#!/usr/bin/env perl
#
#                     The LLVM Compiler Infrastructure
#
# This file is distributed under the University of Illinois Open Source
# License. See LICENSE.TXT for details.
#
##===----------------------------------------------------------------------===##
#
# A script designed to wrap a build so that all calls to gcc are intercepted
# and piped to the static analyzer.
#
##===----------------------------------------------------------------------===##

use strict;
use warnings;
use FindBin qw($RealBin);
use Digest::MD5;
use File::Basename;
use Term::ANSIColor;
use Term::ANSIColor qw(:constants);
use Cwd qw/ getcwd abs_path /;
use Sys::Hostname;

my $Verbose = 0;       # Verbose output from this script.
my $Prog = "scan-build";
my $BuildName;
my $BuildDate;

my $TERM = $ENV{'TERM'};
my $UseColor = (defined $TERM and $TERM =~ 'xterm-.*color' and -t STDOUT
                and defined $ENV{'SCAN_BUILD_COLOR'});

my $UserName = HtmlEscape(getpwuid($<) || 'unknown');
my $HostName = HtmlEscape(hostname() || 'unknown');
my $CurrentDir = HtmlEscape(getcwd());
my $CurrentDirSuffix = basename($CurrentDir);

my @PluginsToLoad;
my $CmdArgs;

my $HtmlTitle;

my $Date = localtime();

##----------------------------------------------------------------------------##
# Diagnostics
##----------------------------------------------------------------------------##

sub Diag {
  if ($UseColor) {
    print BOLD, MAGENTA "$Prog: @_";
    print RESET;
  }
  else {
    print "$Prog: @_";
  }  
}

sub DiagCrashes {
  my $Dir = shift;
  Diag ("The analyzer encountered problems on some source files.\n");
  Diag ("Preprocessed versions of these sources were deposited in '$Dir/failures'.\n");
  Diag ("Please consider submitting a bug report using these files:\n");
  Diag ("  http://clang-analyzer.llvm.org/filing_bugs.html\n")
}

sub DieDiag {
  if ($UseColor) {
    print BOLD, RED "$Prog: ";
    print RESET, RED @_;
    print RESET;
  }
  else {
    print "$Prog: ", @_;
  }
  exit(0);
}

##----------------------------------------------------------------------------##
# Print default checker names
##----------------------------------------------------------------------------##

if (grep /^--help-checkers$/, @ARGV) {
    my @options = qx($0 -h);
    foreach (@options) {
	next unless /^ \+/;
	s/^\s*//;
	my ($sign, $name, @text) = split ' ', $_;
	print $name, $/ if $sign eq '+';
    }
    exit 1;
}

##----------------------------------------------------------------------------##
# Some initial preprocessing of Clang options.
##----------------------------------------------------------------------------##

# Find 'clang'
my $ClangSB = Cwd::realpath("$RealBin/bin/clang");
if (!defined $ClangSB || ! -x $ClangSB) {
  $ClangSB = Cwd::realpath("$RealBin/clang");
}
my $Clang;
if (!defined $ClangSB || ! -x $ClangSB) {
  # Default to looking for 'clang' in the path.
  $Clang = `which clang`;
  chomp $Clang;
  if ($Clang eq "") {
    DieDiag("No 'clang' executable found in path.\n");
  }
}
else {
  $Clang = $ClangSB;
}
my $ClangCXX = $Clang . "++";
my $ClangVersion = HtmlEscape(`$Clang --version`);

##----------------------------------------------------------------------------##
# GetHTMLRunDir - Construct an HTML directory name for the current sub-run.
##----------------------------------------------------------------------------##

sub GetHTMLRunDir {  
  die "Not enough arguments." if (@_ == 0);  
  my $Dir = shift @_;    
  my $TmpMode = 0;
  if (!defined $Dir) {
    if (`uname` =~ /Darwin/) {
      $Dir = $ENV{'TMPDIR'};
      if (!defined $Dir) { $Dir = "/tmp"; }
    }
    else {
      $Dir = "/tmp";
    }    
    $TmpMode = 1;
  }
  
  # Chop off any trailing '/' characters.
  while ($Dir =~ /\/$/) { chop $Dir; }

  # Get current date and time.
  my @CurrentTime = localtime();  
  my $year  = $CurrentTime[5] + 1900;
  my $day   = $CurrentTime[3];
  my $month = $CurrentTime[4] + 1;
  my $DateString = sprintf("%d-%02d-%02d", $year, $month, $day);
  
  # Determine the run number.  
  my $RunNumber;
  
  if (-d $Dir) {    
    if (! -r $Dir) {
      DieDiag("directory '$Dir' exists but is not readable.\n");
    }    
    # Iterate over all files in the specified directory.    
    my $max = 0;    
    opendir(DIR, $Dir);
    my @FILES = grep { -d "$Dir/$_" } readdir(DIR);
    closedir(DIR);

    foreach my $f (@FILES) {
      # Strip the prefix '$Prog-' if we are dumping files to /tmp.
      if ($TmpMode) {
        next if (!($f =~ /^$Prog-(.+)/));
        $f = $1;
      }

      my @x = split/-/, $f;
      next if (scalar(@x) != 4);
      next if ($x[0] != $year);
      next if ($x[1] != $month);
      next if ($x[2] != $day);
      
      if ($x[3] > $max) {
        $max = $x[3];
      }      
    }
    
    $RunNumber = $max + 1;
  }
  else {
    
    if (-x $Dir) {
      DieDiag("'$Dir' exists but is not a directory.\n");
    }

    if ($TmpMode) {
      DieDiag("The directory '/tmp' does not exist or cannot be accessed.\n");
    }

    # $Dir does not exist.  It will be automatically created by the 
    # clang driver.  Set the run number to 1.  

    $RunNumber = 1;
  }
  
  die "RunNumber must be defined!" if (!defined $RunNumber);
  
