diff --git a/doc/html/boost_regex/background_information.html b/doc/html/boost_regex/background_information.html
index db5c0165..76048eed 100644
--- a/doc/html/boost_regex/background_information.html
+++ b/doc/html/boost_regex/background_information.html
@@ -3,7 +3,7 @@
Background Information
Acknowledgements
Test and Example Programs
captures_test.cpp .
regex_timer.cpp .
FAQ
References and Further Information
Headers
History
History
syntax_option_type no_empty_expressions #1081 .
Added support for Perl style ${n} expressions in format strings (issue
@@ -53,7 +53,7 @@
Localization
< boost / regex / user . hpp > Performance
Redistributables
Standards Conformance
@@ -36,7 +36,7 @@
Report on C++ Library Extensions.
@@ -62,7 +62,7 @@
(??{code}) Not implementable in a compiled strongly typed language.
@@ -82,7 +82,7 @@
a custom traits class.
diff --git a/doc/html/boost_regex/background_information/thread_safety.html b/doc/html/boost_regex/background_information/thread_safety.html
index eac436ac..c67aa107 100644
--- a/doc/html/boost_regex/background_information/thread_safety.html
+++ b/doc/html/boost_regex/background_information/thread_safety.html
@@ -3,7 +3,7 @@
Thread Safety
Understanding Marked Sub-Expressions and Captures
@@ -218,7 +218,7 @@
output stream.
@@ -231,7 +231,7 @@
you can determine which sub-expressions matched by accessing the sub_match :: matched
diff --git a/doc/html/boost_regex/configuration.html b/doc/html/boost_regex/configuration.html
index 4489bc41..7b656e1a 100644
--- a/doc/html/boost_regex/configuration.html
+++ b/doc/html/boost_regex/configuration.html
@@ -3,7 +3,7 @@
Configuration
Algorithm Selection
Compiler Setup
Linkage Options
Locale and traits class selection
Algorithm Tuning
Search and Replace Format String Syntax
Boost-Extended Format String Syntax
@@ -40,7 +40,7 @@
you want a to output literal parenthesis.
@@ -66,7 +66,7 @@
with "bar" otherwise.
Perl Format String Syntax
Sed Format String Syntax
Building and Installing the Library
@@ -58,7 +58,7 @@
started guide for more information.
@@ -166,7 +166,7 @@
a lot in compile times!
@@ -302,7 +302,7 @@
see the config library documentation .
@@ -347,7 +347,7 @@
will build v9 variants of the regex library named libboost_regex_v9.a etc.
Introduction and Overview
Partial Matches
Reference
bad_expression
bad_expression
#include < boost / pattern_except . hpp >
@@ -54,7 +54,7 @@
} regex_error ( const std :: string & s , regex_constants :: error_type err , std :: ptrdiff_t pos );
diff --git a/doc/html/boost_regex/ref/basic_regex.html b/doc/html/boost_regex/ref/basic_regex.html
index 264865a5..30cb5969 100644
--- a/doc/html/boost_regex/ref/basic_regex.html
+++ b/doc/html/boost_regex/ref/basic_regex.html
@@ -3,7 +3,7 @@
basic_regex
basic_regex
#include < boost / regex . hpp >
@@ -244,7 +244,7 @@
}
@@ -327,7 +327,7 @@
basic_regex
-
Table 1. basic_regex default construction postconditions
+
Table 1. basic_regex default construction postconditions
f
-
Table 2. Postconditions for basic_regex construction
+
Table 2. Postconditions for basic_regex construction
f
-
Table 3. Postconditions for basic_regex construction
+
Table 3. Postconditions for basic_regex construction
f
-
Table 4. Postconditions for basic_regex construction
+
Table 4. Postconditions for basic_regex construction
f
-
Table 5. Postconditions for basic_regex construction
+
Table 5. Postconditions for basic_regex construction
f
-
Table 6. Postconditions for basic_regex construction
+
Table 6. Postconditions for basic_regex construction
f
-
Table 7. Postconditions for basic_regex::assign
+
Table 7. Postconditions for basic_regex::assign
Concepts
charT Requirements
Iterator Requirements
Traits Class Requirements
