blob: e91c24e9e82ed18234ec4d16ada06d3b2b20d3f9 [file] [log] [blame]
/-- These tests for Unicode property support test PCRE's API and show some of
the compiled code. They are not Perl-compatible. --/
/[\p{L}]/DZ
/[\p{^L}]/DZ
/[\P{L}]/DZ
/[\P{^L}]/DZ
/[abc\p{L}\x{0660}]/8DZ
/[\p{Nd}]/8DZ
1234
/[\p{Nd}+-]+/8DZ
1234
12-34
12+\x{661}-34
** Failers
abcd
/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iDZ
/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8DZ
/AB\x{1fb0}/8DZ
/AB\x{1fb0}/8DZi
/[\x{105}-\x{109}]/8iDZ
\x{104}
\x{105}
\x{109}
** Failers
\x{100}
\x{10a}
/[z-\x{100}]/8iDZ
Z
z
\x{39c}
\x{178}
|
\x{80}
\x{ff}
\x{100}
\x{101}
** Failers
\x{102}
Y
y
/[z-\x{100}]/8DZi
/(?:[\PPa*]*){8,}/
/[\P{Any}]/BZ
/[\P{Any}\E]/BZ
/(\P{Yi}+\277)/
/(\P{Yi}+\277)?/
/(?<=\P{Yi}{3}A)X/
/\p{Yi}+(\P{Yi}+)(?1)/
/(\P{Yi}{2}\277)?/
/[\P{Yi}A]/
/[\P{Yi}\P{Yi}\P{Yi}A]/
/[^\P{Yi}A]/
/[^\P{Yi}\P{Yi}\P{Yi}A]/
/(\P{Yi}*\277)*/
/(\P{Yi}*?\277)*/
/(\p{Yi}*+\277)*/
/(\P{Yi}?\277)*/
/(\P{Yi}??\277)*/
/(\p{Yi}?+\277)*/
/(\P{Yi}{0,3}\277)*/
/(\P{Yi}{0,3}?\277)*/
/(\p{Yi}{0,3}+\277)*/
/\p{Zl}{2,3}+/8BZ
\xe2\x80\xa8\xe2\x80\xa8
\x{2028}\x{2028}\x{2028}
/\p{Zl}/8BZ
/\p{Lu}{3}+/8BZ
/\pL{2}+/8BZ
/\p{Cc}{2}+/8BZ
/^\p{Cs}/8
\?\x{dfff}
** Failers
\x{09f}
/^\p{Sc}+/8
$\x{a2}\x{a3}\x{a4}\x{a5}\x{a6}
\x{9f2}
** Failers
X
\x{2c2}
/^\p{Zs}/8
\ \
\x{a0}
\x{1680}
\x{180e}
\x{2000}
\x{2001}
** Failers
\x{2028}
\x{200d}
/-- These four are here rather than in test 6 because Perl has problems with
the negative versions of the properties. --/
/\p{^Lu}/8i
1234
** Failers
ABC
/\P{Lu}/8i
1234
** Failers
ABC
/\p{Ll}/8i
a
Az
** Failers
ABC
/\p{Lu}/8i
A
a\x{10a0}B
** Failers
a
\x{1d00}
/[\x{c0}\x{391}]/8i
\x{c0}
\x{e0}
/-- The next two are special cases where the lengths of the different cases of
the same character differ. The first went wrong with heap frame storage; the
second was broken in all cases. --/
/^\x{023a}+?(\x{0130}+)/8i
\x{023a}\x{2c65}\x{0130}
/^\x{023a}+([^X])/8i
\x{023a}\x{2c65}X
/\x{c0}+\x{116}+/8i
\x{c0}\x{e0}\x{116}\x{117}
/[\x{c0}\x{116}]+/8i
\x{c0}\x{e0}\x{116}\x{117}
/(\x{de})\1/8i
\x{de}\x{de}
\x{de}\x{fe}
\x{fe}\x{fe}
\x{fe}\x{de}
/^\x{c0}$/8i
\x{c0}
\x{e0}
/^\x{e0}$/8i
\x{c0}
\x{e0}
/-- The next two should be Perl-compatible, but it fails to match \x{e0}. PCRE
will match it only with UCP support, because without that it has no notion
of case for anything other than the ASCII letters. --/
/((?i)[\x{c0}])/8
\x{c0}
\x{e0}
/(?i:[\x{c0}])/8
\x{c0}
\x{e0}
/-- This should be Perl-compatible but Perl 5.11 gets \x{300} wrong. --/8
