienkowski),<br />
+ HTML noscript should not close p (Daniel Veillard),<br />
+ Don't change context node in xmlXPathRoot (Nick Wellnhofer),<br />
+ Stop using XPATH_OP_RESET (Nick Wellnhofer),<br />
+ Revert "Change calls to xmlCharEncInput to set flush false" (Nick Wellnhofer)<br />
+ </li>
+
+ <li>Improvements:<br />
+ Fix "Problem with data in interleave in RelaxNG validation" (Nikolai Weibull),<br />
+ cleanup: remove some unreachable code (Thomas Holder),<br />
+ add --relative to testURI (Thomas Holder),<br />
+ Remove redefined starts and defines inside include elements (Nikolai Weibull),<br />
+ Allow choice within choice in nameClass in RELAX NG (Nikolai Weibull),<br />
+ Look inside divs for starts and defines inside include (Nikolai Weibull),<br />
+ Add compile and libxml2-config.cmake to .gitignore (Nikolai Weibull),<br />
+ Stop using doc->charset outside parser code (Nick Wellnhofer),<br />
+ Add newlines to 'xmllint --xpath' output (Nick Wellnhofer),<br />
+ Don't include SAX.h from globals.h (Nick Wellnhofer),<br />
+ Support xmlTextReaderNextSibling w/o preparsed doc (Felix Bünemann),<br />
+ Don't instruct user to run make when autogen.sh failed (林博仁(Buo-ren Lin)),<br />
+ Run Travis ASan tests with "sudo: required" (Nick Wellnhofer),<br />
+ Improve restoring of context size and position (Nick Wellnhofer),<br />
+ Simplify and harden nodeset filtering (Nick Wellnhofer),<br />
+ Avoid unnecessary backups of the context node (Nick Wellnhofer),<br />
+ Fix inconsistency in xmlXPathIsInf (Nick Wellnhofer)<br />
+ </li>
+
+ <li>Cleanups:<br />
+ </li>
+</ul><h3>v2.9.8: Mar 05 2018</h3><ul>
<li>Portability:<br />
python: remove single use of _PyVerify_fd (Patrick Welche),<br />
Build more test executables on Windows/MSVC (Nick Wellnhofer),<br />
diff --git a/doc/xml.html b/doc/xml.html
index 4059fb7..53388a2 100644
--- a/doc/xml.html
+++ b/doc/xml.html
@@ -709,6 +709,73 @@
<p>Here is the list of public releases:</p>
+<h3>v2.9.9: Jan 03 2019</h3>
+<ul>
+ <li>Security:<br/>
+ CVE-2018-9251 CVE-2018-14567 Fix infinite loop in LZMA decompression (Nick Wellnhofer),<br/>
+ CVE-2018-14404 Fix nullptr deref with XPath logic ops (Nick Wellnhofer),<br/>
+ </li>
+
+ <li>Documentation:<br/>
+ reader: Fix documentation comment (Mohammed Sadiq)<br/>
+ </li>
+
+ <li>Portability:<br/>
+ Fix MSVC build with lzma (Nick Wellnhofer),<br/>
+ Variables need 'extern' in static lib on Cygwin (Michael Haubenwallner),<br/>
+ Really declare dllexport/dllimport for Cygwin (Michael Haubenwallner),<br/>
+ Merge branch 'patch-2' into 'master' (Nick Wellnhofer),<br/>
+ Change dir to $THEDIR after ACLOCAL_PATH check autoreconf creates aclocal.m4 in $srcdir (Vitaly Buka),<br/>
+ Improve error message if pkg.m4 couldn't be found (Nick Wellnhofer),<br/>
+ NaN and Inf fixes for pre-C99 compilers (Nick Wellnhofer)<br/>
+ </li>
+
+ <li>Bug Fixes:<br/>
+ Revert "Support xmlTextReaderNextSibling w/o preparsed doc" (Nick Wellnhofer),<br/>
+ Fix building relative URIs (Thomas Holder),<br/>
+ Problem with data in interleave in RelaxNG validation (Nikolai Weibull),<br/>
+ Fix memory leak in xmlSwitchInputEncodingInt error path (Nick Wellnhofer),<br/>
+ Set doc on element obtained from freeElems (Nick Wellnhofer),<br/>
+ Fix HTML serialization with UTF-8 encoding (Nick Wellnhofer),<br/>
+ Use actual doc in xmlTextReaderRead*Xml (Nick Wellnhofer),<br/>
+ Unlink node before freeing it in xmlSAX2StartElement (Nick Wellnhofer),<br/>
+ Check return value of nodePush in xmlSAX2StartElement (Nick Wellnhofer),<br/>
+ Free input buffer in xmlHaltParser (Nick Wellnhofer),<br/>
+ Reset HTML parser input pointers on encoding failure (Nick Wellnhofer),<br/>
+ Don't run icu_parse_test if EUC-JP is unsupported (Nick Wellnhofer),<br/>
+ Fix xmlSchemaValidCtxtPtr reuse memory leak (Greg Hildstrom),<br/>
+ Fix xmlTextReaderNext with preparsed document (Felix Bünemann),<br/>
+ Remove stray character from comment (Nick Wellnhofer),<br/>
+ Remove a misleading line from xmlCharEncOutput (Andrey Bienkowski),<br/>
+ HTML noscript should not close p (Daniel Veillard),<br/>
+ Don't change context node in xmlXPathRoot (Nick Wellnhofer),<br/>
+ Stop using XPATH_OP_RESET (Nick Wellnhofer),<br/>
+ Revert "Change calls to xmlCharEncInput to set flush false" (Nick Wellnhofer)<br/>
+ </li>
+
+ <li>Improvements:<br/>
+ Fix "Problem with data in interleave in RelaxNG validation" (Nikolai Weibull),<br/>
+ cleanup: remove some unreachable code (Thomas Holder),<br/>
+ add --relative to testURI (Thomas Holder),<br/>
+ Remove redefined starts and defines inside include elements (Nikolai Weibull),<br/>
+ Allow choice within choice in nameClass in RELAX NG (Nikolai Weibull),<br/>
+ Look inside divs for starts and defines inside include (Nikolai Weibull),<br/>
+ Add compile and libxml2-config.cmake to .gitignore (Nikolai Weibull),<br/>
+ Stop using doc->charset outside parser code (Nick Wellnhofer),<br/>
+ Add newlines to 'xmllint --xpath' output (Nick Wellnhofer),<br/>
+ Don't include SAX.h from globals.h (Nick Wellnhofer),<br/>
+ Support xmlTextReaderNextSibling w/o preparsed doc (Felix Bünemann),<br/>
+ Don't instruct user to run make when autogen.sh failed (林博仁(Buo-ren Lin)),<br/>
+ Run Travis ASan tests with "sudo: required" (Nick Wellnhofer),<br/>
+ Improve restoring of context size and position (Nick Wellnhofer),<br/>
+ Simplify and harden nodeset filtering (Nick Wellnhofer),<br/>
+ Avoid unnecessary backups of the context node (Nick Wellnhofer),<br/>
+ Fix inconsistency in xmlXPathIsInf (Nick Wellnhofer)<br/>
+ </li>
+
+ <li>Cleanups:<br/>
+ </li>
+</ul>
<h3>v2.9.8: Mar 05 2018</h3>
<ul>
<li>Portability:<br/>
diff --git a/encoding.c b/encoding.c
index de7b511..a3aaf10 100644
--- a/encoding.c
+++ b/encoding.c
@@ -2460,8 +2460,6 @@
ret = -3;
}
- if (ret >= 0) output += ret;