diff --git a/doc/html/boost_regex/ref/deprecated_interfaces.html b/doc/html/boost_regex/ref/deprecated_interfaces.html
index caa8cd1f..784d146a 100644
--- a/doc/html/boost_regex/ref/deprecated_interfaces.html
+++ b/doc/html/boost_regex/ref/deprecated_interfaces.html
@@ -3,7 +3,7 @@
Deprecated Interfaces
-
+
diff --git a/doc/html/boost_regex/ref/deprecated_interfaces/old_regex.html b/doc/html/boost_regex/ref/deprecated_interfaces/old_regex.html
index a360c329..a79abdd6 100644
--- a/doc/html/boost_regex/ref/deprecated_interfaces/old_regex.html
+++ b/doc/html/boost_regex/ref/deprecated_interfaces/old_regex.html
@@ -3,7 +3,7 @@
High Level Class RegEx (Deprecated)
-
+
diff --git a/doc/html/boost_regex/ref/deprecated_interfaces/regex_format.html b/doc/html/boost_regex/ref/deprecated_interfaces/regex_format.html
index 62293232..02829968 100644
--- a/doc/html/boost_regex/ref/deprecated_interfaces/regex_format.html
+++ b/doc/html/boost_regex/ref/deprecated_interfaces/regex_format.html
@@ -3,7 +3,7 @@
regex_format (Deprecated)
-
+
@@ -34,7 +34,7 @@
previous version of Boost.Regex and will not be further updated:
diff --git a/doc/html/boost_regex/ref/deprecated_interfaces/regex_grep.html b/doc/html/boost_regex/ref/deprecated_interfaces/regex_grep.html
index 1bd7f867..ae24f41c 100644
--- a/doc/html/boost_regex/ref/deprecated_interfaces/regex_grep.html
+++ b/doc/html/boost_regex/ref/deprecated_interfaces/regex_grep.html
@@ -3,7 +3,7 @@
regex_grep (Deprecated)
-
+
diff --git a/doc/html/boost_regex/ref/deprecated_interfaces/regex_split.html b/doc/html/boost_regex/ref/deprecated_interfaces/regex_split.html
index 1c34e94d..26643536 100644
--- a/doc/html/boost_regex/ref/deprecated_interfaces/regex_split.html
+++ b/doc/html/boost_regex/ref/deprecated_interfaces/regex_split.html
@@ -3,7 +3,7 @@
regex_split (deprecated)
-
+
diff --git a/doc/html/boost_regex/ref/error_type.html b/doc/html/boost_regex/ref/error_type.html
index d91f0e9e..baa2636a 100644
--- a/doc/html/boost_regex/ref/error_type.html
+++ b/doc/html/boost_regex/ref/error_type.html
@@ -3,7 +3,7 @@
error_type
-
+
@@ -27,7 +27,7 @@
error_type
@@ -57,7 +57,7 @@
}
diff --git a/doc/html/boost_regex/ref/match_flag_type.html b/doc/html/boost_regex/ref/match_flag_type.html
index 50784ec1..8ee8482a 100644
--- a/doc/html/boost_regex/ref/match_flag_type.html
+++ b/doc/html/boost_regex/ref/match_flag_type.html
@@ -3,7 +3,7 @@
match_flag_type
}
diff --git a/doc/html/boost_regex/ref/match_results.html b/doc/html/boost_regex/ref/match_results.html
index 20f0bd3d..94438118 100644
--- a/doc/html/boost_regex/ref/match_results.html
+++ b/doc/html/boost_regex/ref/match_results.html
@@ -3,7 +3,7 @@
match_results
match_results
#include < boost / regex . hpp >
@@ -142,7 +142,7 @@
match_results < BidirectionalIterator , Allocator >& m2 );
diff --git a/doc/html/boost_regex/ref/non_std_strings.html b/doc/html/boost_regex/ref/non_std_strings.html
index 5f38cc26..4681e22b 100644
--- a/doc/html/boost_regex/ref/non_std_strings.html
+++ b/doc/html/boost_regex/ref/non_std_strings.html
@@ -3,7 +3,7 @@
Interfacing With Non-Standard String Types
Working With Unicode and ICU String Types
Introduction to using Regex with ICU
Unicode Regular Expression Algorithms
@@ -89,7 +89,7 @@
}
@@ -128,7 +128,7 @@
}
diff --git a/doc/html/boost_regex/ref/non_std_strings/icu/unicode_iter.html b/doc/html/boost_regex/ref/non_std_strings/icu/unicode_iter.html
index 0fbd6483..be77c195 100644
--- a/doc/html/boost_regex/ref/non_std_strings/icu/unicode_iter.html
+++ b/doc/html/boost_regex/ref/non_std_strings/icu/unicode_iter.html
@@ -3,7 +3,7 @@
Unicode Aware Regex Iterators
@@ -126,7 +126,7 @@
Provided of course that the input is encoded as UTF-8.
