6 require Config; import Config;
7 if ($Config{ccflags} =~ /\bD?NO_LOCALE\b/) {
15 my $have_setlocale = 0;
18 import POSIX ':locale_h';
22 print "1..", ($have_setlocale ? 102 : 98), "\n";
25 $English $German $French $Spanish
26 @C @English @German @French @Spanish
27 $Locale @Locale %iLocale %UPPER %lower @Neoalpha);
32 my ($n, $result) = @_;
34 print 'not ' unless ($result);
38 # First we'll do a lot of taint checking for locales.
39 # This is the easiest to test, actually, as any locale,
40 # even the default locale will taint under 'use locale'.
42 sub is_tainted { # hello, camel two.
44 not eval { $dummy = join("", @_), kill 0; 1 }
47 sub check_taint ($$) {
48 ok $_[0], is_tainted($_[1]);
51 sub check_taint_not ($$) {
52 ok $_[0], not is_tainted($_[1]);
55 use locale; # engage locale and therefore locale taint.
57 check_taint_not 1, $a;
59 check_taint 2, uc($a);
60 check_taint 3, "\U$a";
61 check_taint 4, ucfirst($a);
62 check_taint 5, "\u$a";
63 check_taint 6, lc($a);
64 check_taint 7, "\L$a";
65 check_taint 8, lcfirst($a);
66 check_taint 9, "\l$a";
68 check_taint 10, sprintf('%e', 123.456);
69 check_taint 11, sprintf('%f', 123.456);
70 check_taint 12, sprintf('%g', 123.456);
71 check_taint_not 13, sprintf('%d', 123.456);
72 check_taint_not 14, sprintf('%x', 123.456);
76 $_ = uc($a); # taint $_
80 /(\w)/; # taint $&, $`, $', $+, $1.
86 check_taint_not 21, $2;
88 /(.)/; # untaint $&, $`, $', $+, $1.
89 check_taint_not 22, $&;
90 check_taint_not 23, $`;
91 check_taint_not 24, $';
92 check_taint_not 25, $+;
93 check_taint_not 26, $1;
94 check_taint_not 27, $2;
96 /(\W)/; # taint $&, $`, $', $+, $1.
102 check_taint_not 33, $2;
104 /(\s)/; # taint $&, $`, $', $+, $1.
110 check_taint_not 39, $2;
112 /(\S)/; # taint $&, $`, $', $+, $1.
118 check_taint_not 45, $2;
120 $_ = $a; # untaint $_
122 check_taint_not 46, $_;
124 /(b)/; # this must not taint
125 check_taint_not 47, $&;
126 check_taint_not 48, $`;
127 check_taint_not 49, $';
128 check_taint_not 50, $+;
129 check_taint_not 51, $1;
130 check_taint_not 52, $2;
132 $_ = $a; # untaint $_
134 check_taint_not 53, $_;
136 $b = uc($a); # taint $b
137 s/(.+)/$b/; # this must taint only the $_
140 check_taint_not 55, $&;
141 check_taint_not 56, $`;
142 check_taint_not 57, $';
143 check_taint_not 58, $+;
144 check_taint_not 59, $1;
145 check_taint_not 60, $2;
147 $_ = $a; # untaint $_
149 s/(.+)/b/; # this must not taint
150 check_taint_not 61, $_;
151 check_taint_not 62, $&;
152 check_taint_not 63, $`;
153 check_taint_not 64, $';
154 check_taint_not 65, $+;
155 check_taint_not 66, $1;
156 check_taint_not 67, $2;
158 $b = $a; # untaint $b
160 ($b = $a) =~ s/\w/$&/;
161 check_taint 68, $b; # $b should be tainted.
162 check_taint_not 69, $a; # $a should be not.
164 $_ = $a; # untaint $_
166 s/(\w)/\l$1/; # this must taint
173 check_taint_not 76, $2;
175 $_ = $a; # untaint $_
177 s/(\w)/\L$1/; # this must taint
184 check_taint_not 83, $2;
186 $_ = $a; # untaint $_
188 s/(\w)/\u$1/; # this must taint
195 check_taint_not 90, $2;
197 $_ = $a; # untaint $_
199 s/(\w)/\U$1/; # this must taint
206 check_taint_not 97, $2;
208 # After all this tainting $a should be cool.
210 check_taint_not 98, $a;
212 # I think we've seen quite enough of taint.
213 # Let us do some *real* locale work now,
214 # unless setlocale() is missing (i.e. minitest).