  # Append the run number.
  my $NewDir;
  if ($TmpMode) {
    $NewDir = "$Dir/$Prog-$DateString-$RunNumber";
  }
  else {
    $NewDir = "$Dir/$DateString-$RunNumber";
  }
  system 'mkdir','-p',$NewDir;
  return $NewDir;
}

sub SetHtmlEnv {
  
  die "Wrong number of arguments." if (scalar(@_) != 2);
  
  my $Args = shift;
  my $Dir = shift;
  
  die "No build command." if (scalar(@$Args) == 0);
  
  my $Cmd = $$Args[0];

  if ($Cmd =~ /configure/ || $Cmd =~ /autogen/) {
    return;
  }
  
  if ($Verbose) {
    Diag("Emitting reports for this run to '$Dir'.\n");
  }
  
  $ENV{'CCC_ANALYZER_HTML'} = $Dir;
}

##----------------------------------------------------------------------------##
# ComputeDigest - Compute a digest of the specified file.
##----------------------------------------------------------------------------##

sub ComputeDigest {
  my $FName = shift;
  DieDiag("Cannot read $FName to compute Digest.\n") if (! -r $FName);  
  
  # Use Digest::MD5.  We don't have to be cryptographically secure.  We're
  # just looking for duplicate files that come from a non-malicious source.
  # We use Digest::MD5 because it is a standard Perl module that should
  # come bundled on most systems.  
  open(FILE, $FName) or DieDiag("Cannot open $FName when computing Digest.\n");
  binmode FILE;
  my $Result = Digest::MD5->new->addfile(*FILE)->hexdigest;
  close(FILE);
  
  # Return the digest.  
  return $Result;
}

##----------------------------------------------------------------------------##
#  UpdatePrefix - Compute the common prefix of files.
##----------------------------------------------------------------------------##

my $Prefix;

sub UpdatePrefix {
  my $x = shift;
  my $y = basename($x);
  $x =~ s/\Q$y\E$//;

  if (!defined $Prefix) {
    $Prefix = $x;
    return;
  }
  
  chop $Prefix while (!($x =~ /^\Q$Prefix/));
}

sub GetPrefix {
  return $Prefix;
}

##----------------------------------------------------------------------------##
#  UpdateInFilePath - Update the path in the report file.
##----------------------------------------------------------------------------##

sub UpdateInFilePath {
  my $fname = shift;
  my $regex = shift;
  my $newtext = shift;

  open (RIN, $fname) or die "cannot open $fname";
  open (ROUT, ">", "$fname.tmp") or die "cannot open $fname.tmp";

  while (<RIN>) {
    s/$regex/$newtext/;
    print ROUT $_;
  }

  close (ROUT);
  close (RIN);
  system("mv", "$fname.tmp", $fname);
}

##----------------------------------------------------------------------------##
# AddStatLine - Decode and insert a statistics line into the database.
##----------------------------------------------------------------------------##

sub AddStatLine {
  my $Line  = shift;
  my $Stats = shift;

  print $Line . "\n";

  my $Regex = qr/(.*?)\ :\ (.*?)\ ->\ Total\ CFGBlocks:\ (\d+)\ \|\ Unreachable
      \ CFGBlocks:\ (\d+)\ \|\ Exhausted\ Block:\ (yes|no)\ \|\ Empty\ WorkList:
      \ (yes|no)/x;

  if ($Line !~ $Regex) {
    return;
  }

  # Create a hash of the interesting fields
  my $Row = {
    Filename    => $1,
    Function    => $2,
    Total       => $3,
    Unreachable => $4,
    Aborted     => $5,
    Empty       => $6
  };

  # Add them to the stats array
  push @$Stats, $Row;
}

##----------------------------------------------------------------------------##
# ScanFile - Scan a report file for various identifying attributes.
##----------------------------------------------------------------------------##

# Sometimes a source file is scanned more than once, and thus produces
# multiple error reports.  We use a cache to solve this problem.

my %AlreadyScanned;

sub ScanFile {
  
  my $Index = shift;
  my $Dir = shift;
  my $FName = shift;
  my $Stats = shift;
  
  # Compute a digest for the report file.  Determine if we have already
  # scanned a file that looks just like it.
  
  my $digest = ComputeDigest("$Dir/$FName");

  if (defined $AlreadyScanned{$digest}) {
    # Redundant file.  Remove it.
    system ("rm", "-f", "$Dir/$FName");
    return;
  }
  
  $AlreadyScanned{$digest} = 1;
  
  # At this point the report file is not world readable.  Make it happen.
  system ("chmod", "644", "$Dir/$FName");
  
  # Scan the report file for tags.
  open(IN, "$Dir/$FName") or DieDiag("Cannot open '$Dir/$FName'\n");

  my $BugType        = "";
  my $BugFile        = "";
  my $BugCategory    = "";
  my $BugDescription = "";
  my $BugPathLength  = 1;
  my $BugLine        = 0;

  while (<IN>) {
    last if (/<!-- BUGMETAEND -->/);

    if (/<!-- BUGTYPE (.*) -->$/) {
      $BugType = $1;
    }
    elsif (/<!-- BUGFILE (.*) -->$/) {
      $BugFile = abs_path($1);
      UpdatePrefix($BugFile);
    }
    elsif (/<!-- BUGPATHLENGTH (.*) -->$/) {
      $BugPathLength = $1;
    }
    elsif (/<!-- BUGLINE (.*) -->$/) {
      $BugLine = $1;    
    }
    elsif (/<!-- BUGCATEGORY (.*) -->$/) {
      $BugCategory = $1;
    }
    elsif (/<!-- BUGDESC (.*) -->$/) {
      $BugDescription = $1;
    }
  }

  close(IN);
  
  if (!defined $BugCategory) {
    $BugCategory = "Other";
  }

  # Don't add internal statistics to the bug reports
  if ($BugCategory =~ /statistics/i) {
    AddStatLine($BugDescription, $Stats);
    return;
  }
  
  push @$Index,[ $FName, $BugCategory, $BugType, $BugFile, $BugLine,
                 $BugPathLength ];
}

##----------------------------------------------------------------------------##
# CopyFiles - Copy resource files to target directory.
##----------------------------------------------------------------------------##

sub CopyFiles {

  my $Dir = shift;

  my $JS = Cwd::realpath("$RealBin/sorttable.js");
  
