prest/doc/tools/htmldiff.prl

   1 #!/usr/local/bin/perl
   2
   3 =pod
   4 =begin reST
   5 =begin Id
   6 $Id: htmldiff.prl 767 2006-01-28 03:29:15Z marknodine $
   7 Copyright (C) 2002-2005 Freescale Semiconductor, Inc.
   8 Distributed under terms of the GNU General Public License (GPL).
   9 =end Id
  10
  11 =begin Description
  12 Description: This tool takes two HTML files and creates an HTML file
  13 that has markings of text insertions and deletions.
  14
  15 =end Description
  16 =begin Usage
  17 Usage: htmldiff [options] file1 file2
  18
  19 Options:
  20    -b string      String to insert for backward link with -l.
  21                   Default: "${opt_b}"
  22    -d string      Deletion marking string, where "$_" specifies the
  23                   deleted words.
  24                   Default: "${opt_d}"
  25    -f string      String to insert for forward link with -l.
  26                   Default: "${opt_f}"
  27    -h             Prints help
  28    -i string      Insertion marking string, where "$_" specifies the
  29                   inserted words.
  30                   Default: "${opt_i}"
  31    -k             Create key describing difference format
  32    -l             Create links between difference regions
  33    -p string      Program to run for diff (program must support -D option).
  34                   Default: ${opt_p}
  35    -V             Prints version information
  36 =end Usage
  37 =end reST
  38 =cut
  39
  40 BEGIN {
  41     $opt_b = '<font color=orange>&lt;</font>';
  42     $opt_f = '<font color=orange>&gt;</font>';
  43     $opt_i = '<U><FONT COLOR=#800080>$_</FONT></U>';
  44     $opt_d = '<S><FONT COLOR=#800000>$_</FONT></S>';
  45     $opt_p = 'diff';
  46 }
  47
  48 # Get options and process them
  49 use Getopt::Std;
  50 Usage() unless getopts("b:d:f:hi:klp:V");
  51
  52 Usage('Id') if $opt_V;
  53 Usage('Description') if $opt_h;
  54
  55 die "Wrong number of arguments" unless @ARGV == 2;
  56
  57 $opt_i =~ /(.*)\$_(.*)/;
  58 ($INSERT_PFX,$INSERT_SFX) = ($1, $2);
  59 $opt_d =~ /(.*)\$_(.*)/;
  60 ($DELETE_PFX,$DELETE_SFX) = ($1, $2);
  61
  62 #sub INSERTION { "$INSERT_PFX$_[0]$INSERT_SFX" }
  63 #sub DELETION { OptLString() . "$DELETE_PFX$_[0]$DELETE_SFX" }
  64 ($ctx_both, $ctx_1, $ctx_2) = (0 .. 2);
  65
  66 $args = join(' ',@ARGV);
  67 $sym = "A__";
  68 open (DIFF, "$opt_p -D $sym $args|") || die "Cannot run diff";
  69 $context = $ctx_both;
  70 $DIFF = 0;
  71 if ($opt_k) {
  72     print "Key: ${DELETE_PFX}Deleted text.$DELETE_SFX\n";
  73     print "${INSERT_PFX}Inserted text.$INSERT_SFX\n";
  74     print "Use <A HREF='#'>$opt_b</A> and <A HREF='#__diff${\scalar($DIFF+1)}'>$opt_f</A> to jump to previous/next difference region.\n"
  75         if $opt_l;
  76     print "<hr />\n";
  77 }
  78 print  "<A HREF='#__diff${\scalar($DIFF+1)}' NAME='__diff$DIFF'>$opt_f</A> "
  79     if $opt_l;
  80
  81 while (<DIFF>) {
  82     if (/^#ifndef $sym\b/o) {
  83         $context = $ctx_1; next;
  84     }
  85     if (/^#ifdef $sym\b/o) {
  86         $context = $ctx_2; next;
  87     }
  88     if (/^#else /o) {
  89         $context = 3 - $context; next;
  90     }
  91     if (/^#endif /o) {
  92         $context = $ctx_both;
  93         # Process the diff
  94         # Get the word lists for each section
  95         $diff1 = join('', @diff1);
  96         $diff2 = join('', @diff2);
  97         if ($diff1 eq '') {
  98             my $s = MarkNonHTML($diff2, $INSERT_PFX, $INSERT_SFX);
  99             $s =~ s/($INSERT_PFX)/OptLString() . $1/eo;
 100             print $s;
 101         }
 102         elsif ($diff2 eq '') {
 103             my $s = MarkNonHTML($diff1, $DELETE_PFX, $DELETE_SFX);
 104             $s =~ s/($DELETE_PFX)/OptLString() . $1/eo;
 105             print $s;
 106         }
 107         else {
 108             @words1 = GetWordList($diff1);
 109             @words2 = GetWordList($diff2);
 110             # Use diff to find out which words were added/deleted/changed
 111             open (O1, "> /tmp/1.$$") || die "Cannot write to /tmp/1.$$";
 112             print O1 join("\n",@words1),"\n";
 113             close O1;
 114             open (O2, "> /tmp/2.$$") || die "Cannot write to /tmp/2.$$";
 115             print O2 join("\n",@words2),"\n";
 116             close O2;
 117             open (DIFF2, "diff /tmp/1.$$ /tmp/2.$$|");
 118             # Record for each difference section the type of difference
 119             # (a, c, or d), the beginning and ending words in diff2, and
 120             # any non-HTML words deleted from diff1.
