feedc0de/boost_regex
1
0
Fork 0
forked from boostorg/regex
Code Pull Requests Activity
Files
19dab6c60a289e2f4422dadd3f8889f1cb5339b4
boost_regex/test/regress/tests.txt
John Maddock ed21a5183b reworked regex directory structure to match boost requirements. 
[SVN r9752]
2001-04-10 11:44:47 +00:00

874 lines
19 KiB
Plaintext
Raw Blame History

;
;
; this file contains a script of tests to run through regress.exe
;
; comments start with a semicolon and proceed to the end of the line
;
; changes to regular expression compile flags start with a "-" as the first
; non-whitespace character and consist of a list of the printable names
; of the flags, for example "match_default"
;
; Other lines contain a test to perform using the current flag status
; the first token contains the expression to compile, the second the string
; to match it against. If the second string is "!" then the expression should
; not compile, that is the first string is an invalid regular expression.
; This is then followed by a list of integers that specify what should match,
; each pair represents the starting and ending positions of a subexpression
; starting with the zeroth subexpression (the whole match).
; A value of -1 indicates that the subexpression should not take part in the
; match at all, if the first value is -1 then no part of the expression should
; match the string.
;
- match_default normal REG_EXTENDED
;
; try some really simple literals:
a a 0 1
Z Z 0 1
Z aaa -1 -1
Z xxxxZZxxx 4 5
; and some simple brackets:
(a) zzzaazz 3 4 3 4
() zzz 0 0 0 0
() "" 0 0 0 0
( !
) !
(aa !
aa) !
a b -1 -1
\(\) () 0 2
\(a\) (a) 0 3
\() !
(\) !
p(a)rameter ABCparameterXYZ 3 12 4 5
[pq](a)rameter ABCparameterXYZ 3 12 4 5
; now try escaped brackets:
- match_default bk_parens REG_BASIC
\(a\) zzzaazz 3 4 3 4
\(\) zzz 0 0 0 0
\(\) "" 0 0 0 0
\( !
\) !
\(aa !
aa\) !
() () 0 2
(a) (a) 0 3
(\) !
\() !
; now move on to "." wildcards
- match_default normal REG_EXTENDED REG_STARTEND
. a 0 1
. \n 0 1
. \r 0 1
. \0 0 1
- match_default normal match_not_dot_newline REG_EXTENDED REG_STARTEND REG_NEWLINE
. a 0 1
. \n -1 -1
. \r -1 -1
. \0 0 1
- match_default normal match_not_dot_null match_not_dot_newline REG_EXTENDED REG_STARTEND REG_NEWLINE
. \n -1 -1
. \r -1 -1
; this *WILL* produce an error from the POSIX API functions:
- match_default normal match_not_dot_null match_not_dot_newline REG_EXTENDED REG_STARTEND REG_NEWLINE REG_NO_POSIX_TEST
. \0 -1 -1
;
; now move on to the repetion ops,
; starting with operator *
- match_default normal REG_EXTENDED
a* b 0 0
ab* a 0 1
ab* ab 0 2
ab* sssabbbbbbsss 3 10
ab*c* a 0 1
ab*c* abbb 0 4
ab*c* accc 0 4
ab*c* abbcc 0 5
*a !
\<* !
\>* !
\n* \n\n 0 2
\** ** 0 2
\* * 0 1
; now try operator +
ab+ a -1 -1
ab+ ab 0 2
ab+ sssabbbbbbsss 3 10
ab+c+ a -1 -1
ab+c+ abbb -1 -1
ab+c+ accc -1 -1
ab+c+ abbcc 0 5
+a !
\<+ !
\>+ !
\n+ \n\n 0 2
\+ + 0 1
\+ ++ 0 1
\++ ++ 0 2
- match_default normal bk_plus_qm REG_EXTENDED REG_NO_POSIX_TEST
+ + 0 1
\+ !
a\+ aa 0 2
; now try operator ?
- match_default normal REG_EXTENDED
a? b 0 0
ab? a 0 1
ab? ab 0 2
ab? sssabbbbbbsss 3 5
ab?c? a 0 1
ab?c? abbb 0 2
ab?c? accc 0 2
ab?c? abcc 0 3
?a !
\<? !
\>? !