/^\X/8
A
A\x{300}BC
A\x{300}\x{301}\x{302}BC
*** Failers
\x{300}
/-- These are PCRE's extra properties to help with Unicodizing \d etc. --/
/^\p{Xan}/8
ABCD
1234
\x{6ca}
\x{a6c}
\x{10a7}
** Failers
_ABC
/^\p{Xan}+/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
** Failers
_ABC
/^\p{Xan}+?/8
\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xan}*/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xan}{2,9}/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xan}{2,9}?/8
\x{6ca}\x{a6c}\x{10a7}_
/^[\p{Xan}]/8
ABCD1234_
1234abcd_
\x{6ca}
\x{a6c}
\x{10a7}
** Failers
_ABC
/^[\p{Xan}]+/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
** Failers
_ABC
/^>\p{Xsp}/8
>\x{1680}\x{2028}\x{0b}
>\x{a0}
** Failers
\x{0b}
/^>\p{Xsp}+/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xsp}+?/8
>\x{1680}\x{2028}\x{0b}
/^>\p{Xsp}*/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xsp}{2,9}/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xsp}{2,9}?/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>[\p{Xsp}]/8
>\x{2028}\x{0b}
/^>[\p{Xsp}]+/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xps}/8
>\x{1680}\x{2028}\x{0b}
>\x{a0}
** Failers
\x{0b}
/^>\p{Xps}+/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xps}+?/8
>\x{1680}\x{2028}\x{0b}
/^>\p{Xps}*/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xps}{2,9}/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>\p{Xps}{2,9}?/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^>[\p{Xps}]/8
>\x{2028}\x{0b}
/^>[\p{Xps}]+/8
> \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
/^\p{Xwd}/8
ABCD
1234
\x{6ca}
\x{a6c}
\x{10a7}
_ABC
** Failers
[]
/^\p{Xwd}+/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xwd}+?/8
\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xwd}*/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
/^\p{Xwd}{2,9}/8
A_B12\x{6ca}\x{a6c}\x{10a7}
/^\p{Xwd}{2,9}?/8
\x{6ca}\x{a6c}\x{10a7}_
/^[\p{Xwd}]/8
ABCD1234_
1234abcd_
\x{6ca}
\x{a6c}
\x{10a7}
_ABC
** Failers
[]
/^[\p{Xwd}]+/8
ABCD1234\x{6ca}\x{a6c}\x{10a7}_
/-- A check not in UTF-8 mode --/
/^[\p{Xwd}]+/
ABCD1234_
/-- Some negative checks --/
/^[\P{Xwd}]+/8
!.+\x{019}\x{35a}AB
/^[\p{^Xwd}]+/8
!.+\x{019}\x{35a}AB
/[\D]/WBZ8
1\x{3c8}2
/[\d]/WBZ8
>\x{6f4}<
/[\S]/WBZ8
\x{1680}\x{6f4}\x{1680}
/[\s]/WBZ8
>\x{1680}<
/[\W]/WBZ8
A\x{1712}B
/[\w]/WBZ8
>\x{1723}<
/\D/WBZ8
1\x{3c8}2
/\d/WBZ8
>\x{6f4}<
/\S/WBZ8
\x{1680}\x{6f4}\x{1680}
/\s/WBZ8
>\x{1680}>
/\W/WBZ8
A\x{1712}B
/\w/WBZ8
>\x{1723}<
