X-Git-Url: http://git.projectaon.org/?p=project-aon.git;a=blobdiff_plain;f=scripts%2Fcheckerrata.pl;fp=scripts%2Fcheckerrata.pl;h=0000000000000000000000000000000000000000;hp=258ce2d2d17007d47297c4ac2d0faabef7119545;hb=97545603aea8298f1eceb604ff65085cc7adfced;hpb=53d5f97cd1474231d3f343c70918a91ade6466e2

diff --git a/scripts/checkerrata.pl b/scripts/checkerrata.pl
deleted file mode 100755
index 258ce2d..0000000
--- a/scripts/checkerrata.pl
+++ /dev/null
@@ -1,88 +0,0 @@
-#!/usr/bin/perl -w
-# This script parses the errata list of a book in XML form and tries to
-# locate the added or replaced strings in the sections where they are now
-# supposed to exist. Things that are not found by the simple errata list
-# parsing code are just skipped. Nesting frontmatter sections are also
-# skipped.
-
-# Old History:
-#  2005-08-23 Tried to improve handling of <link-text> tags
-#  2005-03-12 Treat &thinspace; correctly.
-#  2003-03-16 Fixed reporting of false positived due to <link-text> tags
-#  2003-03-15 First version.
-
-$xmlfile = shift @ARGV;
-$xmlfile || die("usage: $0 <xml file>\n");
-$invalid = shift @ARGV;
-$invalid && die("usage: $0 <xml file>\n");
-
-open(ERRATA, $xmlfile) || die("Cannot read input file $xmlfile\n");
-open(SECTIONS, $xmlfile) || die("Cannot read input file $xmlfile\n");
-
-while(<ERRATA>) { # find errata list
-  last if m/<section.*id="errerr".*>/;
-}
-die("No errata section!? Aborting.\n") if eof(ERRATA);
-while(<ERRATA>) { # find start of actual list
-  last if m/<data>/;
-}
-die("No data in the errata section!? Aborting.\n") if eof(ERRATA);
-
-print("No matches for these errata entries were found:\n");
-
-# Now pick each paragraph (= errata) line
-ERRATALOOP: while(<ERRATA>) {
-  next unless m/^\s*<p>/; # no paragraph = not an errata line
-  last if m|</data>|; # end of the errata list
-  ($sect) = m|\(<a (?:id="\w+?" )?idref="(\w+?)">.*</a>\)|;
-  @reps = m|<quote>.*?</quote> with <quote>(.*?)</quote>|g;
-  @adds = m|[Aa]dded <quote>(.*?(?=</quote>))</quote>|g;
-  next unless @reps || @adds;
-
-  # Now find section and append all contents into one string
-  while(<SECTIONS>) { # locate section
-    last if m/<section.*id=\"$sect\">/;
-  }
-  die("Could not find section $sect!? (This might be because the errata\nentry for it is not placed in the correct section order.) Aborting.\n") if eof(SECTIONS);
-  while(<SECTIONS>) { # locate data
-    last if m/<data>/;
-  }
-  die("Could not find any data in $sect!? Aborting.\n") if eof(SECTIONS);
-  $text = "";
-  while(<SECTIONS>) { # grab all section contents
-    last if m|</data>|;
-    if (m/<section/) { # oh no, nested sections - we cannot handle this well
-      seek(SECTIONS, -length(), 1); # give the next section a chance
-      next ERRATALOOP; # give up on this section
-    }
-    # Before adding the text to the section blob, modify <link-text> tags
-    # since they are commented out in the errata entries
-    s|<(/?)link-text>|<!--${1}link-text-->|g;
-    $text .= $_;
-  }
-  die("Could not find the end of $sect!? Aborting.\n") if eof(SECTIONS);
-
-  # The replacement may contain &[lr][sd]quot; which in the sections are
-  # <quote> thingies. Translate these. Also ignore &thinspace; first and
-  # last in the replacements.
-  # Refactor the duplicated code below some day!
-  foreach $rep (@reps) {
-    $rep =~ s/\&l[sd]quot;/<quote>/g;
-    $rep =~ s/\&r[sd]quot;/<\/quote>/g;
-    $rep =~ s/^\&thinspace;//g;
-    $rep =~ s/\&thinspace;$//g;
-    if ($text !~ m/\Q$rep\E/) {
-      print("Replacement \"$rep\" in $sect\n");
-    }
-  }
-  foreach $add (@adds) {
-    $add =~ s/\&l[sd]quot;/<quote>/g;
-    $add =~ s/\&r[sd]quot;/<\/quote>/g;
-    $add =~ s/^\&thinspace;//g;
-    $add =~ s/\&thinspace;$//g;
-    if ($text !~ m/\Q$add\E/) {
-      print("Addition    \"$add\" in $sect\n");
-    }
-  }
-}
-print("Checking finished!\n");