\n? \n\n 0 1
\? ? 0 1
\? ?? 0 1
\?? ?? 0 1
- match_default normal bk_plus_qm REG_EXTENDED REG_NO_POSIX_TEST
? ? 0 1
\? !
a\? aa 0 1
a\? b 0 0
- match_default normal limited_ops
a? a? 0 2
a+ a+ 0 2
a\? a? 0 2
a\+ a+ 0 2
; now try operator {}
- match_default normal REG_EXTENDED
a{2} a -1 -1
a{2} aa 0 2
a{2} aaa 0 2
a{2,} a -1 -1
a{2,} aa 0 2
a{2,} aaaaa 0 5
a{2,4} a -1 -1
a{2,4} aa 0 2
a{2,4} aaa 0 3
a{2,4} aaaa 0 4
a{2,4} aaaaa 0 4
; spaces are now allowed inside {}
"a{ 2 , 4 }" aaaaa 0 4
a{} !
"a{ }" !
a{2 !
a} !
\{\} {} 0 2
- match_default normal bk_braces
a\{2\} a -1 -1
a\{2\} aa 0 2
a\{2\} aaa 0 2
a\{2,\} a -1 -1
a\{2,\} aa 0 2
a\{2,\} aaaaa 0 5
a\{2,4\} a -1 -1
a\{2,4\} aa 0 2
a\{2,4\} aaa 0 3
a\{2,4\} aaaa 0 4
a\{2,4\} aaaaa 0 4
"a\{ 2 , 4 \}" aaaaa 0 4
{} {} 0 2
; now test the alternation operator |
- match_default normal REG_EXTENDED
a|b a 0 1
a|b b 0 1
a(b|c) ab 0 2 1 2
a(b|c) ac 0 2 1 2
a(b|c) ad -1 -1 -1 -1
|c !
c| !
(|) !
(a|) !
(|a) !
a\| a| 0 2
- match_default normal limited_ops
a| a| 0 2
a\| a| 0 2
| | 0 1
- match_default normal bk_vbar REG_NO_POSIX_TEST
a| a| 0 2
a\|b a 0 1
a\|b b 0 1
; now test the set operator []
- match_default normal REG_EXTENDED
; try some literals first
[abc] a 0 1
[abc] b 0 1
[abc] c 0 1
[abc] d -1 -1
[^bcd] a 0 1
[^bcd] b -1 -1
[^bcd] d -1 -1
[^bcd] e 0 1
a[b]c abc 0 3
a[ab]c abc 0 3
a[^ab]c adc 0 3
a[]b]c a]c 0 3
a[[b]c a[c 0 3
a[-b]c a-c 0 3
a[^]b]c adc 0 3
a[^-b]c adc 0 3
a[b-]c a-c 0 3
a[b !
a[] !
; then some ranges
[b-e] a -1 -1
[b-e] b 0 1
[b-e] e 0 1
[b-e] f -1 -1
[^b-e] a 0 1
[^b-e] b -1 -1
[^b-e] e -1 -1
[^b-e] f 0 1
a[1-3]c a2c 0 3
a[3-1]c !
a[1-3-5]c !
a[1- !
; and some classes
a[[:alpha:]]c abc 0 3
a[[:unknown:]]c !
a[[: !
a[[:alpha !
a[[:alpha:] !
a[[:alpha,:] !
a[[:]:]]b !
a[[:-:]]b !
a[[:alph:]] !
a[[:alphabet:]] !
[[:alnum:]]+ -%@a0X- 3 6
[[:alpha:]]+ -%@aX0- 3 5
[[:blank:]]+ "a \tb" 1 4
[[:cntrl:]]+ a\n\tb 1 3
[[:digit:]]+ a019b 1 4
[[:graph:]]+ " a%b " 1 4
[[:lower:]]+ AabC 1 3
; This test fails with STLPort, disable for now as this is a corner case anyway...