-
/*
* Attempt to handle error cases
*/
diff --git a/include/libxml/globals.h b/include/libxml/globals.h
index a9305aa..5e41b7b 100644
--- a/include/libxml/globals.h
+++ b/include/libxml/globals.h
@@ -17,7 +17,6 @@
#include <libxml/xmlversion.h>
#include <libxml/parser.h>
#include <libxml/xmlerror.h>
-#include <libxml/SAX.h>
#include <libxml/SAX2.h>
#include <libxml/xmlmemory.h>
diff --git a/include/libxml/tree.h b/include/libxml/tree.h
index 4a9b3bc..626ed6a 100644
--- a/include/libxml/tree.h
+++ b/include/libxml/tree.h
@@ -575,7 +575,7 @@
void *ids; /* Hash table for ID attributes if any */
void *refs; /* Hash table for IDREFs attributes if any */
const xmlChar *URL; /* The URI for that document */
- int charset; /* encoding of the in-memory content
+ int charset; /* Internal flag for charset handling,
actually an xmlCharEncoding */
struct _xmlDict *dict; /* dict used to allocate names or NULL */
void *psvi; /* for type/PSVI informations */
diff --git a/include/libxml/xmlexports.h b/include/libxml/xmlexports.h
index 2c79f81..31ab8a1 100644
--- a/include/libxml/xmlexports.h
+++ b/include/libxml/xmlexports.h
@@ -131,8 +131,8 @@
#endif
#endif
-/* Cygwin platform, GNU compiler */
-#if defined(_WIN32) && defined(__CYGWIN__)
+/* Cygwin platform (does not define _WIN32), GNU compiler */
+#if defined(__CYGWIN__)
#undef XMLPUBFUN
#undef XMLPUBVAR
#undef XMLCALL
@@ -145,7 +145,7 @@
#if !defined(LIBXML_STATIC)
#define XMLPUBVAR __declspec(dllimport) extern
#else
- #define XMLPUBVAR
+ #define XMLPUBVAR extern
#endif
#endif
#define XMLCALL __cdecl
diff --git a/include/libxml/xmlversion.h b/include/libxml/xmlversion.h
index eb18ea0..af022db 100644
--- a/include/libxml/xmlversion.h
+++ b/include/libxml/xmlversion.h
@@ -29,28 +29,28 @@
*
* the version string like "1.2.3"
*/
-#define LIBXML_DOTTED_VERSION "2.9.2"
+#define LIBXML_DOTTED_VERSION "2.9.9"
/**
* LIBXML_VERSION:
*
* the version number: 1.2.3 value is 10203
*/
-#define LIBXML_VERSION 20902
+#define LIBXML_VERSION 20909
/**
* LIBXML_VERSION_STRING:
*
* the version number string, 1.2.3 value is "10203"
*/
-#define LIBXML_VERSION_STRING "20902"
+#define LIBXML_VERSION_STRING "20909"
/**
* LIBXML_VERSION_EXTRA:
*
* extra version information, used to show a CVS compilation
*/
-#define LIBXML_VERSION_EXTRA "-GITv2.9.2-48-g4b31d61"
+#define LIBXML_VERSION_EXTRA "-GITv2.9.9-rc2-2-g7c4949afa"
/**
* LIBXML_TEST_VERSION:
@@ -58,7 +58,7 @@
* Macro to check that the libxml version in use is compatible with
* the version the software has been compiled against
*/
-#define LIBXML_TEST_VERSION xmlCheckVersion(20902);
+#define LIBXML_TEST_VERSION xmlCheckVersion(20909);
#ifndef VMS
#if 0
@@ -174,8 +174,8 @@
*
* Whether the FTP support is configured in
*/
-#if 0
-#define LIBXML_FTP_ENABLED
+#if 1
+#undef LIBXML_FTP_ENABLED
#endif
/**
@@ -183,8 +183,8 @@
*
* Whether the HTTP support is configured in
*/
-#if 0
-#define LIBXML_HTTP_ENABLED
+#if 1
+#undef LIBXML_HTTP_ENABLED
#endif
/**
@@ -201,13 +201,8 @@
*
* Whether the HTML support is configured in
*/
-// BEGIN android-changed
-// Was: #if 1
-// HTML support has security vulnerabilities (b/27338391)
-// and is not needed in Android.
-#if 0
-// END android-changed
-#define LIBXML_HTML_ENABLED
+#if 1
+#undef LIBXML_HTML_ENABLED
#endif
/**
@@ -215,8 +210,8 @@
*
* Whether the deprecated APIs are compiled in for compatibility
*/
-#if 0
-#define LIBXML_LEGACY_ENABLED
+#if 1
+#undef LIBXML_LEGACY_ENABLED
#endif
/**
@@ -278,22 +273,18 @@
*
* Whether iconv support is available
*/
-#if 0
-#define LIBXML_ICONV_ENABLED
+#if 1
+#undef LIBXML_ICONV_ENABLED
#endif
/**
* LIBXML_ICU_ENABLED:
*
* Whether icu support is available
- *
- * This is disabled when libxml2 is built for the VNDK.
- * libicuuc.so isn't available in the VNDK.
*/
+#undef LIBXML_ICU_ENABLED
#ifndef __ANDROID_VNDK__
#define LIBXML_ICU_ENABLED
-#else
-#undef LIBXML_ICU_ENABLED
#endif
/**
@@ -406,8 +397,8 @@
*
* Whether the Zlib support is compiled in
*/
-#if 0
-#define LIBXML_ZLIB_ENABLED
+#if 1
+#undef LIBXML_ZLIB_ENABLED
#endif
/**
@@ -415,14 +406,11 @@
*
* Whether the Lzma support is compiled in
*/
-#if 0
-#define LIBXML_LZMA_ENABLED
+#if 1
+#undef LIBXML_LZMA_ENABLED
#endif
#ifdef __GNUC__
-#ifdef HAVE_ANSIDECL_H
-#include <ansidecl.h>
-#endif
/**
* ATTRIBUTE_UNUSED:
@@ -445,7 +433,7 @@
*/
#ifndef LIBXML_ATTR_ALLOC_SIZE
-# if ((__GNUC__ > 4) || ((__GNUC__ == 4) && (__GNUC_MINOR__ >= 3)))
+# if (!defined(__clang__) && ((__GNUC__ > 4) || ((__GNUC__ == 4) && (__GNUC_MINOR__ >= 3))))
# define LIBXML_ATTR_ALLOC_SIZE(x) __attribute__((alloc_size(x)))
# else
# define LIBXML_ATTR_ALLOC_SIZE(x)
diff --git a/parser.c b/parser.c
index ca9fde2..5813a66 100644
--- a/parser.c
+++ b/parser.c
@@ -12462,7 +12462,12 @@
ctxt->input->free((xmlChar *) ctxt->input->base);
ctxt->input->free = NULL;
}
+ if (ctxt->input->buf != NULL) {
+ xmlFreeParserInputBuffer(ctxt->input->buf);
+ ctxt->input->buf = NULL;
+ }
ctxt->input->cur = BAD_CAST"";
+ ctxt->input->length = 0;
ctxt->input->base = ctxt->input->cur;
ctxt->input->end = ctxt->input->cur;
}
diff --git a/parserInternals.c b/parserInternals.c
index 8c0cd57..0f015de 100644
--- a/parserInternals.c
+++ b/parserInternals.c
@@ -1214,7 +1214,7 @@
/*
* convert as much as possible of the buffer
*/
- nbchars = xmlCharEncInput(input->buf, 0);
+ nbchars = xmlCharEncInput(input->buf, 1);
} else {
/*
* convert just enough to get
@@ -1240,8 +1240,18 @@
* size to be able to convert the buffer.
*/
xmlErrInternal(ctxt, "switching encoding : no input\n", NULL);
+ /*
+ * Callers assume that the input buffer takes ownership of the
+ * encoding handler. xmlCharEncCloseFunc frees unregistered
+ * handlers and avoids a memory leak.
+ */
+ xmlCharEncCloseFunc(handler);
return (-1);
}
+ /*
+ * We should actually raise an error here, see issue #34.
