From 582b500cd996c96054615870fd13d6ab0ea77428 Mon Sep 17 00:00:00 2001
From: Jay Berkenbilt <ejb@ql.org>
Date: Sat, 10 Oct 2009 15:10:05 +0000
Subject: start integrating windows port

git-svn-id: svn+q:///qpdf/trunk@757 71b93d88-0707-0410-a8cf-f5a4172ac649
---
 external-libs/pcre/perltest | 211 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 211 insertions(+)
 create mode 100644 external-libs/pcre/perltest

(limited to 'external-libs/pcre/perltest')

diff --git a/external-libs/pcre/perltest b/external-libs/pcre/perltest
new file mode 100644
index 00000000..bb34cc83
--- /dev/null
+++ b/external-libs/pcre/perltest
@@ -0,0 +1,211 @@
+#! /usr/bin/perl
+
+# Program for testing regular expressions with perl to check that PCRE handles
+# them the same. This is the version that supports /8 for UTF-8 testing. As it
+# stands, it requires at least Perl 5.8 for UTF-8 support. For Perl 5.6, it
+# can be used as is for non-UTF-8 testing, but you have to uncomment the
+# "use utf8" lines in order to to UTF-8 stuff (and you mustn't uncomment them
+# for non-UTF-8 use).
+
+
+# Function for turning a string into a string of printing chars. There are
+# currently problems with UTF-8 strings; this fudges round them.
+
+sub pchars {
+my($t) = "";
+
+if ($utf8)
+  {
+#  use utf8;                    <=============== For UTF-8 in Perl 5.6
+  @p = unpack('U*', $_[0]);
+  foreach $c (@p)
+    {
+    if ($c >= 32 && $c < 127) { $t .= chr $c; }
+      else { $t .= sprintf("\\x{%02x}", $c); }
+    }
+  }
+
+else
+  {
+  foreach $c (split(//, $_[0]))
+    {
+    if (ord $c >= 32 && ord $c < 127) { $t .= $c; }
+      else { $t .= sprintf("\\x%02x", ord $c); }
+    }
+  }
+
+$t;
+}
+
+
+
+# Read lines from named file or stdin and write to named file or stdout; lines
+# consist of a regular expression, in delimiters and optionally followed by
+# options, followed by a set of test data, terminated by an empty line.
+
+# Sort out the input and output files
+
+if (@ARGV > 0)
+  {
+  open(INFILE, "<$ARGV[0]") || die "Failed to open $ARGV[0]\n";
+  $infile = "INFILE";
+  }
+else { $infile = "STDIN"; }
+
+if (@ARGV > 1)
+  {
+  open(OUTFILE, ">$ARGV[1]") || die "Failed to open $ARGV[1]\n";
+  $outfile = "OUTFILE";
+  }
+else { $outfile = "STDOUT"; }
+
+printf($outfile "Perl $] Regular Expressions\n\n");
+
+# Main loop
+
+NEXT_RE:
+for (;;)
+  {
+  printf "  re> " if $infile eq "STDIN";
+  last if ! ($_ = <$infile>);
+  printf $outfile "$_" if $infile ne "STDIN";
+  next if ($_ eq "");
+
+  $pattern = $_;
+
+  while ($pattern !~ /^\s*(.).*\1/s)
+    {
+    printf "    > " if $infile eq "STDIN";
+    last if ! ($_ = <$infile>);
+    printf $outfile "$_" if $infile ne "STDIN";
+    $pattern .= $_;
+    }
+
+   chomp($pattern);
+   $pattern =~ s/\s+$//;
+
+  # The private /+ modifier means "print $' afterwards".
+
+  $showrest = ($pattern =~ s/\+(?=[a-z]*$)//);
+
+  # The private /8 modifier means "operate in UTF-8". Currently, Perl
+  # has bugs that we try to work around using this flag.
+
+  $utf8 = ($pattern =~ s/8(?=[a-z]*$)//);
+
+  # Check that the pattern is valid
+
+  if ($utf8)
+    {
+#    use utf8;                    <=============== For UTF-8 in Perl 5.6
+    eval "\$_ =~ ${pattern}";
+    }
+  else
+    {
+    eval "\$_ =~ ${pattern}";
+    }
+
+  if ($@)
+    {
+    printf $outfile "Error: $@";
+    next NEXT_RE;
+    }
+
+  # If the /g modifier is present, we want to put a loop round the matching;
+  # otherwise just a single "if".
+
+  $cmd = ($pattern =~ /g[a-z]*$/)? "while" : "if";
+
+  # If the pattern is actually the null string, Perl uses the most recently
+  # executed (and successfully compiled) regex is used instead. This is a
+  # nasty trap for the unwary! The PCRE test suite does contain null strings
+  # in places - if they are allowed through here all sorts of weird and
+  # unexpected effects happen. To avoid this, we replace such patterns with
+  # a non-null pattern that has the same effect.
+
+  $pattern = "/(?#)/$2" if ($pattern =~ /^(.)\1(.*)$/);
+
+  # Read data lines and test them
+
+  for (;;)
+    {
+    printf "data> " if $infile eq "STDIN";
+    last NEXT_RE if ! ($_ = <$infile>);
+    chomp;
+    printf $outfile "$_\n" if $infile ne "STDIN";
+
+    s/\s+$//;
+    s/^\s+//;
+
+    last if ($_ eq "");
+
+    $x = eval "\"$_\"";   # To get escapes processed
+
+    # Empty array for holding results, then do the matching.
+
+    @subs = ();
+
+    $pushes = "push \@subs,\$&;" .
+         "push \@subs,\$1;" .
+         "push \@subs,\$2;" .
+         "push \@subs,\$3;" .
+         "push \@subs,\$4;" .
+         "push \@subs,\$5;" .
+         "push \@subs,\$6;" .
+         "push \@subs,\$7;" .
+         "push \@subs,\$8;" .
+         "push \@subs,\$9;" .
+         "push \@subs,\$10;" .
+         "push \@subs,\$11;" .
+         "push \@subs,\$12;" .
+         "push \@subs,\$13;" .
+         "push \@subs,\$14;" .
+         "push \@subs,\$15;" .
+         "push \@subs,\$16;" .
+         "push \@subs,\$'; }";
+
+    if ($utf8)
+      {
+#      use utf8;                    <=============== For UTF-8 in Perl 5.6
+      eval "${cmd} (\$x =~ ${pattern}) {" . $pushes;
+      }
+    else
+      {
+      eval "${cmd} (\$x =~ ${pattern}) {" . $pushes;
+      }
+
+    if ($@)
+      {
+      printf $outfile "Error: $@\n";
+      next NEXT_RE;
+      }
+    elsif (scalar(@subs) == 0)
+      {
+      printf $outfile "No match\n";
+      }
+    else
+      {
+      while (scalar(@subs) != 0)
+        {
+        printf $outfile (" 0: %s\n", &pchars($subs[0]));
+        printf $outfile (" 0+ %s\n", &pchars($subs[17])) if $showrest;
+        $last_printed = 0;
+        for ($i = 1; $i <= 16; $i++)
+          {
+          if (defined $subs[$i])
+            {
+            while ($last_printed++ < $i-1)
+              { printf $outfile ("%2d: <unset>\n", $last_printed); }
+            printf $outfile ("%2d: %s\n", $i, &pchars($subs[$i]));
+            $last_printed = $i;
+            }
+          }
+        splice(@subs, 0, 18);
+        }
+      }
+    }
+  }
+
+printf $outfile "\n";
+
+# End
-- 
cgit v1.2.3-54-g00ecf