;[[:print:]]+ "\na b\n" 1 4
[[:punct:]]+ " %-&\t" 1 4
[[:space:]]+ "a \n\t\rb" 1 5
[[:upper:]]+ aBCd 1 3
[[:xdigit:]]+ p0f3Cx 1 5
; now test flag settings:
- escape_in_lists REG_NO_POSIX_TEST
[\n] \n 0 1
- REG_NO_POSIX_TEST
[\n] \n -1 -1
[\n] \\ 0 1
[[:class:] : 0 1
[[:class:] [ 0 1
[[:class:] c 0 1
; line anchors
- match_default normal REG_EXTENDED
^ab ab 0 2
^ab xxabxx -1 -1
^ab xx\nabzz 3 5
ab$ ab 0 2
ab$ abxx -1 -1
ab$ ab\nzz 0 2
- match_default match_not_bol match_not_eol normal REG_EXTENDED REG_NOTBOL REG_NOTEOL
^ab ab -1 -1
^ab xxabxx -1 -1
^ab xx\nabzz 3 5
ab$ ab -1 -1
ab$ abxx -1 -1
ab$ ab\nzz 0 2
; back references
- match_default normal REG_EXTENDED
a(b)\2c !
a(b\1)c !
a(b*)c\1d abbcbbd 0 7 1 3
a(b*)c\1d abbcbd -1 -1
a(b*)c\1d abbcbbbd -1 -1
^(.)\1 abc -1 -1
a([bc])\1d abcdabbd 4 8 5 6
; strictly speaking this is at best ambiguous, at worst wrong, this is what most
; re implimentations will match though.
a(([bc])\2)*d abbccd 0 6 3 5 3 4
a(([bc])\2)*d abbcbd -1 -1
a((b)*\2)*d abbbd 0 5 1 4 2 3
(ab*)[ab]*\1 ababaaa 0 7 0 1
(a)\1bcd aabcd 0 5 0 1
(a)\1bc*d aabcd 0 5 0 1
(a)\1bc*d aabd 0 4 0 1
(a)\1bc*d aabcccd 0 7 0 1
(a)\1bc*[ce]d aabcccd 0 7 0 1
^(a)\1b(c)*cd$ aabcccd 0 7 0 1 4 5
;
; characters by code:
- match_default normal REG_EXTENDED REG_STARTEND
\0101 A 0 1
\00 \0 0 1
\0 \0 0 1
\0172 z 0 1
;
; word operators:
\w a 0 1
\w z 0 1
\w A 0 1
\w Z 0 1
\w _ 0 1
\w } -1 -1
\w ` -1 -1
\w [ -1 -1
\w @ -1 -1
; non-word:
\W a -1 -1
\W z -1 -1
\W A -1 -1
\W Z -1 -1
\W _ -1 -1
\W } 0 1
\W ` 0 1
\W [ 0 1
\W @ 0 1
; word start:
\<abcd " abcd" 2 6
\<ab cab -1 -1
\<ab "\nab" 1 3
\<tag ::tag 2 5
;word end:
abc\> abc 0 3
abc\> abcd -1 -1
abc\> abc\n 0 3
abc\> abc:: 0 3
; word boundary:
\babcd " abcd" 2 6
\bab cab -1 -1
\bab "\nab" 1 3
\btag ::tag 2 5
abc\b abc 0 3
abc\b abcd -1 -1
abc\b abc\n 0 3
abc\b abc:: 0 3
; within word:
\B ab 1 1
a\Bb ab 0 2
a\B ab 0 1
a\B a -1 -1
a\B "a " -1 -1
;
; buffer operators:
\`abc abc 0 3
\`abc \nabc -1 -1
\`abc " abc" -1 -1
abc\' abc 0 3
abc\' abc\n -1 -1
abc\' "abc " -1 -1
;
; extra escape sequences:
\a \a 0 1
\f \f 0 1
\n \n 0 1
\r \r 0 1
\t \t 0 1
\v \v 0 1
;
; now follows various complex expressions designed to try and bust the matcher:
a(((b)))c abc 0 3 1 2 1 2 1 2
a(b|(c))d abd 0 3 1 2 -1 -1
a(b|(c))d acd 0 3 1 2 1 2
a(b*|c)d abbd 0 4 1 3
; just gotta have one DFA-buster, of course
a[ab]{20} aaaaabaaaabaaaabaaaab 0 21
; and an inline expansion in case somebody gets tricky
a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab] aaaaabaaaabaaaabaaaab 0 21
; and in case somebody just slips in an NFA...