+ */
+ xmlCharEncCloseFunc(handler);
return (0);
}
diff --git a/relaxng.c b/relaxng.c
index 88d351d..914706a 100644
--- a/relaxng.c
+++ b/relaxng.c
@@ -1573,6 +1573,9 @@
#endif
}
}
+ if (xmlRelaxNGRemoveRedefine(ctxt, URL, tmp->children, name) == 1) {
+ found = 1;
+ }
}
tmp = tmp2;
}
@@ -1739,7 +1742,18 @@
xmlFree(name);
}
}
- cur = cur->next;
+ if (IS_RELAXNG(cur, "div") && cur->children != NULL) {
+ cur = cur->children;
+ } else {
+ if (cur->next != NULL) {
+ cur = cur->next;
+ } else {
+ while (cur->parent != node && cur->parent->next == NULL) {
+ cur = cur->parent;
+ }
+ cur = cur->parent != node ? cur->parent->next : NULL;
+ }
+ }
}
@@ -3979,7 +3993,7 @@
* xmlRelaxNGGetElements:
* @ctxt: a Relax-NG parser context
* @def: the definition definition
- * @eora: gather elements (0) or attributes (1)
+ * @eora: gather elements (0), attributes (1) or elements and text (2)
*
* Compute the list of top elements a definition can generate
*
@@ -4005,7 +4019,12 @@
while (cur != NULL) {
if (((eora == 0) && ((cur->type == XML_RELAXNG_ELEMENT) ||
(cur->type == XML_RELAXNG_TEXT))) ||
- ((eora == 1) && (cur->type == XML_RELAXNG_ATTRIBUTE))) {
+ ((eora == 1) && (cur->type == XML_RELAXNG_ATTRIBUTE)) ||
+ ((eora == 2) && ((cur->type == XML_RELAXNG_DATATYPE) ||
+ (cur->type == XML_RELAXNG_ELEMENT) ||
+ (cur->type == XML_RELAXNG_LIST) ||
+ (cur->type == XML_RELAXNG_TEXT) ||
+ (cur->type == XML_RELAXNG_VALUE)))) {
if (ret == NULL) {
max = 10;
ret = (xmlRelaxNGDefinePtr *)
@@ -4360,7 +4379,7 @@
if (cur->type == XML_RELAXNG_TEXT)
is_mixed++;
groups[nbgroups]->rule = cur;
- groups[nbgroups]->defs = xmlRelaxNGGetElements(ctxt, cur, 0);
+ groups[nbgroups]->defs = xmlRelaxNGGetElements(ctxt, cur, 2);
groups[nbgroups]->attrs = xmlRelaxNGGetElements(ctxt, cur, 1);
nbgroups++;
cur = cur->next;
@@ -5347,11 +5366,15 @@
xmlNodePtr child;
xmlRelaxNGDefinePtr last = NULL;
- ret = xmlRelaxNGNewDefine(ctxt, node);
- if (ret == NULL)
- return (NULL);
- ret->parent = def;
- ret->type = XML_RELAXNG_CHOICE;
+ if (def->type == XML_RELAXNG_CHOICE) {
+ ret = def;
+ } else {
+ ret = xmlRelaxNGNewDefine(ctxt, node);
+ if (ret == NULL)
+ return (NULL);
+ ret->parent = def;
+ ret->type = XML_RELAXNG_CHOICE;
+ }
if (node->children == NULL) {
xmlRngPErr(ctxt, node, XML_RNGP_CHOICE_EMPTY,
@@ -5363,7 +5386,7 @@
tmp = xmlRelaxNGParseNameClass(ctxt, child, ret);
if (tmp != NULL) {
if (last == NULL) {
- last = ret->nameClass = tmp;
+ last = tmp;
} else {
last->next = tmp;
last = tmp;
@@ -9262,7 +9285,10 @@
return (1);
} else if (((node->type == XML_TEXT_NODE) ||
(node->type == XML_CDATA_SECTION_NODE)) &&
- (cur->type == XML_RELAXNG_TEXT)) {
+ ((cur->type == XML_RELAXNG_DATATYPE) ||
+ (cur->type == XML_RELAXNG_LIST) ||
+ (cur->type == XML_RELAXNG_TEXT) ||
+ (cur->type == XML_RELAXNG_VALUE))) {
return (1);
}
cur = list[i++];
diff --git a/result/errors/759573.xml.err b/result/errors/759573.xml.err
index 554039f..38ef5c4 100644
--- a/result/errors/759573.xml.err
+++ b/result/errors/759573.xml.err
@@ -21,14 +21,11 @@
^
./test/errors/759573.xml:1: parser error : internal error: xmlParseInternalSubset: error detected in Markup declaration
-<?h?><!DOCTYPEt[<!ELEMENT t (A)><!ENTITY % xx '%<![INCLUDE[000%ஸ00
- ^
+
+^
./test/errors/759573.xml:1: parser error : DOCTYPE improperly terminated
-<?h?><!DOCTYPEt[<!ELEMENT t (A)><!ENTITY % xx '%<![INCLUDE[000%ஸ00
- ^
-./test/errors/759573.xml:1: parser error : StartTag: invalid element name
-<?h?><!DOCTYPEt[<!ELEMENT t (A)><!ENTITY % xx '%<![INCLUDE[000%ஸ00
- ^
-./test/errors/759573.xml:1: parser error : Extra content at the end of the document
-<?h?><!DOCTYPEt[<!ELEMENT t (A)><!ENTITY % xx '%<![INCLUDE[000%ஸ00
- ^
+
+^
+./test/errors/759573.xml:1: parser error : Start tag expected, '<' not found
+
+^
diff --git a/result/relaxng/anyName0_0 b/result/relaxng/anyName0_0
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/anyName0_0
diff --git a/result/relaxng/anyName0_0.err b/result/relaxng/anyName0_0.err
new file mode 100644
index 0000000..f660c26
--- /dev/null
+++ b/result/relaxng/anyName0_0.err
@@ -0,0 +1 @@
+./test/relaxng/anyName0_0.xml validates
diff --git a/result/relaxng/anyName0_err b/result/relaxng/anyName0_err
new file mode 100644
index 0000000..ab6f08d
--- /dev/null
+++ b/result/relaxng/anyName0_err
@@ -0,0 +1 @@
+./test/relaxng/anyName0.rng validates
diff --git a/result/relaxng/anyName0_valid b/result/relaxng/anyName0_valid
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/anyName0_valid
diff --git a/result/relaxng/include0_0 b/result/relaxng/include0_0
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/include0_0
diff --git a/result/relaxng/include0_0.err b/result/relaxng/include0_0.err
new file mode 100644
index 0000000..24013bc
--- /dev/null
+++ b/result/relaxng/include0_0.err
@@ -0,0 +1 @@
+./test/relaxng/include0_0.xml validates
diff --git a/result/relaxng/include0_err b/result/relaxng/include0_err
new file mode 100644
index 0000000..cd16cbe
--- /dev/null
+++ b/result/relaxng/include0_err
@@ -0,0 +1 @@
+./test/relaxng/include0.rng validates
diff --git a/result/relaxng/include0_valid b/result/relaxng/include0_valid
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/include0_valid
diff --git a/result/relaxng/include1_0 b/result/relaxng/include1_0
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/include1_0
diff --git a/result/relaxng/include1_0.err b/result/relaxng/include1_0.err
new file mode 100644
index 0000000..8c13f4c
--- /dev/null
+++ b/result/relaxng/include1_0.err
@@ -0,0 +1 @@
+./test/relaxng/include1_0.xml validates
diff --git a/result/relaxng/include1_err b/result/relaxng/include1_err
new file mode 100644
index 0000000..ad4850b
--- /dev/null
+++ b/result/relaxng/include1_err
@@ -0,0 +1 @@
+./test/relaxng/include1.rng validates
diff --git a/result/relaxng/include1_valid b/result/relaxng/include1_valid
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/result/relaxng/include1_valid
diff --git a/runtest.c b/runtest.c
index 605c1d1..addda5c 100644
--- a/runtest.c
+++ b/runtest.c
@@ -4413,9 +4413,10 @@
char *result;
char *error;
int mem;
- xmlCharEncodingHandlerPtr ebcdicHandler;
+ xmlCharEncodingHandlerPtr ebcdicHandler, eucJpHandler;
ebcdicHandler = xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EBCDIC);
+ eucJpHandler = xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EUC_JP);
if (tst == NULL) return(-1);
if (tst->in != NULL) {
@@ -4426,8 +4427,10 @@
for (i = 0;i < globbuf.gl_pathc;i++) {
if (!checkTestFile(globbuf.gl_pathv[i]))
continue;
- if ((ebcdicHandler == NULL) &&
- (strstr(globbuf.gl_pathv[i], "ebcdic") != NULL))
+ if (((ebcdicHandler == NULL) &&
+ (strstr(globbuf.gl_pathv[i], "ebcdic") != NULL)) ||
+ ((eucJpHandler == NULL) &&
+ (strstr(globbuf.gl_pathv[i], "icu_parse_test") != NULL)))
continue;
if (tst->suffix != NULL) {
result = resultFilename(globbuf.gl_pathv[i], tst->out,