  DieDiag("Cannot find 'sorttable.js'.\n")
    if (! -r $JS);  

  system ("cp", $JS, "$Dir");

  DieDiag("Could not copy 'sorttable.js' to '$Dir'.\n")
    if (! -r "$Dir/sorttable.js");
    
  my $CSS = Cwd::realpath("$RealBin/scanview.css");
  
  DieDiag("Cannot find 'scanview.css'.\n")
    if (! -r $CSS);

  system ("cp", $CSS, "$Dir");

  DieDiag("Could not copy 'scanview.css' to '$Dir'.\n")
    if (! -r $CSS);
}

##----------------------------------------------------------------------------##
# CalcStats - Calculates visitation statistics and returns the string.
##----------------------------------------------------------------------------##

sub CalcStats {
  my $Stats = shift;

  my $TotalBlocks = 0;
  my $UnreachedBlocks = 0;
  my $TotalFunctions = scalar(@$Stats);
  my $BlockAborted = 0;
  my $WorkListAborted = 0;
  my $Aborted = 0;

  # Calculate the unique files
  my $FilesHash = {};

  foreach my $Row (@$Stats) {
    $FilesHash->{$Row->{Filename}} = 1;
    $TotalBlocks += $Row->{Total};
    $UnreachedBlocks += $Row->{Unreachable};
    $BlockAborted++ if $Row->{Aborted} eq 'yes';
    $WorkListAborted++ if $Row->{Empty} eq 'no';
    $Aborted++ if $Row->{Aborted} eq 'yes' || $Row->{Empty} eq 'no';
  }

  my $TotalFiles = scalar(keys(%$FilesHash));

  # Calculations
  my $PercentAborted = sprintf("%.2f", $Aborted / $TotalFunctions * 100);
  my $PercentBlockAborted = sprintf("%.2f", $BlockAborted / $TotalFunctions
      * 100);
  my $PercentWorkListAborted = sprintf("%.2f", $WorkListAborted /
      $TotalFunctions * 100);
  my $PercentBlocksUnreached = sprintf("%.2f", $UnreachedBlocks / $TotalBlocks
      * 100);

  my $StatsString = "Analyzed $TotalBlocks blocks in $TotalFunctions functions"
    . " in $TotalFiles files\n"
    . "$Aborted functions aborted early ($PercentAborted%)\n"
    . "$BlockAborted had aborted blocks ($PercentBlockAborted%)\n"
    . "$WorkListAborted had unfinished worklists ($PercentWorkListAborted%)\n"
    . "$UnreachedBlocks blocks were never reached ($PercentBlocksUnreached%)\n";

  return $StatsString;
}

##----------------------------------------------------------------------------##
# Postprocess - Postprocess the results of an analysis scan.
##----------------------------------------------------------------------------##

sub Postprocess {
  
  my $Dir           = shift;
  my $BaseDir       = shift;
  my $AnalyzerStats = shift;
  
  die "No directory specified." if (!defined $Dir);
  
  if (! -d $Dir) {
    Diag("No bugs found.\n");
    return 0;
  }
  
  opendir(DIR, $Dir);
  my @files = grep { /^report-.*\.html$/ } readdir(DIR);
  closedir(DIR);

  if (scalar(@files) == 0 and ! -e "$Dir/failures") {
    Diag("Removing directory '$Dir' because it contains no reports.\n");
    system ("rm", "-fR", $Dir);
    return 0;
  }
  
  # Scan each report file and build an index.  
  my @Index;
  my @Stats;
  foreach my $file (@files) { ScanFile(\@Index, $Dir, $file, \@Stats); }
  
  # Scan the failures directory and use the information in the .info files
  # to update the common prefix directory.
  my @failures;
  my @attributes_ignored;
  if (-d "$Dir/failures") {
    opendir(DIR, "$Dir/failures");
    @failures = grep { /[.]info.txt$/ && !/attribute_ignored/; } readdir(DIR);
    closedir(DIR);
    opendir(DIR, "$Dir/failures");        
    @attributes_ignored = grep { /^attribute_ignored/; } readdir(DIR);
    closedir(DIR);
    foreach my $file (@failures) {
      open IN, "$Dir/failures/$file" or DieDiag("cannot open $file\n");
      my $Path = <IN>;
      if (defined $Path) { UpdatePrefix($Path); }
      close IN;
    }    
  }
  
  # Generate an index.html file.  
  my $FName = "$Dir/index.html";  
  open(OUT, ">", $FName) or DieDiag("Cannot create file '$FName'\n");
  
  # Print out the header.
  
print OUT <<ENDTEXT;
<html>
<head>
<title>${HtmlTitle}</title>
<link type="text/css" rel="stylesheet" href="scanview.css"/>
<script src="sorttable.js"></script>
<script language='javascript' type="text/javascript">
function SetDisplay(RowClass, DisplayVal)
{
  var Rows = document.getElementsByTagName("tr");
  for ( var i = 0 ; i < Rows.length; ++i ) {
    if (Rows[i].className == RowClass) {
      Rows[i].style.display = DisplayVal;
    }
  }
}

function CopyCheckedStateToCheckButtons(SummaryCheckButton) {
  var Inputs = document.getElementsByTagName("input");
  for ( var i = 0 ; i < Inputs.length; ++i ) {
    if (Inputs[i].type == "checkbox") {
      if(Inputs[i] != SummaryCheckButton) {
        Inputs[i].checked = SummaryCheckButton.checked;
        Inputs[i].onclick();
	  }
    }
  }
}

function returnObjById( id ) {
    if (document.getElementById) 
        var returnVar = document.getElementById(id);
    else if (document.all)
        var returnVar = document.all[id];
    else if (document.layers) 
        var returnVar = document.layers[id];
    return returnVar; 
}

var NumUnchecked = 0;

function ToggleDisplay(CheckButton, ClassName) {
  if (CheckButton.checked) {
    SetDisplay(ClassName, "");
    if (--NumUnchecked == 0) {
      returnObjById("AllBugsCheck").checked = true;
    }
  }
  else {
    SetDisplay(ClassName, "none");
    NumUnchecked++;
    returnObjById("AllBugsCheck").checked = false;
  }
}
</script>
<!-- SUMMARYENDHEAD -->
</head>
<body>
<h1>${HtmlTitle}</h1>