a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab](wee|week)(knights|night) aaaaabaaaabaaaabaaaabweeknights 0 31 21 24 24 31
; one really big one
1234567890123456789012345678901234567890123456789012345678901234567890 a1234567890123456789012345678901234567890123456789012345678901234567890b 1 71
; fish for problems as brackets go past 8
[ab][cd][ef][gh][ij][kl][mn] xacegikmoq 1 8
[ab][cd][ef][gh][ij][kl][mn][op] xacegikmoq 1 9
[ab][cd][ef][gh][ij][kl][mn][op][qr] xacegikmoqy 1 10
[ab][cd][ef][gh][ij][kl][mn][op][q] xacegikmoqy 1 10
; and as parenthesis go past 9:
(a)(b)(c)(d)(e)(f)(g)(h) zabcdefghi 1 9 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9
(a)(b)(c)(d)(e)(f)(g)(h)(i) zabcdefghij 1 10 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10
(a)(b)(c)(d)(e)(f)(g)(h)(i)(j) zabcdefghijk 1 11 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11
(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k) zabcdefghijkl 1 12 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12
(a)d|(b)c abc 1 3 -1 -1 1 2
"_+((www)|(ftp)|(mailto)):_*" "_wwwnocolon _mailto:" 12 20 13 19 -1 -1 -1 -1 13 19
; subtleties of matching
a(b)?c\1d acd 0 3 -1 -1
a(b?c)+d accd 0 4 2 3
(wee|week)(knights|night) weeknights 0 10 0 3 3 10
.* abc 0 3
a(b|(c))d abd 0 3 1 2 -1 -1
a(b|(c))d acd 0 3 1 2 1 2
a(b*|c|e)d abbd 0 4 1 3
a(b*|c|e)d acd 0 3 1 2
a(b*|c|e)d ad 0 2 1 1
a(b?)c abc 0 3 1 2
a(b?)c ac 0 2 1 1
a(b+)c abc 0 3 1 2
a(b+)c abbbc 0 5 1 4
a(b*)c ac 0 2 1 1
(a|ab)(bc([de]+)f|cde) abcdef 0 6 0 1 1 6 3 5
a([bc]?)c abc 0 3 1 2
a([bc]?)c ac 0 2 1 1
a([bc]+)c abc 0 3 1 2
a([bc]+)c abcc 0 4 1 3
a([bc]+)bc abcbc 0 5 1 3
a(bb+|b)b abb 0 3 1 2
a(bbb+|bb+|b)b abb 0 3 1 2
a(bbb+|bb+|b)b abbb 0 4 1 3
a(bbb+|bb+|b)bb abbb 0 4 1 2
(.*).* abcdef 0 6 0 6
(a*)* bc 0 0 0 0
; do we get the right subexpression when it is used more than once?
a(b|c)*d ad 0 2 -1 -1
a(b|c)*d abcd 0 4 2 3
a(b|c)+d abd 0 3 1 2
a(b|c)+d abcd 0 4 2 3
a(b|c?)+d ad 0 2 1 1
a(b|c?)+d abcd 0 4 2 3
a(b|c){0,0}d ad 0 2 -1 -1
a(b|c){0,1}d ad 0 2 -1 -1
a(b|c){0,1}d abd 0 3 1 2
a(b|c){0,2}d ad 0 2 -1 -1
a(b|c){0,2}d abcd 0 4 2 3
a(b|c){0,}d ad 0 2 -1 -1
a(b|c){0,}d abcd 0 4 2 3
a(b|c){1,1}d abd 0 3 1 2
a(b|c){1,2}d abd 0 3 1 2
a(b|c){1,2}d abcd 0 4 2 3