diff --git a/doc/html/boost_regex/ref/non_std_strings/icu/unicode_types.html b/doc/html/boost_regex/ref/non_std_strings/icu/unicode_types.html
index cefbc15f..6c64febb 100644
--- a/doc/html/boost_regex/ref/non_std_strings/icu/unicode_types.html
+++ b/doc/html/boost_regex/ref/non_std_strings/icu/unicode_types.html
@@ -3,7 +3,7 @@
Unicode regular expression types
Using Boost Regex With MFC Strings
Overloaded Algorithms For MFC String Types
@@ -82,7 +82,7 @@
}
@@ -110,7 +110,7 @@
}
@@ -149,7 +149,7 @@
}
@@ -164,7 +164,7 @@
+ s . GetLength (), e , f );
diff --git a/doc/html/boost_regex/ref/non_std_strings/mfc_strings/mfc_intro.html b/doc/html/boost_regex/ref/non_std_strings/mfc_strings/mfc_intro.html
index b9c60c1b..9ccedc23 100644
--- a/doc/html/boost_regex/ref/non_std_strings/mfc_strings/mfc_intro.html
+++ b/doc/html/boost_regex/ref/non_std_strings/mfc_strings/mfc_intro.html
@@ -3,7 +3,7 @@
Introduction to Boost.Regex and MFC Strings
Iterating Over the Matches Within An MFC String
regex_iterator regex_token_iterator }
Regular Expression Creation From an MFC String
Regex Types Used With MFC Strings
POSIX Compatible C API's
@@ -379,7 +379,7 @@
@@ -467,7 +467,7 @@
@@ -537,7 +537,7 @@
diff --git a/doc/html/boost_regex/ref/regex_iterator.html b/doc/html/boost_regex/ref/regex_iterator.html
index 80561efa..a3c9d35c 100644
--- a/doc/html/boost_regex/ref/regex_iterator.html
+++ b/doc/html/boost_regex/ref/regex_iterator.html
@@ -3,7 +3,7 @@
regex_iterator
regex_constants :: match_flag_type m = regex_constants :: match_default );
@@ -436,7 +436,7 @@
m
diff --git a/doc/html/boost_regex/ref/regex_match.html b/doc/html/boost_regex/ref/regex_match.html
index 7e8ea6b0..72e113b4 100644
--- a/doc/html/boost_regex/ref/regex_match.html
+++ b/doc/html/boost_regex/ref/regex_match.html
@@ -3,7 +3,7 @@
regex_match
match_flag_type flags = match_default );
template < class BidirectionalIterator , class Allocator , class charT , class traits >
@@ -360,7 +360,7 @@
Effects regex_match ( s . begin (), s . end (), e , flags )
diff --git a/doc/html/boost_regex/ref/regex_replace.html b/doc/html/boost_regex/ref/regex_replace.html
index 1135203f..72e84c91 100644
--- a/doc/html/boost_regex/ref/regex_replace.html
+++ b/doc/html/boost_regex/ref/regex_replace.html
@@ -3,7 +3,7 @@
regex_replace
match_flag_type flags = match_default );
template < class OutputIterator , class BidirectionalIterator , class traits , class charT >
@@ -163,7 +163,7 @@
and then returns result
diff --git a/doc/html/boost_regex/ref/regex_search.html b/doc/html/boost_regex/ref/regex_search.html
index 35f3a630..8c10b0ec 100644
--- a/doc/html/boost_regex/ref/regex_search.html
+++ b/doc/html/boost_regex/ref/regex_search.html
@@ -3,7 +3,7 @@
regex_search
match_flag_type flags = match_default );
template < class BidirectionalIterator , class Allocator , class charT , class traits >
@@ -355,7 +355,7 @@