<table>
<tr><th>User:</th><td>${UserName}\@${HostName}</td></tr>
<tr><th>Working Directory:</th><td>${CurrentDir}</td></tr>
<tr><th>Command Line:</th><td>${CmdArgs}</td></tr>
<tr><th>Clang Version:</th><td>${ClangVersion}</td></tr>
<tr><th>Date:</th><td>${Date}</td></tr>
ENDTEXT

print OUT "<tr><th>Version:</th><td>${BuildName} (${BuildDate})</td></tr>\n"
  if (defined($BuildName) && defined($BuildDate));

print OUT <<ENDTEXT;
</table>
ENDTEXT

  if (scalar(@files)) {
    # Print out the summary table.
    my %Totals;

    for my $row ( @Index ) {
      my $bug_type = ($row->[2]);
      my $bug_category = ($row->[1]);
      my $key = "$bug_category:$bug_type";

      if (!defined $Totals{$key}) { $Totals{$key} = [1,$bug_category,$bug_type]; }
      else { $Totals{$key}->[0]++; }
    }

    print OUT "<h2>Bug Summary</h2>";

    if (defined $BuildName) {
      print OUT "\n<p>Results in this analysis run are based on analyzer build <b>$BuildName</b>.</p>\n"
    }
  
  my $TotalBugs = scalar(@Index);
print OUT <<ENDTEXT;
<table>
<thead><tr><td>Bug Type</td><td>Quantity</td><td class="sorttable_nosort">Display?</td></tr></thead>
<tr style="font-weight:bold"><td class="SUMM_DESC">All Bugs</td><td class="Q">$TotalBugs</td><td><center><input type="checkbox" id="AllBugsCheck" onClick="CopyCheckedStateToCheckButtons(this);" checked/></center></td></tr>
ENDTEXT
  
    my $last_category;

    for my $key (
      sort {
        my $x = $Totals{$a};
        my $y = $Totals{$b};
        my $res = $x->[1] cmp $y->[1];
        $res = $x->[2] cmp $y->[2] if ($res == 0);
        $res
      } keys %Totals ) 
    {
      my $val = $Totals{$key};
      my $category = $val->[1];
      if (!defined $last_category or $last_category ne $category) {
        $last_category = $category;
        print OUT "<tr><th>$category</th><th colspan=2></th></tr>\n";
      }      
      my $x = lc $key;
      $x =~ s/[ ,'":\/()]+/_/g;
      print OUT "<tr><td class=\"SUMM_DESC\">";
      print OUT $val->[2];
      print OUT "</td><td class=\"Q\">";
      print OUT $val->[0];
      print OUT "</td><td><center><input type=\"checkbox\" onClick=\"ToggleDisplay(this,'bt_$x');\" checked/></center></td></tr>\n";
    }

  # Print out the table of errors.

print OUT <<ENDTEXT;
</table>
<h2>Reports</h2>

<table class="sortable" style="table-layout:automatic">
<thead><tr>
  <td>Bug Group</td>
  <td class="sorttable_sorted">Bug Type<span id="sorttable_sortfwdind">&nbsp;&#x25BE;</span></td>
  <td>File</td>
  <td class="Q">Line</td>
  <td class="Q">Path Length</td>
  <td class="sorttable_nosort"></td>
  <!-- REPORTBUGCOL -->
</tr></thead>
<tbody>
ENDTEXT

    my $prefix = GetPrefix();
    my $regex;
    my $InFileRegex;
    my $InFilePrefix = "File:</td><td>";
  
    if (defined $prefix) { 
      $regex = qr/^\Q$prefix\E/is;    
      $InFileRegex = qr/\Q$InFilePrefix$prefix\E/is;
    }    

    for my $row ( sort { $a->[2] cmp $b->[2] } @Index ) {
      my $x = "$row->[1]:$row->[2]";
      $x = lc $x;
      $x =~ s/[ ,'":\/()]+/_/g;
    
      my $ReportFile = $row->[0];
          
      print OUT "<tr class=\"bt_$x\">";
      print OUT "<td class=\"DESC\">";
      print OUT $row->[1];
      print OUT "</td>";
      print OUT "<td class=\"DESC\">";
      print OUT $row->[2];
      print OUT "</td>";
      
      # Update the file prefix.      
      my $fname = $row->[3];

      if (defined $regex) {
        $fname =~ s/$regex//;
        UpdateInFilePath("$Dir/$ReportFile", $InFileRegex, $InFilePrefix)
      }
      
      print OUT "<td>";      
      my @fname = split /\//,$fname;
      if ($#fname > 0) {
        while ($#fname >= 0) {
          my $x = shift @fname;
          print OUT $x;
          if ($#fname >= 0) {
            print OUT "<span class=\"W\"> </span>/";
          }
        }
      }
      else {
        print OUT $fname;
      }      
      print OUT "</td>";
      
      # Print out the quantities.
      for my $j ( 4 .. 5 ) {
        print OUT "<td class=\"Q\">$row->[$j]</td>";        
      }
      
      # Print the rest of the columns.
      for (my $j = 6; $j <= $#{$row}; ++$j) {
        print OUT "<td>$row->[$j]</td>"
      }