216 exit unless $have_setlocale;
219 sort grep /\w/, map { chr } 0..255
222 sub locatelocale ($$@) {
223 my ($lcall, $alnum, @try) = @_;
228 local $^W = 0; # suppress "Subroutine LC_ALL redefined"
229 if (setlocale(&LC_ALL, $_)) {
236 @$alnum = () unless (defined $$lcall);
239 # Find some default locale
241 locatelocale(\$Locale, \@Locale, qw(C POSIX));
243 # Find some English locale
245 locatelocale(\$English, \@English,
246 qw(en_US.ISO8859-1 en_GB.ISO8859-1
247 en en_US en_UK en_IE en_CA en_AU en_NZ
248 english english.iso88591
249 american american.iso88591
250 british british.iso88591
253 # Find some German locale
255 locatelocale(\$German, \@German,
256 qw(de_DE.ISO8859-1 de_AT.ISO8859-1 de_CH.ISO8859-1
258 german german.iso88591));
260 # Find some French locale
262 locatelocale(\$French, \@French,
263 qw(fr_FR.ISO8859-1 fr_BE.ISO8859-1 fr_CA.ISO8859-1 fr_CH.ISO8859-1
264 fr fr_FR fr_BE fr_CA fr_CH
265 french french.iso88591));
267 # Find some Spanish locale
269 locatelocale(\$Spanish, \@Spanish,
270 qw(es_AR.ISO8859-1 es_BO.ISO8859-1 es_CL.ISO8859-1
271 es_CO.ISO8859-1 es_CR.ISO8859-1 es_EC.ISO8859-1
272 es_ES.ISO8859-1 es_GT.ISO8859-1 es_MX.ISO8859-1
273 es_NI.ISO8859-1 es_PA.ISO8859-1 es_PE.ISO8859-1
274 es_PY.ISO8859-1 es_SV.ISO8859-1 es_UY.ISO8859-1 es_VE.ISO8859-1
279 es_PY es_SV es_UY es_VE
280 spanish spanish.iso88591));
282 # Select the largest of the alpha(num)bets.
284 ($Locale, @Locale) = ($English, @English)
285 if (length(@English) > length(@Locale));
286 ($Locale, @Locale) = ($German, @German)
287 if (length(@German) > length(@Locale));
288 ($Locale, @Locale) = ($French, @French)
289 if (length(@French) > length(@Locale));
290 ($Locale, @Locale) = ($Spanish, @Spanish)
291 if (length(@Spanish) > length(@Locale));
293 print "# Locale = $Locale\n";
294 print "# Alnum_ = @Locale\n";
298 setlocale(&LC_ALL, $Locale);
309 # Sieve the uppercase and the lowercase.
312 if (/[^\d_]/) { # skip digits and the _
321 # Find the alphabets that are not alphabets in the default locale.
326 for (keys %UPPER, keys %lower) {
327 push(@Neoalpha, $_) if (/\W/);
331 @Neoalpha = sort @Neoalpha;
336 my $word = join('', @Neoalpha);
340 print 'not ' if ($1 ne $word);
344 # Find places where the collation order differs from the default locale.
346 print "# testing 100\n";
353 @k = sort (keys %UPPER, keys %lower);
356 for ($i = 0; $i < @k; $i++) {
357 for ($j = $i + 1; $j < @k; $j++) {
358 if ($iLocale{$k[$j]} < $iLocale{$k[$i]}) {
359 push(@d, [$k[$j], $k[$i]]);
364 # Cross-check those places.
369 print "# failed 100 at:\n";
370 print "# i = $i, j = $j, i ",
371 $i le $j ? 'le' : 'gt', " j\n";
379 # Cross-check whole character set.
381 print "# testing 101\n";
382 for (map { chr } 0..255) {
383 if (/\w/ and /\W/) { print 'not '; last }
384 if (/\d/ and /\D/) { print 'not '; last }
385 if (/\s/ and /\S/) { print 'not '; last }
386 if (/\w/ and /\D/ and not /_/ and
387 not (exists $UPPER{$_} or exists $lower{$_})) {
388 print "# failed 101 at:\n";
389 print "# ", ord($_), " '$_'\n";
396 # The @Locale should be internally consistent.
398 print "# testing 102\n";
400 my ($from, $to, $lesser, $greater, @test, %test, $test);
404 $from = int(($_*@Locale)/10);
405 $to = $from + int(@Locale/10);
406 $to = $#Locale if ($to > $#Locale);
407 $lesser = join('', @Locale[$from..$to]);
408 # Select a slice one character on.
410 $to = $#Locale if ($to > $#Locale);
411 $greater = join('', @Locale[$from..$to]);
414 'not ($lesser lt $greater)', # 0
415 'not ($lesser le $greater)', # 1
416 'not ($lesser ne $greater)', # 2
417 ' ($lesser eq $greater)', # 3
418 ' ($lesser ge $greater)', # 4
419 ' ($lesser gt $greater)', # 5
420 ' ($greater lt $lesser )', # 6
421 ' ($greater le $lesser )', # 7
422 'not ($greater ne $lesser )', # 8
423 ' ($greater eq $lesser )', # 9
424 'not ($greater ge $lesser )', # 10
425 'not ($greater gt $lesser )', # 11
426 # Well, these two are sort of redundant
427 # because @Locale was derived using cmp.
428 'not (($lesser cmp $greater) == -1)', # 12
429 'not (($greater cmp $lesser ) == 1)' # 13
431 @test{@test} = 0 x @test;
433 for my $ti (@test) { $test{$ti} = eval $ti ; $test ||= $test{$ti} }
435 print "# failed 102 at:\n";
436 print "# lesser = '$lesser'\n";
437 print "# greater = '$greater'\n";
438 print "# (greater) from = $from, to = $to\n";
440 printf("# %-40s %-4s", $ti,
441 $test{$ti} ? 'FAIL' : 'ok');
442 if ($ti =~ /\(\.*(\$.+ +cmp +\$[^\)]+)\.*\)/) {
443 printf("(%s == %4d)", $1, eval $1);