blob: 880538d4e98a218908c8d5beb7116a40c3f628bf [file] [log] [blame]
/*
* Copyright 2004 The WebRTC Project Authors. All rights reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "webrtc/base/stringencode.h"
#include <stdio.h>
#include <stdlib.h>
#include "webrtc/base/basictypes.h"
#include "webrtc/base/checks.h"
#include "webrtc/base/stringutils.h"
namespace rtc {
/////////////////////////////////////////////////////////////////////////////
// String Encoding Utilities
/////////////////////////////////////////////////////////////////////////////
size_t escape(char * buffer, size_t buflen,
const char * source, size_t srclen,
const char * illegal, char escape) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
char ch = source[srcpos++];
if ((ch == escape) || ::strchr(illegal, ch)) {
if (bufpos + 2 >= buflen)
break;
buffer[bufpos++] = escape;
}
buffer[bufpos++] = ch;
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t unescape(char * buffer, size_t buflen,
const char * source, size_t srclen,
char escape) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
char ch = source[srcpos++];
if ((ch == escape) && (srcpos < srclen)) {
ch = source[srcpos++];
}
buffer[bufpos++] = ch;
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t encode(char * buffer, size_t buflen,
const char * source, size_t srclen,
const char * illegal, char escape) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
char ch = source[srcpos++];
if ((ch != escape) && !::strchr(illegal, ch)) {
buffer[bufpos++] = ch;
} else if (bufpos + 3 >= buflen) {
break;
} else {
buffer[bufpos+0] = escape;
buffer[bufpos+1] = hex_encode((static_cast<unsigned char>(ch) >> 4) & 0xF);
buffer[bufpos+2] = hex_encode((static_cast<unsigned char>(ch) ) & 0xF);
bufpos += 3;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t decode(char * buffer, size_t buflen,
const char * source, size_t srclen,
char escape) {
if (buflen <= 0)
return 0;
unsigned char h1, h2;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
char ch = source[srcpos++];
if ((ch == escape)
&& (srcpos + 1 < srclen)
&& hex_decode(source[srcpos], &h1)
&& hex_decode(source[srcpos+1], &h2)) {
buffer[bufpos++] = (h1 << 4) | h2;
srcpos += 2;
} else {
buffer[bufpos++] = ch;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
const char* unsafe_filename_characters() {
// It might be better to have a single specification which is the union of
// all operating systems, unless one system is overly restrictive.
#if defined(WEBRTC_WIN)
return "\\/:*?\"<>|";
#else // !WEBRTC_WIN
// TODO
DCHECK(false);
return "";
#endif // !WEBRTC_WIN
}
const unsigned char URL_UNSAFE = 0x1; // 0-33 "#$%&+,/:;<=>?@[\]^`{|} 127
const unsigned char XML_UNSAFE = 0x2; // "&'<>
const unsigned char HTML_UNSAFE = 0x2; // "&'<>
// ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 6 5 7 8 9 : ; < = > ?