      # Emit the "View" link.
      print OUT "<td><a href=\"$ReportFile#EndPath\">View Report</a></td>";
        
      # Emit REPORTBUG markers.
      print OUT "\n<!-- REPORTBUG id=\"$ReportFile\" -->\n";
        
      # End the row.
      print OUT "</tr>\n";
    }
  
    print OUT "</tbody>\n</table>\n\n";
  }

  if (scalar (@failures) || scalar(@attributes_ignored)) {
    print OUT "<h2>Analyzer Failures</h2>\n";
    
    if (scalar @attributes_ignored) {
      print OUT "The analyzer's parser ignored the following attributes:<p>\n";
      print OUT "<table>\n";
      print OUT "<thead><tr><td>Attribute</td><td>Source File</td><td>Preprocessed File</td><td>STDERR Output</td></tr></thead>\n";
      foreach my $file (sort @attributes_ignored) {
        die "cannot demangle attribute name\n" if (! ($file =~ /^attribute_ignored_(.+).txt/));
        my $attribute = $1;
        # Open the attribute file to get the first file that failed.
        next if (!open (ATTR, "$Dir/failures/$file"));
        my $ppfile = <ATTR>;
        chomp $ppfile;
        close ATTR;
        next if (! -e "$Dir/failures/$ppfile");
        # Open the info file and get the name of the source file.
        open (INFO, "$Dir/failures/$ppfile.info.txt") or
          die "Cannot open $Dir/failures/$ppfile.info.txt\n";
        my $srcfile = <INFO>;
        chomp $srcfile;
        close (INFO);
        # Print the information in the table.
        my $prefix = GetPrefix();
        if (defined $prefix) { $srcfile =~ s/^\Q$prefix//; }
        print OUT "<tr><td>$attribute</td><td>$srcfile</td><td><a href=\"failures/$ppfile\">$ppfile</a></td><td><a href=\"failures/$ppfile.stderr.txt\">$ppfile.stderr.txt</a></td></tr>\n";
        my $ppfile_clang = $ppfile;
        $ppfile_clang =~ s/[.](.+)$/.clang.$1/;
        print OUT "  <!-- REPORTPROBLEM src=\"$srcfile\" file=\"failures/$ppfile\" clangfile=\"failures/$ppfile_clang\" stderr=\"failures/$ppfile.stderr.txt\" info=\"failures/$ppfile.info.txt\" -->\n";
      }
      print OUT "</table>\n";
    }
    
    if (scalar @failures) {
      print OUT "<p>The analyzer had problems processing the following files:</p>\n";
      print OUT "<table>\n";
      print OUT "<thead><tr><td>Problem</td><td>Source File</td><td>Preprocessed File</td><td>STDERR Output</td></tr></thead>\n";
      foreach my $file (sort @failures) {
        $file =~ /(.+).info.txt$/;
        # Get the preprocessed file.
        my $ppfile = $1;
        # Open the info file and get the name of the source file.
        open (INFO, "$Dir/failures/$file") or
          die "Cannot open $Dir/failures/$file\n";
        my $srcfile = <INFO>;
        chomp $srcfile;
        my $problem = <INFO>;
        chomp $problem;
        close (INFO);
        # Print the information in the table.
        my $prefix = GetPrefix();
        if (defined $prefix) { $srcfile =~ s/^\Q$prefix//; }
        print OUT "<tr><td>$problem</td><td>$srcfile</td><td><a href=\"failures/$ppfile\">$ppfile</a></td><td><a href=\"failures/$ppfile.stderr.txt\">$ppfile.stderr.txt</a></td></tr>\n";
        my $ppfile_clang = $ppfile;
        $ppfile_clang =~ s/[.](.+)$/.clang.$1/;
        print OUT "  <!-- REPORTPROBLEM src=\"$srcfile\" file=\"failures/$ppfile\" clangfile=\"failures/$ppfile_clang\" stderr=\"failures/$ppfile.stderr.txt\" info=\"failures/$ppfile.info.txt\" -->\n";
      }
      print OUT "</table>\n";
    }    
    print OUT "<p>Please consider submitting preprocessed files as <a href=\"http://clang-analyzer.llvm.org/filing_bugs.html\">bug reports</a>. <!-- REPORTCRASHES --> </p>\n";
  }
  
  print OUT "</body></html>\n";  
  close(OUT);
  CopyFiles($Dir);

  # Make sure $Dir and $BaseDir are world readable/executable.
  system("chmod", "755", $Dir);
  if (defined $BaseDir) { system("chmod", "755", $BaseDir); }

  # Print statistics
  print CalcStats(\@Stats) if $AnalyzerStats;

  my $Num = scalar(@Index);
  Diag("$Num bugs found.\n");
  if ($Num > 0 && -r "$Dir/index.html") {
    Diag("Run 'scan-view $Dir' to examine bug reports.\n");
  }
  
  DiagCrashes($Dir) if (scalar @failures || scalar @attributes_ignored);
  
  return $Num;
}

##----------------------------------------------------------------------------##
# RunBuildCommand - Run the build command.
##----------------------------------------------------------------------------##

sub AddIfNotPresent {
  my $Args = shift;
  my $Arg = shift;  
  my $found = 0;
  
  foreach my $k (@$Args) {
    if ($k eq $Arg) {
      $found = 1;
      last;
    }
  }
  
  if ($found == 0) {
    push @$Args, $Arg;
  }
}

sub RunBuildCommand {
  
  my $Args = shift;
  my $IgnoreErrors = shift;
  my $Cmd = $Args->[0];
  my $CCAnalyzer = shift;
  my $CXXAnalyzer = shift;
  
  # Get only the part of the command after the last '/'.
  if ($Cmd =~ /\/([^\/]+)$/) {
    $Cmd = $1;
  }
  
  if ($Cmd =~ /(.*\/?gcc[^\/]*$)/ or 
      $Cmd =~ /(.*\/?cc[^\/]*$)/ or
      $Cmd =~ /(.*\/?llvm-gcc[^\/]*$)/ or
      $Cmd =~ /(.*\/?clang$)/ or 
      $Cmd =~ /(.*\/?ccc-analyzer[^\/]*$)/) {

    if (!($Cmd =~ /ccc-analyzer/) and !defined $ENV{"CCC_CC"}) {
      $ENV{"CCC_CC"} = $1;      
    }
        