@@ -4496,6 +4499,7 @@
}
xmlCharEncCloseFunc(ebcdicHandler);
+ xmlCharEncCloseFunc(eucJpHandler);
return(err);
}
diff --git a/test/relaxng/anyName0.rng b/test/relaxng/anyName0.rng
new file mode 100644
index 0000000..9fc5ead
--- /dev/null
+++ b/test/relaxng/anyName0.rng
@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <start>
+ <element>
+ <choice>
+ <choice>
+ <name>a</name>
+ <name>b</name>
+ </choice>
+ <name>c</name>
+ </choice>
+ <empty/>
+ </element>
+ </start>
+</grammar>
diff --git a/test/relaxng/anyName0_0.xml b/test/relaxng/anyName0_0.xml
new file mode 100644
index 0000000..fb08e36
--- /dev/null
+++ b/test/relaxng/anyName0_0.xml
@@ -0,0 +1 @@
+<b/>
diff --git a/test/relaxng/include0.inc b/test/relaxng/include0.inc
new file mode 100644
index 0000000..8636ee4
--- /dev/null
+++ b/test/relaxng/include0.inc
@@ -0,0 +1,5 @@
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <start>
+ <notAllowed/>
+ </start>
+</grammar>
diff --git a/test/relaxng/include0.rng b/test/relaxng/include0.rng
new file mode 100644
index 0000000..e8edb92
--- /dev/null
+++ b/test/relaxng/include0.rng
@@ -0,0 +1,11 @@
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <include href="include0.inc">
+ <div>
+ <start>
+ <element name="root">
+ <empty/>
+ </element>
+ </start>
+ </div>
+ </include>
+</grammar>
diff --git a/test/relaxng/include0_0.xml b/test/relaxng/include0_0.xml
new file mode 100644
index 0000000..f3f286e
--- /dev/null
+++ b/test/relaxng/include0_0.xml
@@ -0,0 +1 @@
+<root/>
diff --git a/test/relaxng/include1-0.inc b/test/relaxng/include1-0.inc
new file mode 100644
index 0000000..87c6802
--- /dev/null
+++ b/test/relaxng/include1-0.inc
@@ -0,0 +1,12 @@
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <include href="include1-1.inc">
+ <start>
+ <notAllowed/>
+ </start>
+ </include>
+ <start combine="choice">
+ <element name="b">
+ <empty/>
+ </element>
+ </start>
+</grammar>
diff --git a/test/relaxng/include1-1.inc b/test/relaxng/include1-1.inc
new file mode 100644
index 0000000..8af51ec
--- /dev/null
+++ b/test/relaxng/include1-1.inc
@@ -0,0 +1,7 @@
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <start>
+ <element name="c">
+ <empty/>
+ </element>
+ </start>
+</grammar>
diff --git a/test/relaxng/include1.rng b/test/relaxng/include1.rng
new file mode 100644
index 0000000..c36bff9
--- /dev/null
+++ b/test/relaxng/include1.rng
@@ -0,0 +1,12 @@
+<grammar xmlns="http://relaxng.org/ns/structure/1.0">
+ <include href="include1-0.inc">
+ <start>
+ <notAllowed/>
+ </start>
+ </include>
+ <start combine="choice">
+ <element name="a">
+ <empty/>
+ </element>
+ </start>
+</grammar>
diff --git a/test/relaxng/include1_0.xml b/test/relaxng/include1_0.xml
new file mode 100644
index 0000000..d80a5e2
--- /dev/null
+++ b/test/relaxng/include1_0.xml
@@ -0,0 +1 @@
+<a/>
diff --git a/testURI.c b/testURI.c
index d20989d..0043b20 100644
--- a/testURI.c
+++ b/testURI.c
@@ -19,6 +19,7 @@
static const char *base = NULL;
static int escape = 0;
static int debug = 0;
+static int relative = 0;
static void handleURI(const char *str) {
int ret;
@@ -57,7 +58,12 @@
}
}
} else {
- res = xmlBuildURI((xmlChar *)str, (xmlChar *) base);
+ if (relative) {
+ res = xmlBuildRelativeURI((xmlChar *)str, (xmlChar *) base);
+ } else {
+ res = xmlBuildURI((xmlChar *)str, (xmlChar *) base);
+ }
+
if (res != NULL) {
printf("%s\n", (char *) res);
}
@@ -75,6 +81,11 @@
int i, arg = 1;
if ((argc > arg) && (argv[arg] != NULL) &&
+ (!strcmp(argv[arg], "--relative"))) {
+ arg++;
+ relative++;
+ }
+ if ((argc > arg) && (argv[arg] != NULL) &&
((!strcmp(argv[arg], "-base")) || (!strcmp(argv[arg], "--base")))) {
arg++;
base = argv[arg];
diff --git a/uri.c b/uri.c
index 860b203..2cf8d9f 100644
--- a/uri.c
+++ b/uri.c
@@ -12,7 +12,6 @@
#include "libxml.h"
#include <string.h>
-#include <limits.h>
#include <libxml/xmlmemory.h>
#include <libxml/uri.h>
@@ -335,7 +334,7 @@
cur++;
}
if (uri != NULL)
- uri->port = port & USHRT_MAX; /* port value modulo INT_MAX+1 */
+ uri->port = port & INT_MAX; /* port value modulo INT_MAX+1 */
*str = cur;
return(0);
}
@@ -2237,25 +2236,8 @@
* First we take care of the special case where either of the
* two path components may be missing (bug 316224)
*/
- if (bas->path == NULL) {
- if (ref->path != NULL) {
- uptr = (xmlChar *) ref->path;
- if (*uptr == '/')
- uptr++;
- /* exception characters from xmlSaveUri */
- val = xmlURIEscapeStr(uptr, BAD_CAST "/;&=+$,");
- }
- goto done;
- }
bptr = (xmlChar *)bas->path;
- if (ref->path == NULL) {
- for (ix = 0; bptr[ix] != 0; ix++) {
- if (bptr[ix] == '/')
- nbslash++;
- }
- uptr = NULL;
- len = 1; /* this is for a string terminator only */
- } else {
+ {
xmlChar *rptr = (xmlChar *) ref->path;
int pos = 0;
@@ -2281,30 +2263,28 @@
* beginning of the "unique" suffix of URI
*/
ix = pos;
- if ((rptr[ix] == '/') && (ix > 0))
- ix--;
- else if ((rptr[ix] == 0) && (ix > 1) && (rptr[ix - 1] == '/'))
- ix -= 2;
for (; ix > 0; ix--) {
- if (rptr[ix] == '/')
+ if (rptr[ix - 1] == '/')
break;
}
- if (ix == 0) {
- uptr = (xmlChar *)rptr;
- } else {
- ix++;
- uptr = (xmlChar *)&rptr[ix];
- }
+ uptr = (xmlChar *)&rptr[ix];
/*
* In base, count the number of '/' from the differing point
*/
- if (bptr[pos] != rptr[pos]) {/* check for trivial URI == base */
- for (; bptr[ix] != 0; ix++) {
- if (bptr[ix] == '/')
- nbslash++;
- }
+ for (; bptr[ix] != 0; ix++) {
+ if (bptr[ix] == '/')
+ nbslash++;
}
+
+ /*
+ * e.g: URI="foo/" base="foo/bar" -> "./"
+ */
+ if (nbslash == 0 && !uptr[0]) {
+ val = xmlStrdup(BAD_CAST "./");
+ goto done;
+ }
+
len = xmlStrlen (uptr) + 1;
}
diff --git a/valid.c b/valid.c
index 57a1a9e..b1cfede 100644
--- a/valid.c
+++ b/valid.c
@@ -4660,13 +4660,13 @@
#if 0
/* Validity Constraint: ID uniqueness */
if (attrDecl->atype == XML_ATTRIBUTE_ID) {
- if (xmlAddID(ctxt, doc, value, (xmlAttrPtr) attrDecl) == NULL)
+ if (xmlAddID(ctxt, doc, value, (xmlAttrPtr) ns) == NULL)
ret = 0;
}
if ((attrDecl->atype == XML_ATTRIBUTE_IDREF) ||
(attrDecl->atype == XML_ATTRIBUTE_IDREFS)) {
- if (xmlAddRef(ctxt, doc, value, (xmlAttrPtr) attrDecl) == NULL)
+ if (xmlAddRef(ctxt, doc, value, (xmlAttrPtr) ns) == NULL)
ret = 0;
}
#endif
diff --git a/win32/Makefile.msvc b/win32/Makefile.msvc
index 491dc88..ee4250a 100644
--- a/win32/Makefile.msvc
+++ b/win32/Makefile.msvc
@@ -244,6 +244,12 @@
$(XML_INTDIR_A_DLL)\xpointer.obj\
$(XML_INTDIR_A_DLL)\xmlstring.obj
+!if "$(WITH_LZMA)" == "1"
+XML_OBJS = $(XML_OBJS) $(XML_INTDIR)\xzlib.obj
+XML_OBJS_A = $(XML_OBJS_A) $(XML_INTDIR_A)\xzlib.obj
+XML_OBJS_A_DLL = $(XML_OBJS_A_DLL) $(XML_INTDIR_A_DLL)\xzlib.obj
+!endif
+
# Xmllint and friends executables.
UTILS = $(BINDIR)\xmllint.exe\
$(BINDIR)\xmlcatalog.exe\
diff --git a/xmlIO.c b/xmlIO.c
index 8254347..f61dd05 100644
--- a/xmlIO.c
+++ b/xmlIO.c
@@ -3157,7 +3157,7 @@
* convert as much as possible to the parser reading buffer.