//@ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _
//` a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~
const unsigned char ASCII_CLASS[128] = {
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,0,3,1,1,1,3,2,0,0,0,1,1,0,0,1,0,0,0,0,0,0,0,0,0,0,1,1,3,1,3,1,
1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0,
1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,1,
};
size_t url_encode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
if (NULL == buffer)
return srclen * 3 + 1;
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
unsigned char ch = source[srcpos++];
if ((ch < 128) && (ASCII_CLASS[ch] & URL_UNSAFE)) {
if (bufpos + 3 >= buflen) {
break;
}
buffer[bufpos+0] = '%';
buffer[bufpos+1] = hex_encode((ch >> 4) & 0xF);
buffer[bufpos+2] = hex_encode((ch ) & 0xF);
bufpos += 3;
} else {
buffer[bufpos++] = ch;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t url_decode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
if (NULL == buffer)
return srclen + 1;
if (buflen <= 0)
return 0;
unsigned char h1, h2;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
unsigned char ch = source[srcpos++];
if (ch == '+') {
buffer[bufpos++] = ' ';
} else if ((ch == '%')
&& (srcpos + 1 < srclen)
&& hex_decode(source[srcpos], &h1)
&& hex_decode(source[srcpos+1], &h2))
{
buffer[bufpos++] = (h1 << 4) | h2;
srcpos += 2;
} else {
buffer[bufpos++] = ch;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t utf8_decode(const char* source, size_t srclen, unsigned long* value) {
const unsigned char* s = reinterpret_cast<const unsigned char*>(source);
if ((s[0] & 0x80) == 0x00) { // Check s[0] == 0xxxxxxx
*value = s[0];
return 1;
}
if ((srclen < 2) || ((s[1] & 0xC0) != 0x80)) { // Check s[1] != 10xxxxxx
return 0;
}
// Accumulate the trailer byte values in value16, and combine it with the
// relevant bits from s[0], once we've determined the sequence length.
unsigned long value16 = (s[1] & 0x3F);
if ((s[0] & 0xE0) == 0xC0) { // Check s[0] == 110xxxxx
*value = ((s[0] & 0x1F) << 6) | value16;
return 2;
}
if ((srclen < 3) || ((s[2] & 0xC0) != 0x80)) { // Check s[2] != 10xxxxxx
return 0;
}
value16 = (value16 << 6) | (s[2] & 0x3F);
if ((s[0] & 0xF0) == 0xE0) { // Check s[0] == 1110xxxx
*value = ((s[0] & 0x0F) << 12) | value16;
return 3;
}
if ((srclen < 4) || ((s[3] & 0xC0) != 0x80)) { // Check s[3] != 10xxxxxx
return 0;
}
value16 = (value16 << 6) | (s[3] & 0x3F);
if ((s[0] & 0xF8) == 0xF0) { // Check s[0] == 11110xxx
*value = ((s[0] & 0x07) << 18) | value16;
return 4;
}
return 0;
}
size_t utf8_encode(char* buffer, size_t buflen, unsigned long value) {
if ((value <= 0x7F) && (buflen >= 1)) {
buffer[0] = static_cast<unsigned char>(value);
return 1;
}
if ((value <= 0x7FF) && (buflen >= 2)) {
buffer[0] = 0xC0 | static_cast<unsigned char>(value >> 6);
buffer[1] = 0x80 | static_cast<unsigned char>(value & 0x3F);
return 2;
}
if ((value <= 0xFFFF) && (buflen >= 3)) {
buffer[0] = 0xE0 | static_cast<unsigned char>(value >> 12);
buffer[1] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F);
buffer[2] = 0x80 | static_cast<unsigned char>(value & 0x3F);
return 3;
}
if ((value <= 0x1FFFFF) && (buflen >= 4)) {
buffer[0] = 0xF0 | static_cast<unsigned char>(value >> 18);
buffer[1] = 0x80 | static_cast<unsigned char>((value >> 12) & 0x3F);
buffer[2] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F);
buffer[3] = 0x80 | static_cast<unsigned char>(value & 0x3F);
return 4;
}
return 0;
}
size_t html_encode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
unsigned char ch = source[srcpos];
if (ch < 128) {
srcpos += 1;
if (ASCII_CLASS[ch] & HTML_UNSAFE) {
const char * escseq = 0;
size_t esclen = 0;
switch (ch) {
case '<': escseq = "&lt;"; esclen = 4; break;
case '>': escseq = "&gt;"; esclen = 4; break;
case '\'': escseq = "&#39;"; esclen = 5; break;
case '\"': escseq = "&quot;"; esclen = 6; break;
case '&': escseq = "&amp;"; esclen = 5; break;
default: DCHECK(false);
}
if (bufpos + esclen >= buflen) {
break;
}
memcpy(buffer + bufpos, escseq, esclen);
bufpos += esclen;
} else {
buffer[bufpos++] = ch;
}
} else {
// Largest value is 0x1FFFFF => &#2097151; (10 characters)
const size_t kEscseqSize = 11;
char escseq[kEscseqSize];
unsigned long val;
if (size_t vallen = utf8_decode(&source[srcpos], srclen - srcpos, &val)) {
srcpos += vallen;
} else {
// Not a valid utf8 sequence, just use the raw character.
