/[pcre]/code/trunk/perltest.pl
ViewVC logotype

Contents of /code/trunk/perltest.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 518 - (show annotations) (download)
Tue May 18 15:47:01 2010 UTC (4 years, 6 months ago) by ph10
File MIME type: text/plain
File size: 5539 byte(s)
Added PCRE_UCP and related stuff to make \w etc use Unicode properties.

1 #! /usr/bin/env 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. However, it needs to
6 # have "use utf8" at the start for running the UTF-8 tests, but *not* for the
7 # other tests. The only way I've found for doing this is to cat this line in
8 # explicitly in the RunPerlTest script.
9
10 # use locale; # With this included, \x0b matches \s!
11
12 # Function for turning a string into a string of printing chars. There are
13 # currently problems with UTF-8 strings; this fudges round them.
14
15 sub pchars {
16 my($t) = "";
17
18 if ($utf8)
19 {
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 $t;
38 }
39
40
41 # Read lines from named file or stdin and write to named file or stdout; lines
42 # consist of a regular expression, in delimiters and optionally followed by
43 # options, followed by a set of test data, terminated by an empty line.
44
45 # Sort out the input and output files
46
47 if (@ARGV > 0)
48 {
49 open(INFILE, "<$ARGV[0]") || die "Failed to open $ARGV[0]\n";
50 $infile = "INFILE";
51 }
52 else { $infile = "STDIN"; }
53
54 if (@ARGV > 1)
55 {
56 open(OUTFILE, ">$ARGV[1]") || die "Failed to open $ARGV[1]\n";
57 $outfile = "OUTFILE";
58 }
59 else { $outfile = "STDOUT"; }
60
61 printf($outfile "Perl $] Regular Expressions\n\n");
62
63 # Main loop
64
65 NEXT_RE:
66 for (;;)
67 {
68 printf " re> " if $infile eq "STDIN";
69 last if ! ($_ = <$infile>);
70 printf $outfile "$_" if $infile ne "STDIN";
71 next if ($_ eq "");
72
73 $pattern = $_;
74
75 while ($pattern !~ /^\s*(.).*\1/s)
76 {
77 printf " > " if $infile eq "STDIN";
78 last if ! ($_ = <$infile>);
79 printf $outfile "$_" if $infile ne "STDIN";
80 $pattern .= $_;
81 }
82
83 chomp($pattern);
84 $pattern =~ s/\s+$//;
85
86 # The private /+ modifier means "print $' afterwards".
87
88 $showrest = ($pattern =~ s/\+(?=[a-zA-Z]*$)//);
89
90 # Remove /8 from a UTF-8 pattern.
91
92 $utf8 = $pattern =~ s/8(?=[a-zA-Z]*$)//;
93
94 # Remove /J from a pattern with duplicate names.
95
96 $pattern =~ s/J(?=[a-zA-Z]*$)//;
97
98 # Remove /K from a pattern (asks pcretest to check MARK data) */
99
100 $pattern =~ s/K(?=[a-zA-Z]*$)//;
101
102 # Remove /W from a pattern (asks pcretest to set PCRE_UCP)
103
104 $pattern =~ s/W(?=[a-zA-Z]*$)//;
105
106 # Check that the pattern is valid
107
108 eval "\$_ =~ ${pattern}";
109 if ($@)
110 {
111 printf $outfile "Error: $@";
112 next NEXT_RE;
113 }
114
115 # If the /g modifier is present, we want to put a loop round the matching;
116 # otherwise just a single "if".
117
118 $cmd = ($pattern =~ /g[a-z]*$/)? "while" : "if";
119
120 # If the pattern is actually the null string, Perl uses the most recently
121 # executed (and successfully compiled) regex is used instead. This is a
122 # nasty trap for the unwary! The PCRE test suite does contain null strings
123 # in places - if they are allowed through here all sorts of weird and
124 # unexpected effects happen. To avoid this, we replace such patterns with
125 # a non-null pattern that has the same effect.
126
127 $pattern = "/(?#)/$2" if ($pattern =~ /^(.)\1(.*)$/);
128
129 # Read data lines and test them
130
131 for (;;)
132 {
133 printf "data> " if $infile eq "STDIN";
134 last NEXT_RE if ! ($_ = <$infile>);
135 chomp;
136 printf $outfile "$_\n" if $infile ne "STDIN";
137
138 s/\s+$//; # Remove trailing space
139 s/^\s+//; # Remove leading space
140 s/\\Y//g; # Remove \Y (pcretest flag to set PCRE_NO_START_OPTIMIZE)
141
142 last if ($_ eq "");
143 $x = eval "\"$_\""; # To get escapes processed
144
145 # Empty array for holding results, ensure $REGERROR and $REGMARK are
146 # unset, then do the matching.
147
148 @subs = ();
149
150 $pushes = "push \@subs,\$&;" .
151 "push \@subs,\$1;" .
152 "push \@subs,\$2;" .
153 "push \@subs,\$3;" .
154 "push \@subs,\$4;" .
155 "push \@subs,\$5;" .
156 "push \@subs,\$6;" .
157 "push \@subs,\$7;" .
158 "push \@subs,\$8;" .
159 "push \@subs,\$9;" .
160 "push \@subs,\$10;" .
161 "push \@subs,\$11;" .
162 "push \@subs,\$12;" .
163 "push \@subs,\$13;" .
164 "push \@subs,\$14;" .
165 "push \@subs,\$15;" .
166 "push \@subs,\$16;" .
167 "push \@subs,\$'; }";
168
169 undef $REGERROR;
170 undef $REGMARK;
171
172 eval "${cmd} (\$x =~ ${pattern}) {" . $pushes;
173
174 if ($@)
175 {
176 printf $outfile "Error: $@\n";
177 next NEXT_RE;
178 }
179 elsif (scalar(@subs) == 0)
180 {
181 printf $outfile "No match";
182 if (defined $REGERROR && $REGERROR != 1)
183 { print $outfile (", mark = $REGERROR"); }
184 printf $outfile "\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 if (defined $REGMARK && $REGMARK != 1)
206 { print $outfile ("MK: $REGMARK\n"); }
207 }
208 }
209 }
210
211 # printf $outfile "\n";
212
213 # End

Properties

Name Value
svn:executable *

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12