*/
use = xmlBufUse(in->raw);
- nbchars = xmlCharEncInput(in, 0);
+ nbchars = xmlCharEncInput(in, 1);
if (nbchars < 0) {
xmlIOErr(XML_IO_ENCODER, NULL);
in->error = XML_IO_ENCODER;
@@ -3273,7 +3273,7 @@
* convert as much as possible to the parser reading buffer.
*/
use = xmlBufUse(in->raw);
- nbchars = xmlCharEncInput(in, 0);
+ nbchars = xmlCharEncInput(in, 1);
if (nbchars < 0) {
xmlIOErr(XML_IO_ENCODER, NULL);
in->error = XML_IO_ENCODER;
diff --git a/xmllint.c b/xmllint.c
index 6512540..d6f5eb0 100644
--- a/xmllint.c
+++ b/xmllint.c
@@ -2071,51 +2071,52 @@
int i;
xmlNodePtr node;
#ifdef LIBXML_OUTPUT_ENABLED
- xmlSaveCtxtPtr ctxt;
+ xmlOutputBufferPtr buf;
if ((cur->nodesetval == NULL) || (cur->nodesetval->nodeNr <= 0)) {
fprintf(stderr, "XPath set is empty\n");
progresult = XMLLINT_ERR_XPATH;
break;
}
- ctxt = xmlSaveToFd(1, NULL, 0);
- if (ctxt == NULL) {
+ buf = xmlOutputBufferCreateFile(stdout, NULL);
+ if (buf == NULL) {
fprintf(stderr, "Out of memory for XPath\n");
progresult = XMLLINT_ERR_MEM;
return;
}
for (i = 0;i < cur->nodesetval->nodeNr;i++) {
node = cur->nodesetval->nodeTab[i];
- xmlSaveTree(ctxt, node);
+ xmlNodeDumpOutput(buf, node->doc, node, 0, 0, NULL);
+ xmlOutputBufferWrite(buf, 1, "\n");
}
- xmlSaveClose(ctxt);
+ xmlOutputBufferClose(buf);
#else
printf("xpath returned %d nodes\n", cur->nodesetval->nodeNr);
#endif
break;
}
case XPATH_BOOLEAN:
- if (cur->boolval) printf("true");
- else printf("false");
+ if (cur->boolval) printf("true\n");
+ else printf("false\n");
break;
case XPATH_NUMBER:
switch (xmlXPathIsInf(cur->floatval)) {
case 1:
- printf("Infinity");
+ printf("Infinity\n");
break;
case -1:
- printf("-Infinity");
+ printf("-Infinity\n");
break;
default:
if (xmlXPathIsNaN(cur->floatval)) {
- printf("NaN");
+ printf("NaN\n");
} else {
- printf("%0g", cur->floatval);
+ printf("%0g\n", cur->floatval);
}
}
break;
case XPATH_STRING:
- printf("%s", (const char *) cur->stringval);
+ printf("%s\n", (const char *) cur->stringval);
break;
case XPATH_UNDEFINED:
fprintf(stderr, "XPath Object is uninitialized\n");
diff --git a/xmlreader.c b/xmlreader.c
index 4053269..db310c8 100644
--- a/xmlreader.c
+++ b/xmlreader.c
@@ -1711,10 +1711,11 @@
if (xmlTextReaderExpand(reader) == NULL) {
return NULL;
}
- doc = reader->doc;
+ doc = reader->node->doc;
buff = xmlBufferCreate();
for (cur_node = reader->node->children; cur_node != NULL;
cur_node = cur_node->next) {
+ /* XXX: Why is the node copied? */
node = xmlDocCopyNode(cur_node, doc, 1);
buff2 = xmlBufferCreate();
if (xmlNodeDump(buff2, doc, node, 0, 0) == -1) {
@@ -1755,10 +1756,11 @@
xmlDocPtr doc;
node = reader->node;
- doc = reader->doc;
+ doc = node->doc;
if (xmlTextReaderExpand(reader) == NULL) {
return NULL;
}
+ /* XXX: Why is the node copied? */
if (node->type == XML_DTD_NODE) {
node = (xmlNodePtr) xmlCopyDtd((xmlDtdPtr) node);
} else {
@@ -1917,12 +1919,9 @@
/* if reader->node->next is NULL mean no subtree for current node,
so need to move to sibling of parent node if present */
- if ((reader->node->type == XML_ELEMENT_NODE) ||
- (reader->node->type == XML_ATTRIBUTE_NODE)) {
- reader->state = XML_TEXTREADER_BACKTRACK;
- /* This will move to parent if present */
- xmlTextReaderRead(reader);
- }
+ reader->state = XML_TEXTREADER_BACKTRACK;
+ /* This will move to parent if present */
+ xmlTextReaderRead(reader);
}
if (reader->node->next != 0) {
@@ -3006,7 +3005,7 @@
* Reference:
* http://www.gnu.org/software/dotgnu/pnetlib-doc/System/Xml/XmlNodeType.html
*
- * Returns the xmlNodeType of the current node or -1 in case of error
+ * Returns the xmlReaderTypes of the current node or -1 in case of error
*/
int
xmlTextReaderNodeType(xmlTextReaderPtr reader) {
diff --git a/xmlsave.c b/xmlsave.c
index 6c7418e..7a05d83 100644
--- a/xmlsave.c
+++ b/xmlsave.c
@@ -1123,9 +1123,6 @@
cur->encoding = BAD_CAST ctxt->encoding;
} else if (cur->encoding != NULL) {
encoding = cur->encoding;
- } else if (cur->charset != XML_CHAR_ENCODING_UTF8) {
- encoding = (const xmlChar *)
- xmlGetCharEncodingName((xmlCharEncoding) cur->charset);
}
if (((cur->type == XML_HTML_DOCUMENT_NODE) &&
diff --git a/xmlschemas.c b/xmlschemas.c
index 405f72a..019988a 100644
--- a/xmlschemas.c
+++ b/xmlschemas.c
@@ -27653,6 +27653,17 @@
vctxt->nbIdcNodes = 0;
vctxt->sizeIdcNodes = 0;
}
+
+ if (vctxt->idcKeys != NULL) {
+ int i;
+ for (i = 0; i < vctxt->nbIdcKeys; i++)
+ xmlSchemaIDCFreeKey(vctxt->idcKeys[i]);
+ xmlFree(vctxt->idcKeys);
+ vctxt->idcKeys = NULL;
+ vctxt->nbIdcKeys = 0;
+ vctxt->sizeIdcKeys = 0;
+ }
+
/*
* Note that we won't delete the XPath state pool here.
*/
diff --git a/xpath.c b/xpath.c
index d134025..5e3bb9f 100644
--- a/xpath.c
+++ b/xpath.c
@@ -2,7 +2,7 @@
* xpath.c: XML Path Language implementation
* XPath is a language for addressing parts of an XML document,
* designed to be used by both XSLT and XPointer
- *f
+ *
* Reference: W3C Recommendation 16 November 1999
* http://www.w3.org/TR/1999/REC-xpath-19991116
* Public reference:
@@ -477,27 +477,28 @@
* *
************************************************************************/
-#ifndef NAN
-#define NAN (0.0 / 0.0)
-#endif
-
#ifndef INFINITY
-#define INFINITY HUGE_VAL
+#define INFINITY (DBL_MAX * DBL_MAX)
#endif
-double xmlXPathNAN = NAN;
-double xmlXPathPINF = INFINITY;
-double xmlXPathNINF = -INFINITY;
+#ifndef NAN
+#define NAN (INFINITY / INFINITY)
+#endif
+
+double xmlXPathNAN;
+double xmlXPathPINF;
+double xmlXPathNINF;
/**
* xmlXPathInit:
*
* Initialize the XPath environment
- *
- * Does nothing but must be kept as public function.