val = static_cast<unsigned char>(source[srcpos++]);
}
size_t esclen = sprintfn(escseq, kEscseqSize, "&#%lu;", val);
if (bufpos + esclen >= buflen) {
break;
}
memcpy(buffer + bufpos, escseq, esclen);
bufpos += esclen;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t html_decode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
DCHECK(buffer); // TODO: estimate output size
return xml_decode(buffer, buflen, source, srclen);
}
size_t xml_encode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
unsigned char ch = source[srcpos++];
if ((ch < 128) && (ASCII_CLASS[ch] & XML_UNSAFE)) {
const char * escseq = 0;
size_t esclen = 0;
switch (ch) {
case '<': escseq = "&lt;"; esclen = 4; break;
case '>': escseq = "&gt;"; esclen = 4; break;
case '\'': escseq = "&apos;"; esclen = 6; break;
case '\"': escseq = "&quot;"; esclen = 6; break;
case '&': escseq = "&amp;"; esclen = 5; break;
default: DCHECK(false);
}
if (bufpos + esclen >= buflen) {
break;
}
memcpy(buffer + bufpos, escseq, esclen);
bufpos += esclen;
} else {
buffer[bufpos++] = ch;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
size_t xml_decode(char * buffer, size_t buflen,
const char * source, size_t srclen) {
DCHECK(buffer); // TODO: estimate output size
if (buflen <= 0)
return 0;
size_t srcpos = 0, bufpos = 0;
while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
unsigned char ch = source[srcpos++];
if (ch != '&') {
buffer[bufpos++] = ch;
} else if ((srcpos + 2 < srclen)
&& (memcmp(source + srcpos, "lt;", 3) == 0)) {
buffer[bufpos++] = '<';
srcpos += 3;
} else if ((srcpos + 2 < srclen)
&& (memcmp(source + srcpos, "gt;", 3) == 0)) {
buffer[bufpos++] = '>';
srcpos += 3;
} else if ((srcpos + 4 < srclen)
&& (memcmp(source + srcpos, "apos;", 5) == 0)) {
buffer[bufpos++] = '\'';
srcpos += 5;
} else if ((srcpos + 4 < srclen)
&& (memcmp(source + srcpos, "quot;", 5) == 0)) {
buffer[bufpos++] = '\"';
srcpos += 5;
} else if ((srcpos + 3 < srclen)
&& (memcmp(source + srcpos, "amp;", 4) == 0)) {
buffer[bufpos++] = '&';
srcpos += 4;
} else if ((srcpos < srclen) && (source[srcpos] == '#')) {
int int_base = 10;
if ((srcpos + 1 < srclen) && (source[srcpos+1] == 'x')) {
int_base = 16;
srcpos += 1;
}
char * ptr;
// TODO: Fix hack (ptr may go past end of data)
unsigned long val = strtoul(source + srcpos + 1, &ptr, int_base);
if ((static_cast<size_t>(ptr - source) < srclen) && (*ptr == ';')) {
srcpos = ptr - source + 1;
} else {
// Not a valid escape sequence.
break;
}
if (size_t esclen = utf8_encode(buffer + bufpos, buflen - bufpos, val)) {
bufpos += esclen;
} else {
// Not enough room to encode the character, or illegal character
break;
}
} else {
// Unrecognized escape sequence.