 121             my(@deletes, @diffs);
 122             while (<DIFF2>) {
 123                 if (/^[0-9]+(?:,[0-9]+)?([acd][0-9]+(?:,[0-9]+)?)/) {
 124                     # A new difference section
 125                     push(@diffs, $1);
 126                     push(@deletes, []);
 127                 }
 128                 elsif (/^< ([^<].*)/) {
 129                     push(@{$deletes[$#deletes]}, $1);
 130                 }
 131             }
 132             close DIFF2;
 133             unlink "/tmp/1.$$";
 134             unlink "/tmp/2.$$";
 135             # Create the merged version.  We copy words from diff2 to
 136             # $merged until we hit a difference section; for each
 137             # difference section, we put the struck out section in and
 138             # then mark as added any non-HTML words from diff2 for c
 139             # and a type difference sections.
 140             my($merged);
 141             my($i);
 142             my($word) = 1;
 143             for ($i = 0; $i < @diffs; $i++) {
 144                 my($diff) = $diffs[$i];
 145                 my($deletes) = $deletes[$i];
 146                 my($type,$start,$end) =
 147                     $diff =~ /([acd])([0-9]+)(?:,([0-9]+))?/;
 148                 $start++ if $type eq 'd';
 149                 $end = $start unless $end;
 150                 # Copy over stuff
 151                 while ($word < $start) {
 152                     if ($diff2 =~ /^(\s+)/) {
 153                     }
 154                     elsif ($diff2 =~ /^(<.*?>|[^<\s]+)/) {
 155                         $word++;
 156                     }
 157                     else { next; }
 158                     $merged .= $1;
 159                     $diff2 = "$'";
 160                 }
 161                 # Mark any deletions
 162                 my $new;
 163                 $new = ' ' .
 164                     MarkNonHTML((join(' ',@$deletes)), $DELETE_PFX,
 165                                 $DELETE_SFX) . ' '
 166                     if @$deletes;
 167                 if ($type =~ /[ac]/) {
 168                     # Mark any additions.  Copy the whole string to $add.
 169                     my($add);
 170                     while ($word <= $end) {
 171                         if ($diff2 =~ /^(\s+)/) {
 172                         }
 173                         elsif ($diff2 =~ /^(<.*?>|[^<\s]+)/) {
 174                             $word++;
 175                         }
 176                         else { next; }
 177                         $add .= $1;
 178                         $diff2 = "$'";
 179                     }
 180                     # Now mark the non-HTML words.
 181                     $new .= MarkNonHTML($add, $INSERT_PFX, $INSERT_SFX);
 182                 }
 183                 $merged .= (($new =~ /$DELETE_PFX|$INSERT_PFX/o) ?
 184                             OptLString() : "") . $new;
 185             }
 186             # Copy the rest of the diff2 text
 187             $merged .= $diff2;
 188             print $merged;
 189         }
 190         undef @diff1;
 191         undef @diff2;
 192         next;
 193     }
 194     if ($context == $ctx_both) {
 195         print;
 196     }
 197     elsif ($context == $ctx_1) {
 198         push(@diff1, $_);
 199     }
 200     else {
 201         push(@diff2, $_);
 202     }
 203 }
 204 print "<A HREF='#__diff${\scalar($DIFF++)}' NAME='__diff$DIFF'>$opt_b</A>\n"
 205     if $opt_l;
 206 print "($DIFF differences)\n" if $opt_k;
 207
 208 # Returns a list of words.  A word is either a set of non-space
 209 # characters separated by spaces or a string enclosed between '<' and '>'.
 210 # Inputs: string
 211 # Outputs: array of words
 212 sub GetWordList {
 213     $_ = $_[0];
 214     my(@out);
 215     my($last_pos) = -1;
 216     while ($last_pos != pos) {
 217         $last_pos = pos;
 218         next if (/\G(\s+)/gc);
 219         if (/\G(<.*?>)/gc || /\G([^<\s]+)/gc) {
 220             push (@out, $1);
 221         }
 222     }
 223     return @out;
 224 }
 225
 226 # Marks all non-HTML words by putting the prefix and suffix around
 227 # them.
 228 # Inputs: string, prefix, suffix
 229 # Returns: string
 230 sub MarkNonHTML {
 231     my($s, $pfx, $sfx) = @_;
 232     $s =~ s/(>|\A)(?!\s+(?:<|\Z))([^<]+)/$1$pfx$2$sfx/g;
 233     return $s;
 234 }
 235
 236 # Returns a string that links from one difference section to the next
 237 # Inputs: None
 238 # Returns: string
 239 # Uses globals: $opt_l, $DIFF
 240 sub OptLString {
 241     return $opt_l ?
 242         "<A HREF='#__diff${\scalar($DIFF++)}' NAME='__diff$DIFF'>$opt_b</A> <A HREF='#__diff${\scalar($DIFF+1)}'>$opt_f</A> " :
 243         "";
 244 }
 245
 246 # This subroutine extracts and prints usage information
 247 sub Usage {
 248     my ($what) = @_;
 249     $what = "Usage" if ! $what;
 250     my $mark = $what eq 'Description' ? "($what|Usage)" : $what;
 251     if (open(ME,$0) == 1) {
 252         while (<ME>) {
 253             if ((/^=begin $mark/ .. /^=end $mark/) &&
 254                 ! /^=(begin|end) $mark/) {
 255                 s/(\$\{[^\}]+\})/eval($1)/ge;
 256                 print;
 257             }
 258         }
 259         close(ME);
 260     }
 261     else {
 262         print STDERR "Usage not available.\n";
 263     }
 264     exit (1);
 265 }