*/
void
xmlXPathInit(void) {
+ xmlXPathNAN = NAN;
+ xmlXPathPINF = INFINITY;
+ xmlXPathNINF = -INFINITY;
}
/**
@@ -526,9 +527,9 @@
#ifdef isinf
return isinf(val) ? (val > 0 ? 1 : -1) : 0;
#else
- if (val >= HUGE_VAL)
+ if (val >= INFINITY)
return 1;
- if (val <= -HUGE_VAL)
+ if (val <= -INFINITY)
return -1;
return 0;
#endif
@@ -867,15 +868,14 @@
XPATH_OP_UNION,
XPATH_OP_ROOT,
XPATH_OP_NODE,
- XPATH_OP_RESET, /* 10 */
XPATH_OP_COLLECT,
- XPATH_OP_VALUE, /* 12 */
+ XPATH_OP_VALUE, /* 11 */
XPATH_OP_VARIABLE,
XPATH_OP_FUNCTION,
XPATH_OP_ARG,
XPATH_OP_PREDICATE,
- XPATH_OP_FILTER, /* 17 */
- XPATH_OP_SORT /* 18 */
+ XPATH_OP_FILTER, /* 16 */
+ XPATH_OP_SORT /* 17 */
#ifdef LIBXML_XPTR_ENABLED
,XPATH_OP_RANGETO
#endif
@@ -1525,8 +1525,6 @@
fprintf(output, "ROOT"); break;
case XPATH_OP_NODE:
fprintf(output, "NODE"); break;
- case XPATH_OP_RESET:
- fprintf(output, "RESET"); break;
case XPATH_OP_SORT:
fprintf(output, "SORT"); break;
case XPATH_OP_COLLECT: {
@@ -4376,7 +4374,7 @@
}
memset(ret, 0 , (size_t) sizeof(xmlXPathObject));
ret->type = XPATH_XSLT_TREE;
- ret->boolval = 0;
+ ret->boolval = 1;
ret->user = (void *) val;
ret->nodesetval = xmlXPathNodeSetCreate(val);
#ifdef XP_DEBUG_OBJ_USAGE
@@ -8479,9 +8477,8 @@
xmlXPathRoot(xmlXPathParserContextPtr ctxt) {
if ((ctxt == NULL) || (ctxt->context == NULL))
return;
- ctxt->context->node = (xmlNodePtr) ctxt->context->doc;
valuePush(ctxt, xmlXPathCacheNewNodeSet(ctxt->context,
- ctxt->context->node));
+ (xmlNodePtr) ctxt->context->doc));
}
/************************************************************************
@@ -10734,7 +10731,6 @@
PUSH_LONG_EXPR(XPATH_OP_COLLECT, AXIS_DESCENDANT_OR_SELF,
NODE_TEST_TYPE, NODE_TYPE_NODE, NULL, NULL);
- PUSH_UNARY_EXPR(XPATH_OP_RESET, ctxt->comp->last, 1, 0);
xmlXPathCompRelativeLocationPath(ctxt);
} else if (CUR == '/') {
@@ -11665,6 +11661,7 @@
xmlXPathContextPtr xpctxt = ctxt->context;
xmlNodePtr contextNode, oldContextNode;
xmlDocPtr oldContextDoc;
+ int oldcs, oldpp;
int i, res, contextPos = 0, newContextSize;
xmlXPathStepOpPtr exprOp;
xmlXPathObjectPtr contextObj = NULL, exprRes = NULL;
@@ -11701,6 +11698,8 @@
*/
oldContextNode = xpctxt->node;
oldContextDoc = xpctxt->doc;
+ oldcs = xpctxt->contextSize;
+ oldpp = xpctxt->proximityPosition;
/*
* Get the expression of this predicate.
*/
@@ -11787,8 +11786,8 @@
*/
xpctxt->node = oldContextNode;
xpctxt->doc = oldContextDoc;
- xpctxt->contextSize = -1;
- xpctxt->proximityPosition = -1;
+ xpctxt->contextSize = oldcs;
+ xpctxt->proximityPosition = oldpp;
return(newContextSize);
}
return(contextSize);
@@ -11831,6 +11830,7 @@
return (contextSize);
} else {
xmlDocPtr oldContextDoc;
+ int oldcs, oldpp;
int i, pos = 0, newContextSize = 0, contextPos = 0, res;
xmlXPathStepOpPtr exprOp;
xmlXPathObjectPtr contextObj = NULL, exprRes = NULL;
@@ -11851,6 +11851,8 @@
*/
oldContextNode = xpctxt->node;
oldContextDoc = xpctxt->doc;
+ oldcs = xpctxt->contextSize;
+ oldpp = xpctxt->proximityPosition;
/*
* Get the expression of this predicate.
*/
@@ -11987,8 +11989,8 @@
*/
xpctxt->node = oldContextNode;
xpctxt->doc = oldContextDoc;
- xpctxt->contextSize = -1;
- xpctxt->proximityPosition = -1;
+ xpctxt->contextSize = oldcs;
+ xpctxt->proximityPosition = oldpp;
return(newContextSize);
}
return(contextSize);
@@ -12778,15 +12780,6 @@
valuePush(ctxt, xmlXPathCacheNewNodeSet(ctxt->context,
ctxt->context->node));
return (total);
- case XPATH_OP_RESET:
- if (op->ch1 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
- CHECK_ERROR0;
- if (op->ch2 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
- CHECK_ERROR0;
- ctxt->context->node = NULL;
- return (total);
case XPATH_OP_COLLECT:{
if (op->ch1 == -1)
return (total);
@@ -12842,10 +12835,6 @@
int total = 0, cur;
xmlXPathCompExprPtr comp;
xmlXPathObjectPtr arg1, arg2;
- xmlNodePtr bak;
- xmlDocPtr bakd;
- int pp;
- int cs;
CHECK_ERROR0;
comp = ctxt->comp;
@@ -12853,10 +12842,6 @@
case XPATH_OP_END:
return (0);
case XPATH_OP_UNION:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total =
xmlXPathCompOpEvalLast(ctxt, &comp->steps[op->ch1], last);
CHECK_ERROR0;
@@ -12874,10 +12859,6 @@
nodesetval->nodeNr -
1];
}
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
cur =
xmlXPathCompOpEvalLast(ctxt, &comp->steps[op->ch2], last);
CHECK_ERROR0;
@@ -12917,15 +12898,6 @@
valuePush(ctxt, xmlXPathCacheNewNodeSet(ctxt->context,
ctxt->context->node));
return (total);
- case XPATH_OP_RESET:
- if (op->ch1 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
- CHECK_ERROR0;
- if (op->ch2 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
- CHECK_ERROR0;
- ctxt->context->node = NULL;
- return (total);
case XPATH_OP_COLLECT:{
if (op->ch1 == -1)
return (0);
@@ -12970,6 +12942,7 @@
xmlNodeSetPtr oldset;
xmlNodePtr oldnode;
xmlDocPtr oldDoc;
+ int oldcs, oldpp;
int i;
CHECK_ERROR0;
@@ -13021,7 +12994,6 @@
return (total);
#ifdef LIBXML_XPTR_ENABLED
- oldnode = ctxt->context->node;
/*
* Hum are we filtering the result of an XPointer expression
*/
@@ -13036,23 +13008,17 @@
* up a new locset.
*/
CHECK_TYPE0(XPATH_LOCATIONSET);
+
+ if ((ctxt->value->user == NULL) ||
+ (((xmlLocationSetPtr) ctxt->value->user)->locNr == 0))
+ return (total);
+
obj = valuePop(ctxt);
oldlocset = obj->user;
- ctxt->context->node = NULL;
+ oldnode = ctxt->context->node;
+ oldcs = ctxt->context->contextSize;
+ oldpp = ctxt->context->proximityPosition;
- if ((oldlocset == NULL) || (oldlocset->locNr == 0)) {
- ctxt->context->contextSize = 0;
- ctxt->context->proximityPosition = 0;
- if (op->ch2 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
- res = valuePop(ctxt);
- if (res != NULL) {
- xmlXPathReleaseObject(ctxt->context, res);
- }
- valuePush(ctxt, obj);
- CHECK_ERROR0;
- return (total);
- }
newlocset = xmlXPtrLocationSetCreate(NULL);
for (i = 0; i < oldlocset->locNr; i++) {
@@ -13076,8 +13042,8 @@
if (op->ch2 != -1)
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
- xmlXPathFreeObject(obj);
- return(0);
+ xmlXPtrFreeLocationSet(newlocset);
+ goto xptr_error;
}
/*
* The result of the evaluation need to be tested to
@@ -13105,7 +13071,6 @@
/* OLD: xmlXPathFreeObject(res); */
} else
tmp = NULL;
- ctxt->context->node = NULL;
/*
* Only put the first node in the result, then leave.
*/
@@ -13120,12 +13085,12 @@
/*
* The result is used as the new evaluation locset.