/[[:alpha:]]/WBZ
/[[:lower:]]/WBZ
/[[:upper:]]/WBZ
/[[:alnum:]]/WBZ
/[[:ascii:]]/WBZ
/[[:blank:]]/WBZ
/[[:cntrl:]]/WBZ
/[[:digit:]]/WBZ
/[[:graph:]]/WBZ
/[[:print:]]/WBZ
/[[:punct:]]/WBZ
/[[:space:]]/WBZ
/[[:word:]]/WBZ
/[[:xdigit:]]/WBZ
/-- Unicode properties for \b abd \B --/
/\b...\B/8W
abc_
\x{37e}abc\x{376}
\x{37e}\x{376}\x{371}\x{393}\x{394}
!\x{c0}++\x{c1}\x{c2}
!\x{c0}+++++
/-- Without PCRE_UCP, non-ASCII always fail, even if < 256 --/
/\b...\B/8
abc_
** Failers
\x{37e}abc\x{376}
\x{37e}\x{376}\x{371}\x{393}\x{394}
!\x{c0}++\x{c1}\x{c2}
!\x{c0}+++++
/-- With PCRE_UCP, non-UTF8 chars that are < 256 still check properties --/
/\b...\B/W
abc_
!\x{c0}++\x{c1}\x{c2}
!\x{c0}+++++
/-- POSIX interface --/
/\w/P
+++\x{c2}
/\w/WP
+++\x{c2}
/-- Some of these are silly, but they check various combinations --/
/[[:^alpha:][:^cntrl:]]+/8WBZ
123
abc
/[[:^cntrl:][:^alpha:]]+/8WBZ
123
abc
/[[:alpha:]]+/8WBZ
abc
/[[:^alpha:]\S]+/8WBZ
123
abc
/[^\d]+/8WBZ
abc123
abc\x{123}
\x{660}abc
/\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}/8iSI
\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
\x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
/\p{Lu}+9\p{Lu}+B\p{Lu}+b/BZ
/\p{^Lu}+9\p{^Lu}+B\p{^Lu}+b/BZ
/\P{Lu}+9\P{Lu}+B\P{Lu}+b/BZ
/\p{Han}+X\p{Greek}+\x{370}/BZ8
/\p{Xan}+!\p{Xan}+A/BZ
/\p{Xsp}+!\p{Xsp}\t/BZ
/\p{Xps}+!\p{Xps}\t/BZ
/\p{Xwd}+!\p{Xwd}_/BZ
/A+\p{N}A+\dB+\p{N}*B+\d*/WBZ
/-- These behaved oddly in Perl, so they are kept in this test --/
/(\x{23a}\x{23a}\x{23a})?\1/8i
\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
/(ȺȺȺ)?\1/8i
ȺȺȺⱥⱥ
/(\x{23a}\x{23a}\x{23a})?\1/8i
\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
/(ȺȺȺ)?\1/8i
ȺȺȺⱥⱥⱥ
/(\x{23a}\x{23a}\x{23a})\1/8i
\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
/(ȺȺȺ)\1/8i
ȺȺȺⱥⱥ
/(\x{23a}\x{23a}\x{23a})\1/8i
\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
/(ȺȺȺ)\1/8i
ȺȺȺⱥⱥⱥ
/(\x{2c65}\x{2c65})\1/8i
\x{2c65}\x{2c65}\x{23a}\x{23a}
/(ⱥⱥ)\1/8i
ⱥⱥȺȺ
/(\x{23a}\x{23a}\x{23a})\1Y/8i
X\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}YZ
/(\x{2c65}\x{2c65})\1Y/8i
X\x{2c65}\x{2c65}\x{23a}\x{23a}YZ
/-- --/
/-- These scripts weren't yet in Perl when I added Unicode 6.0.0 to PCRE --/
/^[\p{Batak}]/8
\x{1bc0}
\x{1bff}
** Failers
\x{1bf4}
/^[\p{Brahmi}]/8
\x{11000}
\x{1106f}
** Failers
\x{1104e}
/^[\p{Mandaic}]/8
\x{840}
\x{85e}
** Failers
\x{85c}
\x{85d}
/-- --/
/(\X*)(.)/s8
A\x{300}
/^S(\X*)e(\X*)$/8
Stéréo
/^\X/8
́réo
/^a\X41z/<JS>
aX41z
*** Failers
aAz
/(?<=ab\Cde)X/8
/-- End of testinput13 --/