a(b|c){1,}d abd 0 3 1 2
a(b|c){1,}d abcd 0 4 2 3
a(b|c){2,2}d acbd 0 4 2 3
a(b|c){2,2}d abcd 0 4 2 3
a(b|c){2,4}d abcd 0 4 2 3
a(b|c){2,4}d abcbd 0 5 3 4
a(b|c){2,4}d abcbcd 0 6 4 5
a(b|c){2,}d abcd 0 4 2 3
a(b|c){2,}d abcbd 0 5 3 4
a(b+|((c)*))+d abd 0 3 1 2 -1 -1 -1 -1
a(b+|((c)*))+d abcd 0 4 2 3 2 3 2 3
- match_default normal REG_EXTENDED REG_STARTEND REG_NOSPEC literal
\**?/{} \\**?/{} 0 7
- match_default normal REG_EXTENDED REG_NO_POSIX_TEST ; we disable POSIX testing because it can't handle escapes in sets
; try to match C++ syntax elements:
; line comment:
//[^\n]* "++i //here is a line comment\n" 4 28
; block comment:
/\*([^*]|\*+[^*/])*\*+/ "/* here is a block comment */" 0 29 26 27
/\*([^*]|\*+[^*/])*\*+/ "/**/" 0 4 -1 -1
/\*([^*]|\*+[^*/])*\*+/ "/***/" 0 5 -1 -1
/\*([^*]|\*+[^*/])*\*+/ "/****/" 0 6 -1 -1
/\*([^*]|\*+[^*/])*\*+/ "/*****/" 0 7 -1 -1
/\*([^*]|\*+[^*/])*\*+/ "/*****/*/" 0 7 -1 -1
; preprossor directives:
^[[:blank:]]*#([^\n]*\\[[:space:]]+)*[^\n]* "#define some_symbol" 0 19 -1 -1
^[[:blank:]]*#([^\n]*\\[[:space:]]+)*[^\n]* "#define some_symbol(x) #x" 0 25 -1 -1
^[[:blank:]]*#([^\n]*\\[[:space:]]+)*[^\n]* "#define some_symbol(x) \\ \r\n foo();\\\r\n printf(#x);" 0 53 28 42
; literals:
((0x[[:xdigit:]]+)|([[:digit:]]+))u?((int(8|16|32|64))|L)? 0xFF 0 4 0 4 0 4 -1 -1 -1 -1 -1 -1 -1 -1
((0x[[:xdigit:]]+)|([[:digit:]]+))u?((int(8|16|32|64))|L)? 35 0 2 0 2 -1 -1 0 2 -1 -1 -1 -1 -1 -1
((0x[[:xdigit:]]+)|([[:digit:]]+))u?((int(8|16|32|64))|L)? 0xFFu 0 5 0 4 0 4 -1 -1 -1 -1 -1 -1 -1 -1
((0x[[:xdigit:]]+)|([[:digit:]]+))u?((int(8|16|32|64))|L)? 0xFFL 0 5 0 4 0 4 -1 -1 4 5 -1 -1 -1 -1
((0x[[:xdigit:]]+)|([[:digit:]]+))u?((int(8|16|32|64))|L)? 0xFFFFFFFFFFFFFFFFuint64 0 24 0 18 0 18 -1 -1 19 24 19 24 22 24
; strings:
'([^\\']|\\.)*' '\\x3A' 0 6 4 5
'([^\\']|\\.)*' '\\'' 0 4 1 3
'([^\\']|\\.)*' '\\n' 0 4 1 3
; now try and test some unicode specific characters:
- match_default normal REG_EXTENDED REG_UNICODE_ONLY
[[:unicode:]]+ a\0300\0400z 1 3
; finally try some case insensitive matches:
- match_default normal REG_EXTENDED REG_ICASE
; upper and lower have no meaning here so they fail, however these
; may compile with other libraries...
;[[:lower:]] !
;[[:upper:]] !