*/
- xmlXPathReleaseObject(ctxt->context, obj);
- ctxt->context->node = NULL;
- ctxt->context->contextSize = -1;
- ctxt->context->proximityPosition = -1;
valuePush(ctxt, xmlXPtrWrapLocationSet(newlocset));
+xptr_error:
+ xmlXPathReleaseObject(ctxt->context, obj);
ctxt->context->node = oldnode;
+ ctxt->context->contextSize = oldcs;
+ ctxt->context->proximityPosition = oldpp;
return (total);
}
#endif /* LIBXML_XPTR_ENABLED */
@@ -13136,32 +13101,19 @@
* up a new set.
*/
CHECK_TYPE0(XPATH_NODESET);
- obj = valuePop(ctxt);
- oldset = obj->nodesetval;
- oldnode = ctxt->context->node;
- oldDoc = ctxt->context->doc;
- ctxt->context->node = NULL;
-
- if ((oldset == NULL) || (oldset->nodeNr == 0)) {
- ctxt->context->contextSize = 0;
- ctxt->context->proximityPosition = 0;
- /* QUESTION TODO: Why was this code commented out?
- if (op->ch2 != -1)
- total +=
- xmlXPathCompOpEval(ctxt,
- &comp->steps[op->ch2]);
- CHECK_ERROR0;
- res = valuePop(ctxt);
- if (res != NULL)
- xmlXPathFreeObject(res);
- */
- valuePush(ctxt, obj);
- ctxt->context->node = oldnode;
- CHECK_ERROR0;
- } else {
+ if ((ctxt->value->nodesetval != NULL) &&
+ (ctxt->value->nodesetval->nodeNr != 0)) {
xmlNodeSetPtr newset;
xmlXPathObjectPtr tmp = NULL;
+
+ obj = valuePop(ctxt);
+ oldset = obj->nodesetval;
+ oldnode = ctxt->context->node;
+ oldDoc = ctxt->context->doc;
+ oldcs = ctxt->context->contextSize;
+ oldpp = ctxt->context->proximityPosition;
+
/*
* Initialize the new set.
* Also set the xpath document in case things like
@@ -13195,8 +13147,7 @@
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
xmlXPathFreeNodeSet(newset);
- xmlXPathFreeObject(obj);
- return(0);
+ goto error;
}
/*
* The result of the evaluation needs to be tested to
@@ -13223,7 +13174,6 @@
xmlXPathNodeSetClear(tmp->nodesetval, 1);
} else
tmp = NULL;
- ctxt->context->node = NULL;
/*
* Only put the first node in the result, then leave.
*/
@@ -13238,15 +13188,14 @@
/*
* The result is used as the new evaluation set.
*/
- xmlXPathReleaseObject(ctxt->context, obj);
- ctxt->context->node = NULL;
- ctxt->context->contextSize = -1;
- ctxt->context->proximityPosition = -1;
- /* may want to move this past the '}' later */
- ctxt->context->doc = oldDoc;
valuePush(ctxt, xmlXPathCacheWrapNodeSet(ctxt->context, newset));
+error:
+ xmlXPathReleaseObject(ctxt->context, obj);
+ ctxt->context->node = oldnode;
+ ctxt->context->doc = oldDoc;
+ ctxt->context->contextSize = oldcs;
+ ctxt->context->proximityPosition = oldpp;
}
- ctxt->context->node = oldnode;
return(total);
}
#endif /* XP_OPTIMIZED_FILTER_FIRST */
@@ -13266,10 +13215,6 @@
int equal, ret;
xmlXPathCompExprPtr comp;
xmlXPathObjectPtr arg1, arg2;
- xmlNodePtr bak;
- xmlDocPtr bakd;
- int pp;
- int cs;
CHECK_ERROR0;
comp = ctxt->comp;
@@ -13277,68 +13222,42 @@
case XPATH_OP_END:
return (0);
case XPATH_OP_AND:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
xmlXPathBooleanFunction(ctxt, 1);
if ((ctxt->value == NULL) || (ctxt->value->boolval == 0))
return (total);
arg2 = valuePop(ctxt);
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
if (ctxt->error) {
xmlXPathFreeObject(arg2);
return(0);
}
xmlXPathBooleanFunction(ctxt, 1);
- arg1 = valuePop(ctxt);
- arg1->boolval &= arg2->boolval;
- valuePush(ctxt, arg1);
+ if (ctxt->value != NULL)
+ ctxt->value->boolval &= arg2->boolval;
xmlXPathReleaseObject(ctxt->context, arg2);
return (total);
case XPATH_OP_OR:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
xmlXPathBooleanFunction(ctxt, 1);
if ((ctxt->value == NULL) || (ctxt->value->boolval == 1))
return (total);
arg2 = valuePop(ctxt);
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
if (ctxt->error) {
xmlXPathFreeObject(arg2);
return(0);
}
xmlXPathBooleanFunction(ctxt, 1);
- arg1 = valuePop(ctxt);
- arg1->boolval |= arg2->boolval;
- valuePush(ctxt, arg1);
+ if (ctxt->value != NULL)
+ ctxt->value->boolval |= arg2->boolval;
xmlXPathReleaseObject(ctxt->context, arg2);
return (total);
case XPATH_OP_EQUAL:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
CHECK_ERROR0;
if (op->value)
@@ -13348,33 +13267,17 @@
valuePush(ctxt, xmlXPathCacheNewBoolean(ctxt->context, equal));
return (total);
case XPATH_OP_CMP:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
CHECK_ERROR0;
ret = xmlXPathCompareValues(ctxt, op->value, op->value2);
valuePush(ctxt, xmlXPathCacheNewBoolean(ctxt->context, ret));
return (total);
case XPATH_OP_PLUS:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
if (op->ch2 != -1) {
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
}
CHECK_ERROR0;
@@ -13390,16 +13293,8 @@
}
return (total);
case XPATH_OP_MULT:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
CHECK_ERROR0;
if (op->value == 0)
@@ -13410,16 +13305,8 @@
xmlXPathModValues(ctxt);
return (total);
case XPATH_OP_UNION:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
CHECK_ERROR0;
- ctxt->context->doc = bakd;
- ctxt->context->node = bak;
- ctxt->context->proximityPosition = pp;
- ctxt->context->contextSize = cs;
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
CHECK_ERROR0;
@@ -13456,15 +13343,6 @@
valuePush(ctxt, xmlXPathCacheNewNodeSet(ctxt->context,
ctxt->context->node));
return (total);
- case XPATH_OP_RESET:
- if (op->ch1 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
- CHECK_ERROR0;
- if (op->ch2 != -1)
- total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
- CHECK_ERROR0;
- ctxt->context->node = NULL;
- return (total);
case XPATH_OP_COLLECT:{
if (op->ch1 == -1)
return (total);
@@ -13583,24 +13461,12 @@
return (total);
}
case XPATH_OP_ARG:
- bakd = ctxt->context->doc;
- bak = ctxt->context->node;
- pp = ctxt->context->proximityPosition;
- cs = ctxt->context->contextSize;
if (op->ch1 != -1) {
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch1]);
- ctxt->context->contextSize = cs;
- ctxt->context->proximityPosition = pp;
- ctxt->context->node = bak;
- ctxt->context->doc = bakd;
CHECK_ERROR0;
}
if (op->ch2 != -1) {
total += xmlXPathCompOpEval(ctxt, &comp->steps[op->ch2]);
- ctxt->context->contextSize = cs;
- ctxt->context->proximityPosition = pp;
- ctxt->context->node = bak;
- ctxt->context->doc = bakd;
CHECK_ERROR0;
}
return (total);
@@ -13612,6 +13478,7 @@
xmlNodeSetPtr oldset;
xmlNodePtr oldnode;
xmlDocPtr oldDoc;
+ int oldcs, oldpp;
int i;
/*
@@ -13714,8 +13581,6 @@
if (ctxt->value == NULL)
return (total);
- oldnode = ctxt->context->node;
-
#ifdef LIBXML_XPTR_ENABLED
/*
* Hum are we filtering the result of an XPointer expression
@@ -13730,25 +13595,17 @@
* up a new locset.
