/[pcre]/code/tags/pcre-4.3/perltest
ViewVC logotype

Contents of /code/tags/pcre-4.3/perltest

Parent Directory Parent Directory | Revision Log Revision Log


Revision 70 - (show annotations) (download)
Sat Feb 24 21:40:20 2007 UTC (7 years, 8 months ago) by nigel
File size: 5232 byte(s)
Tag code/trunk as code/tags/pcre-4.3.

1 #! /usr/bin/perl
2
3 # Program for testing regular expressions with perl to check that PCRE handles
4 # them the same. This is the version that supports /8 for UTF-8 testing. As it
5 # stands, it requires at least Perl 5.8 for UTF-8 support. For Perl 5.6, it
6 # can be used as is for non-UTF-8 testing, but you have to uncomment the
7 # "use utf8" lines in order to to UTF-8 stuff (and you mustn't uncomment them
8 # for non-UTF-8 use).
9
10
11 # Function for turning a string into a string of printing chars. There are
12 # currently problems with UTF-8 strings; this fudges round them.
13
14 sub pchars {
15 my($t) = "";
16
17 if ($utf8)
18 {
19 # use utf8; <=============== For UTF-8 in Perl 5.6
20 @p = unpack('U*', $_[0]);
21 foreach $c (@p)
22 {
23 if ($c >= 32 && $c < 127) { $t .= chr $c; }
24 else { $t .= sprintf("\\x{%02x}", $c); }
25 }
26 }
27
28 else
29 {
30 foreach $c (split(//, $_[0]))
31 {
32 if (ord $c >= 32 && ord $c < 127) { $t .= $c; }
33 else { $t .= sprintf("\\x%02x", ord $c); }
34 }
35 }
36
37 $t;
38 }
39
40
41
42 # Read lines from named file or stdin and write to named file or stdout; lines
43 # consist of a regular expression, in delimiters and optionally followed by
44 # options, followed by a set of test data, terminated by an empty line.
45
46 # Sort out the input and output files
47
48 if (@ARGV > 0)
49 {
50 open(INFILE, "<$ARGV[0]") || die "Failed to open $ARGV[0]\n";
51 $infile = "INFILE";
52 }
53 else { $infile = "STDIN"; }
54
55 if (@ARGV > 1)
56 {
57 open(OUTFILE, ">$ARGV[1]") || die "Failed to open $ARGV[1]\n";
58 $outfile = "OUTFILE";
59 }
60 else { $outfile = "STDOUT"; }
61
62 printf($outfile "Perl $] Regular Expressions\n\n");
63
64 # Main loop
65
66 NEXT_RE:
67 for (;;)
68 {
69 printf " re> " if $infile eq "STDIN";
70 last if ! ($_ = <$infile>);
71 printf $outfile "$_" if $infile ne "STDIN";
72 next if ($_ eq "");
73
74 $pattern = $_;
75
76 while ($pattern !~ /^\s*(.).*\1/s)
77 {
78 printf " > " if $infile eq "STDIN";
79 last if ! ($_ = <$infile>);
80 printf $outfile "$_" if $infile ne "STDIN";
81 $pattern .= $_;
82 }
83
84 chomp($pattern);
85 $pattern =~ s/\s+$//;
86
87 # The private /+ modifier means "print $' afterwards".
88
89 $showrest = ($pattern =~ s/\+(?=[a-z]*$)//);
90
91 # The private /8 modifier means "operate in UTF-8". Currently, Perl
92 # has bugs that we try to work around using this flag.
93
94 $utf8 = ($pattern =~ s/8(?=[a-z]*$)//);
95
96 # Check that the pattern is valid
97
98 if ($utf8)
99 {
100 # use utf8; <=============== For UTF-8 in Perl 5.6
101 eval "\$_ =~ ${pattern}";
102 }
103 else
104 {
105 eval "\$_ =~ ${pattern}";
106 }
107
108 if ($@)
109 {
110 printf $outfile "Error: $@";
111 next NEXT_RE;
112 }
113
114 # If the /g modifier is present, we want to put a loop round the matching;
115 # otherwise just a single "if".
116
117 $cmd = ($pattern =~ /g[a-z]*$/)? "while" : "if";
118
119 # If the pattern is actually the null string, Perl uses the most recently
120 # executed (and successfully compiled) regex is used instead. This is a
121 # nasty trap for the unwary! The PCRE test suite does contain null strings
122 # in places - if they are allowed through here all sorts of weird and
123 # unexpected effects happen. To avoid this, we replace such patterns with
124 # a non-null pattern that has the same effect.
125
126 $pattern = "/(?#)/$2" if ($pattern =~ /^(.)\1(.*)$/);
127
128 # Read data lines and test them
129
130 for (;;)
131 {
132 printf "data> " if $infile eq "STDIN";
133 last NEXT_RE if ! ($_ = <$infile>);
134 chomp;
135 printf $outfile "$_\n" if $infile ne "STDIN";
136
137 s/\s+$//;
138 s/^\s+//;
139
140 last if ($_ eq "");
141
142 $x = eval "\"$_\""; # To get escapes processed
143
144 # Empty array for holding results, then do the matching.
145
146 @subs = ();
147
148 $pushes = "push \@subs,\$&;" .
149 "push \@subs,\$1;" .
150 "push \@subs,\$2;" .
151 "push \@subs,\$3;" .
152 "push \@subs,\$4;" .
153 "push \@subs,\$5;" .
154 "push \@subs,\$6;" .
155 "push \@subs,\$7;" .
156 "push \@subs,\$8;" .
157 "push \@subs,\$9;" .
158 "push \@subs,\$10;" .
159 "push \@subs,\$11;" .
160 "push \@subs,\$12;" .
161 "push \@subs,\$13;" .
162 "push \@subs,\$14;" .
163 "push \@subs,\$15;" .
164 "push \@subs,\$16;" .
165 "push \@subs,\$'; }";
166
167 if ($utf8)
168 {
169 # use utf8; <=============== For UTF-8 in Perl 5.6
170 eval "${cmd} (\$x =~ ${pattern}) {" . $pushes;
171 }
172 else
173 {
174 eval "${cmd} (\$x =~ ${pattern}) {" . $pushes;
175 }
176
177 if ($@)
178 {
179 printf $outfile "Error: $@\n";
180 next NEXT_RE;
181 }
182 elsif (scalar(@subs) == 0)
183 {
184 printf $outfile "No match\n";
185 }
186 else
187 {
188 while (scalar(@subs) != 0)
189 {
190 printf $outfile (" 0: %s\n", &pchars($subs[0]));
191 printf $outfile (" 0+ %s\n", &pchars($subs[17])) if $showrest;
192 $last_printed = 0;
193 for ($i = 1; $i <= 16; $i++)
194 {
195 if (defined $subs[$i])
196 {
197 while ($last_printed++ < $i-1)
198 { printf $outfile ("%2d: <unset>\n", $last_printed); }
199 printf $outfile ("%2d: %s\n", $i, &pchars($subs[$i]));
200 $last_printed = $i;
201 }
202 }
203 splice(@subs, 0, 18);
204 }
205 }
206 }
207 }
208
209 printf $outfile "\n";
210
211 # End

Properties

Name Value
svn:executable *

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12