break;
}
}
buffer[bufpos] = '\0';
return bufpos;
}
static const char HEX[] = "0123456789abcdef";
char hex_encode(unsigned char val) {
DCHECK_LT(val, 16);
return (val < 16) ? HEX[val] : '!';
}
bool hex_decode(char ch, unsigned char* val) {
if ((ch >= '0') && (ch <= '9')) {
*val = ch - '0';
} else if ((ch >= 'A') && (ch <= 'Z')) {
*val = (ch - 'A') + 10;
} else if ((ch >= 'a') && (ch <= 'z')) {
*val = (ch - 'a') + 10;
} else {
return false;
}
return true;
}
size_t hex_encode(char* buffer, size_t buflen,
const char* csource, size_t srclen) {
return hex_encode_with_delimiter(buffer, buflen, csource, srclen, 0);
}
size_t hex_encode_with_delimiter(char* buffer, size_t buflen,
const char* csource, size_t srclen,
char delimiter) {
DCHECK(buffer); // TODO: estimate output size
if (buflen == 0)
return 0;
// Init and check bounds.
const unsigned char* bsource =
reinterpret_cast<const unsigned char*>(csource);
size_t srcpos = 0, bufpos = 0;
size_t needed = delimiter ? (srclen * 3) : (srclen * 2 + 1);
if (buflen < needed)
return 0;
while (srcpos < srclen) {
unsigned char ch = bsource[srcpos++];
buffer[bufpos ] = hex_encode((ch >> 4) & 0xF);
buffer[bufpos+1] = hex_encode((ch ) & 0xF);
bufpos += 2;
// Don't write a delimiter after the last byte.
if (delimiter && (srcpos < srclen)) {
buffer[bufpos] = delimiter;
++bufpos;
}
}
// Null terminate.
buffer[bufpos] = '\0';
return bufpos;
}
std::string hex_encode(const std::string& str) {
return hex_encode(str.c_str(), str.size());
}
std::string hex_encode(const char* source, size_t srclen) {
return hex_encode_with_delimiter(source, srclen, 0);
}
std::string hex_encode_with_delimiter(const char* source, size_t srclen,
char delimiter) {
const size_t kBufferSize = srclen * 3;
char* buffer = STACK_ARRAY(char, kBufferSize);
size_t length = hex_encode_with_delimiter(buffer, kBufferSize,
source, srclen, delimiter);
DCHECK(srclen == 0 || length > 0);
return std::string(buffer, length);
}
size_t hex_decode(char * cbuffer, size_t buflen,
const char * source, size_t srclen) {
return hex_decode_with_delimiter(cbuffer, buflen, source, srclen, 0);
}
size_t hex_decode_with_delimiter(char* cbuffer, size_t buflen,
const char* source, size_t srclen,
char delimiter) {
DCHECK(cbuffer); // TODO: estimate output size
if (buflen == 0)
return 0;
// Init and bounds check.
unsigned char* bbuffer = reinterpret_cast<unsigned char*>(cbuffer);
size_t srcpos = 0, bufpos = 0;
size_t needed = (delimiter) ? (srclen + 1) / 3 : srclen / 2;
if (buflen < needed)
return 0;
while (srcpos < srclen) {
if ((srclen - srcpos) < 2) {
// This means we have an odd number of bytes.
return 0;
}
unsigned char h1, h2;
if (!hex_decode(source[srcpos], &h1) ||
!hex_decode(source[srcpos + 1], &h2))
return 0;
bbuffer[bufpos++] = (h1 << 4) | h2;
srcpos += 2;
// Remove the delimiter if needed.