    shift @$Args;
    unshift @$Args, $CCAnalyzer;
  }
  elsif ($Cmd =~ /(.*\/?g\+\+[^\/]*$)/ or 
        $Cmd =~ /(.*\/?c\+\+[^\/]*$)/ or
        $Cmd =~ /(.*\/?llvm-g\+\+[^\/]*$)/ or
        $Cmd =~ /(.*\/?clang\+\+$)/ or
        $Cmd =~ /(.*\/?c\+\+-analyzer[^\/]*$)/) {
    if (!($Cmd =~ /c\+\+-analyzer/) and !defined $ENV{"CCC_CXX"}) {
      $ENV{"CCC_CXX"} = $1;      
    }        
    shift @$Args;
    unshift @$Args, $CXXAnalyzer;
  }
  elsif ($IgnoreErrors) {
    if ($Cmd eq "make" or $Cmd eq "gmake") {
      AddIfNotPresent($Args, "CC=$CCAnalyzer");
      AddIfNotPresent($Args, "CXX=$CXXAnalyzer");
      AddIfNotPresent($Args,"-k");
      AddIfNotPresent($Args,"-i");
    }
    elsif ($Cmd eq "xcodebuild") {
      AddIfNotPresent($Args,"-PBXBuildsContinueAfterErrors=YES");
    }
  } 
  
  if ($Cmd eq "xcodebuild") {
    # Check if using iPhone SDK 3.0 (simulator).  If so the compiler being
    # used should be gcc-4.2.
    if (!defined $ENV{"CCC_CC"}) {
      for (my $i = 0 ; $i < scalar(@$Args); ++$i) {
        if ($Args->[$i] eq "-sdk" && $i + 1 < scalar(@$Args)) {
          if (@$Args[$i+1] =~ /^iphonesimulator3/) {
            $ENV{"CCC_CC"} = "gcc-4.2";
            $ENV{"CCC_CXX"} = "g++-4.2";            
          }
        }
      }
    }

    # Disable PCH files until clang supports them.
    AddIfNotPresent($Args,"GCC_PRECOMPILE_PREFIX_HEADER=NO");
    
    # When 'CC' is set, xcodebuild uses it to do all linking, even if we are
    # linking C++ object files.  Set 'LDPLUSPLUS' so that xcodebuild uses 'g++'
    # (via c++-analyzer) when linking such files.
    $ENV{"LDPLUSPLUS"} = $CXXAnalyzer;
  }
  
  return (system(@$Args) >> 8);
}

##----------------------------------------------------------------------------##
# DisplayHelp - Utility function to display all help options.
##----------------------------------------------------------------------------##

sub DisplayHelp {
  
print <<ENDTEXT;
USAGE: $Prog [options] <build command> [build options]

ENDTEXT

  if (defined $BuildName) {
    print "ANALYZER BUILD: $BuildName ($BuildDate)\n\n";
  }

print <<ENDTEXT;
OPTIONS:

 -analyze-headers - Also analyze functions in #included files.
 
 -o             - Target directory for HTML report files.  Subdirectories
                  will be created as needed to represent separate "runs" of
                  the analyzer.  If this option is not specified, a directory
                  is created in /tmp (TMPDIR on Mac OS X) to store the reports.
                  
 -h             - Display this message.
 --help

 -k             - Add a "keep on going" option to the specified build command.
 --keep-going     This option currently supports make and xcodebuild.
                  This is a convenience option; one can specify this
                  behavior directly using build options.

 --html-title [title]       - Specify the title used on generated HTML pages.
 --html-title=[title]         If not specified, a default title will be used.

 -plist         - By default the output of scan-build is a set of HTML files.
                  This option outputs the results as a set of .plist files.
 
 -plist-html    - By default the output of scan-build is a set of HTML files.
                  This option outputs the results as a set of HTML 
                  and .plist files.
 
 --status-bugs  - By default, the exit status of $Prog is the same as the
                  executed build command.  Specifying this option causes the
                  exit status of $Prog to be 1 if it found potential bugs
                  and 0 otherwise.

 --use-cc [compiler path]   - $Prog attempts to guess the default compiler for
 --use-cc=[compiler path]     your C and Objective-C code. Use this option
                              to specify an alternate compiler.

 --use-c++ [compiler path]  - $Prog attempts to guess the default compiler for
 --use-c++=[compiler path]    your C++ and Objective-C++ code. Use this option
                              to specify an alternate compiler.

 -v             - Verbose output from $Prog and the analyzer.
                  A second and third '-v' increases verbosity.

 -V             - View analysis results in a web browser when the build
 --view           completes.

ADVANCED OPTIONS:

 -constraints [model] - Specify the contraint engine used by the analyzer.
                        By default the 'range' model is used.  Specifying 
                        'basic' uses a simpler, less powerful constraint model
                        used by checker-0.160 and earlier.

 -store [model] - Specify the store model used by the analyzer. By default,
                  the 'region' store model is used. 'region' specifies a field-
                  sensitive store model. Users can also specify 'basic', which
                  is far less precise but can more quickly analyze code.
                  'basic' was the default store model for checker-0.221 and
                  earlier.

 -no-failure-reports - Do not create a 'failures' subdirectory that includes
                       analyzer crash reports and preprocessed source files.

 -stats - Generates visitation statistics for the project being analyzed.

 -maxloop N - specifiy the number of times a block can be visited before giving
              up. Default is 4. Increase for more comprehensive coverage at a
              cost of speed.

CONTROLLING CHECKERS:

 A default group of checkers are always run unless explicitly disabled.
 Checkers may be enabled/disabled using the following options:

 -enable-checker [checker name]
 -disable-checker [checker name]
 
LOADING CHECKERS:

 Loading external checkers using the clang plugin interface:

 -load-plugin [plugin library]
ENDTEXT

# Query clang for list of checkers that are enabled.

# create a list to load the plugins via the 'Xclang' command line
# argument
my @PluginLoadCommandline_xclang;
foreach my $param ( @PluginsToLoad ) {
  push ( @PluginLoadCommandline_xclang, "-Xclang" );
  push ( @PluginLoadCommandline_xclang, $param );
}
my %EnabledCheckers;
foreach my $lang ("c", "objective-c", "objective-c++", "c++") {
  pipe(FROM_CHILD, TO_PARENT);
  my $pid = fork();
  if ($pid == 0) {
    close FROM_CHILD;
    open(STDOUT,">&", \*TO_PARENT);
    open(STDERR,">&", \*TO_PARENT);
    exec $Clang, ( @PluginLoadCommandline_xclang, '--analyze', '-x', $lang, '-', '-###'); 
  }
  close(TO_PARENT);
  while(<FROM_CHILD>) {
    foreach my $val (split /\s+/) {
      $val =~ s/\"//g;
      if ($val =~ /-analyzer-checker\=([^\s]+)/) {
        $EnabledCheckers{$1} = 1;
      }
    }
  }
  waitpid($pid,0);
  close(FROM_CHILD);
}