0123456789@abcdefghijklmnopqrstuvwxyz\[\\\]\^_`ABCDEFGHIJKLMNOPQRSTUVWXYZ\{\|\} 0123456789@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]\^_`abcdefghijklmnopqrstuvwxyz\{\|\} 0 72
; known and suspected bugs:
- match_default normal REG_EXTENDED
\( ( 0 1
\) ) 0 1
\$ $ 0 1
\^ ^ 0 1
\. . 0 1
\* * 0 1
\+ + 0 1
\? ? 0 1
\[ [ 0 1
\] ] 0 1
\| | 0 1
\\ \\ 0 1
# # 0 1
\# # 0 1
a- a- 0 2
\- - 0 1
\{ { 0 1
\} } 0 1
0 0 0 1
1 1 0 1
9 9 0 1
b b 0 1
B B 0 1
< < 0 1
> > 0 1
w w 0 1
W W 0 1
` ` 0 1
' ' 0 1
\n \n 0 1
, , 0 1
a a 0 1
f f 0 1
n n 0 1
r r 0 1
t t 0 1
v v 0 1
c c 0 1
x x 0 1
: : 0 1
(\.[[:alnum:]]+){2} "w.a.b " 1 5 3 5
- match_default normal REG_EXTENDED REG_ICASE
a A 0 1
A a 0 1
[abc]+ abcABC 0 6
[ABC]+ abcABC 0 6
[a-z]+ abcABC 0 6
[A-Z]+ abzANZ 0 6
[a-Z]+ abzABZ 0 6
[A-z]+ abzABZ 0 6
[[:lower:]]+ abyzABYZ 0 8
[[:upper:]]+ abzABZ 0 6
[[:word:]]+ abcZZZ 0 6
[[:alpha:]]+ abyzABYZ 0 8
[[:alnum:]]+ 09abyzABYZ 0 10
; updated tests for version 2:
- match_default normal REG_EXTENDED
\x41 A 0 1
\xff \255 0 1
\xFF \255 0 1
- match_default normal REG_EXTENDED REG_NO_POSIX_TEST
\c@ \0 0 1
- match_default normal REG_EXTENDED
\cA \1 0 1
\cz \58 0 1
\c= !
\c? !
=: =: 0 2
; word start:
[[:<:]]abcd " abcd" 2 6
[[:<:]]ab cab -1 -1
[[:<:]]ab "\nab" 1 3
[[:<:]]tag ::tag 2 5
;word end:
abc[[:>:]] abc 0 3
abc[[:>:]] abcd -1 -1
abc[[:>:]] abc\n 0 3
abc[[:>:]] abc:: 0 3
; collating elements and rewritten set code:
- match_default normal REG_EXTENDED REG_STARTEND
[[.zero.]] 0 0 1
[[.one.]] 1 0 1
[[.two.]] 2 0 1
[[.three.]] 3 0 1
[[.a.]] baa 1 2
[[.right-curly-bracket.]] } 0 1
[[.NUL.]] \0 0 1
[[:<:]z] !
[a[:>:]] !
[[=a=]] a 0 1
[[=right-curly-bracket=]] } 0 1
- match_default normal REG_EXTENDED REG_STARTEND REG_ICASE
[[.A.]] A 0 1
[[.A.]] a 0 1
[[.A.]-b]+ AaBb 0 4
[A-[.b.]]+ AaBb 0 4
[[.a.]-B]+ AaBb 0 4
[a-[.B.]]+ AaBb 0 4
- match_default normal REG_EXTENDED REG_NO_POSIX_TEST
[\x61] a 0 1
[\x61-c]+ abcd 0 3
[a-\x63]+ abcd 0 3
- match_default normal REG_EXTENDED REG_STARTEND
[[.a.]-c]+ abcd 0 3
[a-[.c.]]+ abcd 0 3
[[:alpha:]-a] !
[a-[:alpha:]] !
; try mutli-character ligatures:
[[.ae.]] ae 0 2
[[.ae.]] aE -1 -1
[[.AE.]] AE 0 2
[[.Ae.]] Ae 0 2
[[.ae.]-b] a -1 -1
[[.ae.]-b] b 0 1
[[.ae.]-b] ae 0 2
[a-[.ae.]] a 0 1
[a-[.ae.]] b -1 -1
[a-[.ae.]] ae 0 2
- match_default normal REG_EXTENDED REG_STARTEND REG_ICASE
[[.ae.]] AE 0 2
[[.ae.]] Ae 0 2
[[.AE.]] Ae 0 2
[[.Ae.]] aE 0 2
[[.AE.]-B] a -1 -1
[[.Ae.]-b] b 0 1
[[.Ae.]-b] B 0 1
[[.ae.]-b] AE 0 2
- match_default normal REG_EXTENDED REG_STARTEND
;extended perl style escape sequences:
\e \27 0 1
\x1b \27 0 1
\x{1b} \27 0 1
\x{} !
\x{ !
\x} !
\x !
\x{yy !
\x{1b !
- match_default normal REG_EXTENDED REG_STARTEND REG_NO_POSIX_TEST
\l+ ABabcAB 2 5
[\l]+ ABabcAB 2 5
[a-\l] !
[\l-a] !
[\L] !
\L+ abABCab 2 5
\u+ abABCab 2 5
[\u]+ abABCab 2 5
[\U] !