if (delimiter && (srclen - srcpos) > 1) {
if (source[srcpos] != delimiter)
return 0;
++srcpos;
}
}
return bufpos;
}
size_t hex_decode(char* buffer, size_t buflen, const std::string& source) {
return hex_decode_with_delimiter(buffer, buflen, source, 0);
}
size_t hex_decode_with_delimiter(char* buffer, size_t buflen,
const std::string& source, char delimiter) {
return hex_decode_with_delimiter(buffer, buflen,
source.c_str(), source.length(), delimiter);
}
size_t transform(std::string& value, size_t maxlen, const std::string& source,
Transform t) {
char* buffer = STACK_ARRAY(char, maxlen + 1);
size_t length = t(buffer, maxlen + 1, source.data(), source.length());
value.assign(buffer, length);
return length;
}
std::string s_transform(const std::string& source, Transform t) {
// Ask transformation function to approximate the destination size (returns upper bound)
size_t maxlen = t(NULL, 0, source.data(), source.length());
char * buffer = STACK_ARRAY(char, maxlen);
size_t len = t(buffer, maxlen, source.data(), source.length());
std::string result(buffer, len);
return result;
}
size_t tokenize(const std::string& source, char delimiter,
std::vector<std::string>* fields) {
DCHECK(fields);
fields->clear();
size_t last = 0;
for (size_t i = 0; i < source.length(); ++i) {
if (source[i] == delimiter) {
if (i != last) {
fields->push_back(source.substr(last, i - last));
}
last = i + 1;
}
}
if (last != source.length()) {
fields->push_back(source.substr(last, source.length() - last));
}
return fields->size();
}
size_t tokenize_append(const std::string& source, char delimiter,
std::vector<std::string>* fields) {
if (!fields) return 0;
std::vector<std::string> new_fields;
tokenize(source, delimiter, &new_fields);
fields->insert(fields->end(), new_fields.begin(), new_fields.end());
return fields->size();
}
size_t tokenize(const std::string& source, char delimiter, char start_mark,
char end_mark, std::vector<std::string>* fields) {
if (!fields) return 0;
fields->clear();
std::string remain_source = source;
while (!remain_source.empty()) {
size_t start_pos = remain_source.find(start_mark);
if (std::string::npos == start_pos) break;
std::string pre_mark;
if (start_pos > 0) {
pre_mark = remain_source.substr(0, start_pos - 1);
}
++start_pos;
size_t end_pos = remain_source.find(end_mark, start_pos);
if (std::string::npos == end_pos) break;
// We have found the matching marks. First tokenize the pre-mask. Then add
// the marked part as a single field. Finally, loop back for the post-mark.
tokenize_append(pre_mark, delimiter, fields);
fields->push_back(remain_source.substr(start_pos, end_pos - start_pos));
remain_source = remain_source.substr(end_pos + 1);
}
return tokenize_append(remain_source, delimiter, fields);
}
bool tokenize_first(const std::string& source, const char delimiter,
std::string* token, std::string* rest) {
// Find the first delimiter
size_t left_pos = source.find(delimiter);
if (left_pos == std::string::npos) {
return false;
}
// Look for additional occurrances of delimiter.
size_t right_pos = left_pos + 1;
while(source[right_pos] == delimiter) {
right_pos++;
}
*token = source.substr(0, left_pos);
*rest = source.substr(right_pos);
return true;
}
size_t split(const std::string& source, char delimiter,
std::vector<std::string>* fields) {
DCHECK(fields);
fields->clear();
size_t last = 0;
for (size_t i = 0; i < source.length(); ++i) {
if (source[i] == delimiter) {
fields->push_back(source.substr(last, i - last));
last = i + 1;
}
}
fields->push_back(source.substr(last, source.length() - last));
return fields->size();
}
char make_char_safe_for_filename(char c) {
if (c < 32)
return '_';
switch (c) {
case '<':
case '>':
case ':':
case '"':
case '/':
case '\\':
case '|':
case '*':
case '?':
return '_';
default:
return c;
}
}
/*
void sprintf(std::string& value, size_t maxlen, const char * format, ...) {
char * buffer = STACK_ARRAY(char, maxlen + 1);
va_list args;
va_start(args, format);
value.assign(buffer, vsprintfn(buffer, maxlen + 1, format, args));
va_end(args);
}
*/
/////////////////////////////////////////////////////////////////////////////
} // namespace rtc