# Query clang for complete list of checkers.
pipe(FROM_CHILD, TO_PARENT);
my $pid = fork();
if ($pid == 0) {
  close FROM_CHILD;
  open(STDOUT,">&", \*TO_PARENT);
  open(STDERR,">&", \*TO_PARENT);
  exec $Clang, ('-cc1', @PluginsToLoad , '-analyzer-checker-help');
}
close(TO_PARENT);
my $foundCheckers = 0;
while(<FROM_CHILD>) {
  if (/CHECKERS:/) {
    $foundCheckers = 1;
    last;
  }
}
if (!$foundCheckers) {
  print "  *** Could not query Clang for the list of available checkers.";
}
else {
  print("\nAVAILABLE CHECKERS:\n\n");
  my $skip = 0;
  while(<FROM_CHILD>) {
    if (/experimental/) {
      $skip = 1;
      next;
    }
    if ($skip) {
      next if (!/^\s\s[^\s]/);
      $skip = 0;
    }
    s/^\s\s//;
    if (/^([^\s]+)/) {
      # Is the checker enabled?
      my $checker = $1;
      my $enabled = 0;
      my $aggregate = "";
      foreach my $domain (split /\./, $checker) {
        $aggregate .= $domain;
        if ($EnabledCheckers{$aggregate}) {
          $enabled =1;
          last;
        }
        # append a dot, if an additional domain is added in the next iteration
        $aggregate .= ".";
      }
      
      if ($enabled) {
        print " + ";
      }
      else {
        print "   ";
      }
    }
    else {
      print "   ";
    }
    print $_;
  } 
}
waitpid($pid,0);
close(FROM_CHILD);

print <<ENDTEXT

 NOTE: "+" indicates that an analysis is enabled by default.

BUILD OPTIONS

 You can specify any build option acceptable to the build command.

EXAMPLE

 $Prog -o /tmp/myhtmldir make -j4
     
 The above example causes analysis reports to be deposited into
 a subdirectory of "/tmp/myhtmldir" and to run "make" with the "-j4" option.
 A different subdirectory is created each time $Prog analyzes a project.
 The analyzer should support most parallel builds, but not distributed builds.

ENDTEXT
}

##----------------------------------------------------------------------------##
# HtmlEscape - HTML entity encode characters that are special in HTML
##----------------------------------------------------------------------------##

sub HtmlEscape {
  # copy argument to new variable so we don't clobber the original
  my $arg = shift || '';
  my $tmp = $arg;
  $tmp =~ s/&/&amp;/g;
  $tmp =~ s/</&lt;/g;
  $tmp =~ s/>/&gt;/g;
  return $tmp;
}

##----------------------------------------------------------------------------##
# ShellEscape - backslash escape characters that are special to the shell
##----------------------------------------------------------------------------##

sub ShellEscape {
  # copy argument to new variable so we don't clobber the original
  my $arg = shift || '';
  if ($arg =~ /["\s]/) { return "'" . $arg . "'"; }
  return $arg;
}

##----------------------------------------------------------------------------##
# Process command-line arguments.
##----------------------------------------------------------------------------##

my $AnalyzeHeaders = 0;
my $HtmlDir;           # Parent directory to store HTML files.
my $IgnoreErrors = 0;  # Ignore build errors.
my $ViewResults  = 0;  # View results when the build terminates.
my $ExitStatusFoundBugs = 0; # Exit status reflects whether bugs were found
my @AnalysesToRun;
my $StoreModel;
my $ConstraintsModel;
my $OutputFormat = "html";
my $AnalyzerStats = 0;
my $MaxLoop = 0;

if (!@ARGV) {
  DisplayHelp();
  exit 1;
}


my $displayHelp = 0;

while (@ARGV) {
  
  # Scan for options we recognize.
  
  my $arg = $ARGV[0];

  if ($arg eq "-h" or $arg eq "--help") {
    $displayHelp = 1;
    shift @ARGV;
    next;
  }
  
  if ($arg eq '-analyze-headers') {
    shift @ARGV;    
    $AnalyzeHeaders = 1;
    next;
  }
  
  if ($arg eq "-o") {
    shift @ARGV;
        
    if (!@ARGV) {
      DieDiag("'-o' option requires a target directory name.\n");
    }
    
    # Construct an absolute path.  Uses the current working directory
    # as a base if the original path was not absolute.
    $HtmlDir = abs_path(shift @ARGV);
    
    next;
  }

  if ($arg =~ /^--html-title(=(.+))?$/) {
    shift @ARGV;

    if (!defined $2 || $2 eq '') {
      if (!@ARGV) {
        DieDiag("'--html-title' option requires a string.\n");
      }

      $HtmlTitle = shift @ARGV;
    } else {
      $HtmlTitle = $2;
    }

    next;
  }
  
  if ($arg eq "-k" or $arg eq "--keep-going") {
    shift @ARGV;
    $IgnoreErrors = 1;
    next;
  }

  if ($arg =~ /^--use-cc(=(.+))?$/) {
    shift @ARGV;
    my $cc;
    
    if (!defined $2 || $2 eq "") {
      if (!@ARGV) {
        DieDiag("'--use-cc' option requires a compiler executable name.\n");
      }
      $cc = shift @ARGV;
    }
    else {
      $cc = $2;
    }
    
    $ENV{"CCC_CC"} = $cc;
    next;
  }
  
  if ($arg =~ /^--use-c\+\+(=(.+))?$/) {
    shift @ARGV;
    my $cxx;    
    
    if (!defined $2 || $2 eq "") {
      if (!@ARGV) {
        DieDiag("'--use-c++' option requires a compiler executable name.\n");
      }
      $cxx = shift @ARGV;
    }
    else {
      $cxx = $2;
    }
    