\U+ ABabcAB 2 5
\d+ ab012ab 2 5
[\d]+ ab012ab 2 5
[\D] !
\D+ 01abc01 2 5
\s+ "ab ab" 2 5
[\s]+ "ab ab" 2 5
[\S] !
\S+ " abc " 2 5
- match_default normal REG_EXTENDED REG_STARTEND
\Qabc !
\Qabc\E abcd 0 3
\Qabc\Ed abcde 0 4
\Q+*?\\E +*?\\ 0 4
\C+ abcde 0 5
\X+ abcde 0 5
- match_default normal REG_EXTENDED REG_STARTEND REG_UNICODE_ONLY
\X+ a\768\769 0 3
\X+ \2309\2307 0 2 ;DEVANAGARI script
\X+ \2489\2494 0 2 ;BENGALI script
- match_default normal REG_EXTENDED REG_STARTEND
\Aabc abc 0 3
\Aabc aabc -1 -1
abc\z abc 0 3
abc\z abcd -1 -1
abc\Z abc\n\n 0 3
abc\Z abc 0 3
\Gabc abc 0 3
\Gabc dabcd -1 -1
a\Gbc abc -1 -1
a\Aab abc -1 -1
;
; now test grep,
; basically check all our restart types - line, word, etc
; checking each one for null and non-null matches.
;
- match_default normal REG_EXTENDED REG_STARTEND REG_GREP
a " a a a aa" 1 2 3 4 5 6 7 8 8 9
a+b+ "aabaabbb ab" 0 3 3 8 9 11
a(b*|c|e)d adabbdacd 0 2 2 6 6 9
a "\na\na\na\naa" 1 2 3 4 5 6 7 8 8 9
^ " \n\n \n\n\n" 0 0 4 4 5 5 8 8 9 9 10 10
^ab "ab \nab ab\n" 0 2 5 7
^[^\n]*\n " \n \n\n \n" 0 4 4 7 7 8 8 11
\<abc "abcabc abc\n\nabc" 0 3 7 10 12 15
\< " ab a aaa " 2 2 5 5 7 7
\<\w+\W+ " aa aa a " 1 5 5 9 9 11
\Aabc "abc abc" 0 3
\G\w+\W+ "abc abc a cbbb " 0 5 5 9 9 11 11 18
\Ga+b+ "aaababb abb" 0 4 4 7
abc abc 0 3
abc " abc abcabc " 1 4 5 8 8 11
\n\n " \n\n\n \n \n\n\n\n " 1 3 18 20 20 22
$ " \n\n \n\n\n" 3 3 4 4 7 7 8 8 9 9 10 10
\b " abb a abbb " 2 2 5 5 6 6 7 7 8 8 12 12
- match_default normal REG_EXTENDED REG_STARTEND REG_GREP REG_ICASE
A " a a a aa" 1 2 3 4 5 6 7 8 8 9
A+B+ "aabaabbb ab" 0 3 3 8 9 11
A(B*|c|e)D adabbdacd 0 2 2 6 6 9
A "\na\na\na\naa" 1 2 3 4 5 6 7 8 8 9
^aB "Ab \nab Ab\n" 0 2 5 7
\<abc "Abcabc aBc\n\nabc" 0 3 7 10 12 15
ABC abc 0 3
abc " ABC ABCABC " 1 4 5 8 8 11
;
; now test merge,
;
- match_default normal REG_EXTENDED REG_STARTEND REG_MERGE format_no_copy
; start by testing subs:
a+ "...aaa,,," $` "..."
a+ "...aaa,,," $' ",,,"
a+ "...aaa,,," $& "aaa"
a+ "...aaa,,," $0 aaa
a+ "...aaa,,," $1 ""
a+ "...aaa,,," $15 ""
(a+)b+ "...aaabbb,,," $1 aaa
[[:digit:]]* 123ab <$0> <123><><><>
[[:digit:]]* 123ab1 <$0> <123><><><1>
; and now escapes:
a+ "...aaa,,," $x "$x"
a+ "...aaa,,," \a "\a"
a+ "...aaa,,," \f "\f"
a+ "...aaa,,," \n "\n"
a+ "...aaa,,," \r "\r"
a+ "...aaa,,," \t "\t"
a+ "...aaa,,," \v "\v"
a+ "...aaa,,," \x21 "!"