*/
CHECK_TYPE0(XPATH_LOCATIONSET);
+
+ if ((ctxt->value->user == NULL) ||
+ (((xmlLocationSetPtr) ctxt->value->user)->locNr == 0))
+ return (total);
+
obj = valuePop(ctxt);
oldlocset = obj->user;
- ctxt->context->node = NULL;
+ oldnode = ctxt->context->node;
+ oldcs = ctxt->context->contextSize;
+ oldpp = ctxt->context->proximityPosition;
- if ((oldlocset == NULL) || (oldlocset->locNr == 0)) {
- ctxt->context->contextSize = 0;
- ctxt->context->proximityPosition = 0;
- if (op->ch2 != -1)
- total +=
- xmlXPathCompOpEval(ctxt,
- &comp->steps[op->ch2]);
- res = valuePop(ctxt);
- if (res != NULL) {
- xmlXPathReleaseObject(ctxt->context, res);
- }
- valuePush(ctxt, obj);
- CHECK_ERROR0;
- return (total);
- }
newlocset = xmlXPtrLocationSetCreate(NULL);
for (i = 0; i < oldlocset->locNr; i++) {
@@ -13768,8 +13625,8 @@
xmlXPathCompOpEval(ctxt,
&comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
- xmlXPathFreeObject(obj);
- return(0);
+ xmlXPtrFreeLocationSet(newlocset);
+ goto filter_xptr_error;
}
/*
@@ -13793,19 +13650,17 @@
res = valuePop(ctxt);
xmlXPathReleaseObject(ctxt->context, res);
}
-
- ctxt->context->node = NULL;
}
/*
* The result is used as the new evaluation locset.
*/
- xmlXPathReleaseObject(ctxt->context, obj);
- ctxt->context->node = NULL;
- ctxt->context->contextSize = -1;
- ctxt->context->proximityPosition = -1;
valuePush(ctxt, xmlXPtrWrapLocationSet(newlocset));
+filter_xptr_error:
+ xmlXPathReleaseObject(ctxt->context, obj);
ctxt->context->node = oldnode;
+ ctxt->context->contextSize = oldcs;
+ ctxt->context->proximityPosition = oldpp;
return (total);
}
#endif /* LIBXML_XPTR_ENABLED */
@@ -13816,30 +13671,15 @@
* up a new set.
*/
CHECK_TYPE0(XPATH_NODESET);
- obj = valuePop(ctxt);
- oldset = obj->nodesetval;
- oldnode = ctxt->context->node;
- oldDoc = ctxt->context->doc;
- ctxt->context->node = NULL;
-
- if ((oldset == NULL) || (oldset->nodeNr == 0)) {
- ctxt->context->contextSize = 0;
- ctxt->context->proximityPosition = 0;
-/*
- if (op->ch2 != -1)
- total +=
- xmlXPathCompOpEval(ctxt,
- &comp->steps[op->ch2]);
- CHECK_ERROR0;
- res = valuePop(ctxt);
- if (res != NULL)
- xmlXPathFreeObject(res);
-*/
- valuePush(ctxt, obj);
- ctxt->context->node = oldnode;
- CHECK_ERROR0;
- } else {
+ if ((ctxt->value->nodesetval != NULL) &&
+ (ctxt->value->nodesetval->nodeNr != 0)) {
+ obj = valuePop(ctxt);
+ oldset = obj->nodesetval;
+ oldnode = ctxt->context->node;
+ oldDoc = ctxt->context->doc;
+ oldcs = ctxt->context->contextSize;
+ oldpp = ctxt->context->proximityPosition;
tmp = NULL;
/*
* Initialize the new set.
@@ -13905,8 +13745,7 @@
&comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
xmlXPathFreeNodeSet(newset);
- xmlXPathFreeObject(obj);
- return(0);
+ goto filter_error;
}
/*
@@ -13940,23 +13779,21 @@
*/
} else
tmp = NULL;
- ctxt->context->node = NULL;
}
if (tmp != NULL)
xmlXPathReleaseObject(ctxt->context, tmp);
/*
* The result is used as the new evaluation set.
*/
- xmlXPathReleaseObject(ctxt->context, obj);
- ctxt->context->node = NULL;
- ctxt->context->contextSize = -1;
- ctxt->context->proximityPosition = -1;
- /* may want to move this past the '}' later */
- ctxt->context->doc = oldDoc;
valuePush(ctxt,
xmlXPathCacheWrapNodeSet(ctxt->context, newset));
+filter_error:
+ xmlXPathReleaseObject(ctxt->context, obj);
+ ctxt->context->node = oldnode;
+ ctxt->context->doc = oldDoc;
+ ctxt->context->contextSize = oldcs;
+ ctxt->context->proximityPosition = oldpp;
}
- ctxt->context->node = oldnode;
return (total);
}
case XPATH_OP_SORT:
@@ -13979,6 +13816,9 @@
xmlLocationSetPtr newlocset = NULL;
xmlLocationSetPtr oldlocset;
xmlNodeSetPtr oldset;
+ xmlNodePtr oldnode = ctxt->context->node;
+ int oldcs = ctxt->context->contextSize;
+ int oldpp = ctxt->context->proximityPosition;
int i, j;
if (op->ch1 != -1) {
@@ -13999,22 +13839,14 @@
* up a new locset.
*/
CHECK_TYPE0(XPATH_LOCATIONSET);
+
+ if ((ctxt->value->user == NULL) ||
+ (((xmlLocationSetPtr) ctxt->value->user)->locNr == 0))
+ return (total);
+
obj = valuePop(ctxt);
oldlocset = obj->user;
- if ((oldlocset == NULL) || (oldlocset->locNr == 0)) {
- ctxt->context->node = NULL;
- ctxt->context->contextSize = 0;
- ctxt->context->proximityPosition = 0;
- total += xmlXPathCompOpEval(ctxt,&comp->steps[op->ch2]);
- res = valuePop(ctxt);
- if (res != NULL) {
- xmlXPathReleaseObject(ctxt->context, res);
- }
- valuePush(ctxt, obj);
- CHECK_ERROR0;
- return (total);
- }
newlocset = xmlXPtrLocationSetCreate(NULL);
for (i = 0; i < oldlocset->locNr; i++) {
@@ -14034,8 +13866,8 @@
xmlXPathCompOpEval(ctxt,
&comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
- xmlXPathFreeObject(obj);
- return(0);
+ xmlXPtrFreeLocationSet(newlocset);
+ goto rangeto_error;
}
res = valuePop(ctxt);
@@ -14070,14 +13902,11 @@
res = valuePop(ctxt);
xmlXPathReleaseObject(ctxt->context, res);
}
-
- ctxt->context->node = NULL;
}
} else { /* Not a location set */
CHECK_TYPE0(XPATH_NODESET);
obj = valuePop(ctxt);
oldset = obj->nodesetval;
- ctxt->context->node = NULL;
newlocset = xmlXPtrLocationSetCreate(NULL);
@@ -14100,8 +13929,8 @@
xmlXPathCompOpEval(ctxt,
&comp->steps[op->ch2]);
if (ctxt->error != XPATH_EXPRESSION_OK) {
- xmlXPathFreeObject(obj);
- return(0);
+ xmlXPtrFreeLocationSet(newlocset);
+ goto rangeto_error;
}
res = valuePop(ctxt);
@@ -14122,8 +13951,6 @@
res = valuePop(ctxt);
xmlXPathReleaseObject(ctxt->context, res);
}
-
- ctxt->context->node = NULL;
}
}
}
@@ -14131,11 +13958,12 @@
/*
* The result is used as the new evaluation set.
*/
- xmlXPathReleaseObject(ctxt->context, obj);
- ctxt->context->node = NULL;
- ctxt->context->contextSize = -1;
- ctxt->context->proximityPosition = -1;
valuePush(ctxt, xmlXPtrWrapLocationSet(newlocset));
+rangeto_error:
+ xmlXPathReleaseObject(ctxt->context, obj);
+ ctxt->context->node = oldnode;
+ ctxt->context->contextSize = oldcs;
+ ctxt->context->proximityPosition = oldpp;
return (total);
}
#endif /* LIBXML_XPTR_ENABLED */
diff --git a/xzlib.c b/xzlib.c
index a839169..0ba88cf 100644
--- a/xzlib.c
+++ b/xzlib.c
@@ -562,6 +562,10 @@
"internal error: inflate stream corrupt");
return -1;
}
+ /*
+ * FIXME: Remapping a couple of error codes and falling through
+ * to the LZMA error handling looks fragile.
+ */
if (ret == Z_MEM_ERROR)
ret = LZMA_MEM_ERROR;
if (ret == Z_DATA_ERROR)
@@ -587,6 +591,11 @@
xz_error(state, LZMA_PROG_ERROR, "compression error");
return -1;
}
+ if ((state->how != GZIP) &&
+ (ret != LZMA_OK) && (ret != LZMA_STREAM_END)) {
+ xz_error(state, ret, "lzma error");
+ return -1;
+ }
} while (strm->avail_out && ret != LZMA_STREAM_END);
/* update available output and crc check value */