    $ENV{"CCC_CXX"} = $cxx;
    next;
  }
  
  if ($arg eq "-v") {
    shift @ARGV;
    $Verbose++;
    next;
  }
  
  if ($arg eq "-V" or $arg eq "--view") {
    shift @ARGV;
    $ViewResults = 1;    
    next;
  }
  
  if ($arg eq "--status-bugs") {
    shift @ARGV;
    $ExitStatusFoundBugs = 1;
    next;
  }

  if ($arg eq "-store") {
    shift @ARGV;
    $StoreModel = shift @ARGV;
    next;
  }
  
  if ($arg eq "-constraints") {
    shift @ARGV;
    $ConstraintsModel = shift @ARGV;
    next;
  }
  
  if ($arg eq "-plist") {
    shift @ARGV;
    $OutputFormat = "plist";
    next;
  }
  if ($arg eq "-plist-html") {
    shift @ARGV;
    $OutputFormat = "plist-html";
    next;
  }
  
  if ($arg eq "-no-failure-reports") {
    $ENV{"CCC_REPORT_FAILURES"} = 0;
    next;
  }
  if ($arg eq "-stats") {
    shift @ARGV;
    $AnalyzerStats = 1;
    next;
  }
  if ($arg eq "-maxloop") {
    shift @ARGV;
    $MaxLoop = shift @ARGV;
    next;
  }
  if ($arg eq "-enable-checker") {
    shift @ARGV;
    push @AnalysesToRun, "-analyzer-checker", shift @ARGV;
    next;
  }
  if ($arg eq "-disable-checker") {
    shift @ARGV;
    push @AnalysesToRun, "-analyzer-disable-checker", shift @ARGV;
    next;
  }
  if ($arg eq "-load-plugin") {
    shift @ARGV;
    push @PluginsToLoad, "-load", shift @ARGV;
    next;
  }
  
  DieDiag("unrecognized option '$arg'\n") if ($arg =~ /^-/);
  
  last;
}

if (!@ARGV and $displayHelp == 0) {
  Diag("No build command specified.\n\n");
  $displayHelp = 1;
}

if ($displayHelp) {
  DisplayHelp();
  exit 1;
}

# Determine where results go.
$CmdArgs = HtmlEscape(join(' ', map(ShellEscape($_), @ARGV)));
$HtmlTitle = "${CurrentDirSuffix} - scan-build results"
  unless (defined($HtmlTitle));

# Determine the output directory for the HTML reports.
my $BaseDir = $HtmlDir;
$HtmlDir = GetHTMLRunDir($HtmlDir);

# Determine the location of ccc-analyzer.
my $AbsRealBin = Cwd::realpath($RealBin);
my $Cmd = "$AbsRealBin/libexec/ccc-analyzer";
my $CmdCXX = "$AbsRealBin/libexec/c++-analyzer";

if (!defined $Cmd || ! -x $Cmd) {
  $Cmd = "$AbsRealBin/ccc-analyzer";
  DieDiag("Executable 'ccc-analyzer' does not exist at '$Cmd'\n") if(! -x $Cmd);
}
if (!defined $CmdCXX || ! -x $CmdCXX) {
  $CmdCXX = "$AbsRealBin/c++-analyzer";
  DieDiag("Executable 'c++-analyzer' does not exist at '$CmdCXX'\n") if(! -x $CmdCXX);
}

if (!defined $ClangSB || ! -x $ClangSB) {
  Diag("'clang' executable not found in '$RealBin/bin'.\n");
  Diag("Using 'clang' from path: $Clang\n");
}

# Set the appropriate environment variables.
SetHtmlEnv(\@ARGV, $HtmlDir);
$ENV{'CC'} = $Cmd;
$ENV{'CXX'} = $CmdCXX;
$ENV{'CLANG'} = $Clang;
$ENV{'CLANG_CXX'} = $ClangCXX;
if ($Verbose >= 2) {
  $ENV{'CCC_ANALYZER_VERBOSE'} = 1;
}
if ($Verbose >= 3) {
  $ENV{'CCC_ANALYZER_LOG'} = 1;
}
if ($AnalyzeHeaders) {
  push @AnalysesToRun,"-analyzer-opt-analyze-headers";  
}
if ($AnalyzerStats) {
  push @AnalysesToRun, '-analyzer-checker', 'debug.Stats';
}
if ($MaxLoop > 0) {
  push @AnalysesToRun, '-analyzer-max-loop ' . $MaxLoop;
}

$ENV{'CCC_ANALYZER_ANALYSIS'} = join ' ',@AnalysesToRun;

$ENV{'CCC_ANALYZER_PLUGINS'} = join ' ',@PluginsToLoad;

if (defined $StoreModel) {
  $ENV{'CCC_ANALYZER_STORE_MODEL'} = $StoreModel;
}
if (defined $ConstraintsModel) {
  $ENV{'CCC_ANALYZER_CONSTRAINTS_MODEL'} = $ConstraintsModel;
}
if (defined $OutputFormat) {
  $ENV{'CCC_ANALYZER_OUTPUT_FORMAT'} = $OutputFormat;
}

# Run the build.
my $ExitStatus = RunBuildCommand(\@ARGV, $IgnoreErrors, $Cmd, $CmdCXX);

if (defined $OutputFormat) {
  if ($OutputFormat =~ /plist/) {
    Diag "Analysis run complete.\n";
    Diag "Analysis results (plist files) deposited in '$HtmlDir'\n";
  }
  elsif ($OutputFormat =~ /html/) {
    # Postprocess the HTML directory.
    my $NumBugs = Postprocess($HtmlDir, $BaseDir, $AnalyzerStats);

    if ($ViewResults and -r "$HtmlDir/index.html") {
      Diag "Analysis run complete.\n";
      Diag "Viewing analysis results in '$HtmlDir' using scan-view.\n";
      my $ScanView = Cwd::realpath("$RealBin/scan-view");
      if (! -x $ScanView) { $ScanView = "scan-view"; }
      exec $ScanView, "$HtmlDir";
    }

    if ($ExitStatusFoundBugs) {
      exit 1 if ($NumBugs > 0);
      exit 0;
    }
  }
}

exit $ExitStatus;