a+ "...aaa,,," \x{21} "!"
a+ "...aaa,,," \c@ \0
a+ "...aaa,,," \e \27
a+ "...aaa,,," \0101 A
a+ "...aaa,,," (\0101) A
- match_default normal REG_EXTENDED REG_STARTEND REG_MERGE format_sed format_no_copy
(a+)(b+) ...aabb,, \0 aabb
(a+)(b+) ...aabb,, \1 aa
(a+)(b+) ...aabb,, \2 bb
(a+)(b+) ...aabb,, & aabb
(a+)(b+) ...aabb,, $ $
(a+)(b+) ...aabb,, $1 $1
(a+)(b+) ...aabb,, ()?: ()?:
(a+)(b+) ...aabb,, \\ \\
(a+)(b+) ...aabb,, \& &
- match_default normal REG_EXTENDED REG_STARTEND REG_MERGE format_perl format_no_copy
(a+)(b+) ...aabb,, $0 aabb
(a+)(b+) ...aabb,, $1 aa
(a+)(b+) ...aabb,, $2 bb
(a+)(b+) ...aabb,, $& aabb
(a+)(b+) ...aabb,, & &
(a+)(b+) ...aabb,, \0 \0
(a+)(b+) ...aabb,, ()?: ()?:
- match_default normal REG_EXTENDED REG_STARTEND REG_MERGE
; move to copying unmatched data:
a+ "...aaa,,," bbb "...bbb,,,"
a+(b+) "...aaabb,,," $1 "...bb,,,"
a+(b+) "...aaabb,,,ab*abbb?" $1 "...bb,,,b*bbb?"
(a+)|(b+) "...aaabb,,,ab*abbb?" (?1A)(?2B) "...AB,,,AB*AB?"
(a+)|(b+) "...aaabb,,,ab*abbb?" ?1A:B "...AB,,,AB*AB?"
(a+)|(b+) "...aaabb,,,ab*abbb?" (?1A:B)C "...ACBC,,,ACBC*ACBC?"
(a+)|(b+) "...aaabb,,,ab*abbb?" ?1:B "...B,,,B*B?"
;
; changes to newline handling with 2.11:
;
- match_default normal REG_EXTENDED REG_STARTEND REG_GREP
^. " \n \r\n " 0 1 3 4 7 8
.$ " \n \r\n " 1 2 4 5 8 9
- match_default normal REG_EXTENDED REG_STARTEND REG_GREP REG_UNICODE_ONLY
^. " \8232 \8233 " 0 1 3 4 5 6
.$ " \8232 \8233 " 1 2 3 4 6 7
;
; non-greedy repeats added 21/04/00
- match_default normal REG_EXTENDED
a** !
a*? aa 0 0
a?? aa 0 0
a++ !
a+? aa 0 1
a{1,3}{1} !
a{1,3}? aaa 0 1
\w+?w ...ccccccwcccccw 3 10
\W+\w+?w ...ccccccwcccccw 0 10
abc|\w+? abd 0 1
abc|\w+? abcd 0 3
<\s*tag[^>]*>(.*?)<\s*/tag\s*> " <tag>here is some text</tag> <tag></tag>" 1 29 6 23
<\s*tag[^>]*>(.*?)<\s*/tag\s*> " < tag attr=\"something\">here is some text< /tag > <tag></tag>" 1 49 24 41
;
; non-marking parenthesis added 25/04/00
- match_default normal REG_EXTENDED
(?:abc)+ xxabcabcxx 2 8
(?:a+)(b+) xaaabbbx 1 7 4 7
(a+)(?:b+) xaaabbba 1 7 1 4
(?:(a+)b+) xaaabbba 1 7 1 4
(?:a+(b+)) xaaabbba 1 7 4 7
a+(?#b+)b+ xaaabbba 1 7
;
; try some partial matches:
- match_partial match_default normal REG_EXTENDED REG_NO_POSIX_TEST
(xyz)(.*)abc xyzaaab -1 -1 0 3 3 7
(xyz)(.*)abc xyz -1 -1 0 3 3 3
(xyz)(.*)abc xy -1 -1 -1 -1 -1 -1
View Git Blame Copy Permalink