Effects regex_search ( s . begin (), s . end (), e , flags )
diff --git a/doc/html/boost_regex/ref/regex_token_iterator.html b/doc/html/boost_regex/ref/regex_token_iterator.html
index 0df2dcd4..14dedba1 100644
--- a/doc/html/boost_regex/ref/regex_token_iterator.html
+++ b/doc/html/boost_regex/ref/regex_token_iterator.html
@@ -3,7 +3,7 @@
regex_token_iterator
regex_constants :: match_flag_type m = regex_constants :: match_default );
@@ -383,7 +383,7 @@
m
diff --git a/doc/html/boost_regex/ref/regex_traits.html b/doc/html/boost_regex/ref/regex_traits.html
index 0d515761..04a84d31 100644
--- a/doc/html/boost_regex/ref/regex_traits.html
+++ b/doc/html/boost_regex/ref/regex_traits.html
@@ -3,7 +3,7 @@
regex_traits
}
diff --git a/doc/html/boost_regex/ref/sub_match.html b/doc/html/boost_regex/ref/sub_match.html
index bcc5eed7..98bd6f93 100644
--- a/doc/html/boost_regex/ref/sub_match.html
+++ b/doc/html/boost_regex/ref/sub_match.html
@@ -3,7 +3,7 @@
sub_match
}
@@ -473,7 +473,7 @@
@@ -1008,7 +1008,7 @@
+ m2 . str () .
diff --git a/doc/html/boost_regex/ref/syntax_option_type.html b/doc/html/boost_regex/ref/syntax_option_type.html
index 16e179c8..c5a27483 100644
--- a/doc/html/boost_regex/ref/syntax_option_type.html
+++ b/doc/html/boost_regex/ref/syntax_option_type.html
@@ -3,7 +3,7 @@
syntax_option_type
Options for POSIX Basic Regular Expressions
Options for POSIX Extended Regular Expressions
Options for Literal Strings
Overview of syntax_option_type
Options for Perl Regular Expressions
syntax_option_type Synopsis
Regular Expression Syntax
POSIX Extended Regular Expression Syntax
@@ -46,7 +46,7 @@
.[{()\*+?|^$
@@ -74,7 +74,7 @@
@@ -86,7 +86,7 @@
of an expression, or the last character of a sub-expression.
@@ -184,7 +184,7 @@ cab
operator to be applied to.
@@ -214,7 +214,7 @@ cab
@@ -227,7 +227,7 @@ cab
will match either of "abd" or "abef".
@@ -240,7 +240,7 @@ cab
A bracket expression may contain any combination of the following:
@@ -249,7 +249,7 @@ cab
or 'c'.
@@ -265,7 +265,7 @@ cab
the code points of the characters only.
@@ -274,7 +274,7 @@ cab
range a - c
@@ -284,7 +284,7 @@ cab
character class names .
@@ -312,7 +312,7 @@ cab
matches a NUL character.
@@ -329,7 +329,7 @@ cab
or even all locales on one platform.
@@ -337,7 +337,7 @@ cab
[[: digit :] a - c [. NUL .]]
@@ -363,7 +363,7 @@ cab
extensions are also supported by Boost.Regex:
@@ -552,7 +552,7 @@ cab
@@ -706,7 +706,7 @@ cab
@@ -813,7 +813,7 @@ cab
matches any "digit" character, as does
\ p { digit } .
@@ -888,7 +888,7 @@ cab
@@ -979,7 +979,7 @@ cab
@@ -991,7 +991,7 @@ cab
match to start where the last one ended.
@@ -1005,7 +1005,7 @@ cab
\*+ aaa
@@ -1056,7 +1056,7 @@ cab
@@ -1065,7 +1065,7 @@ cab
\@ matches a literal '@'.
@@ -1101,7 +1101,7 @@ cab
@@ -1136,7 +1136,7 @@ cab
used with the -E option.
@@ -1150,7 +1150,7 @@ cab
these by default anyway.
@@ -1163,7 +1163,7 @@ cab
modify how the case and locale sensitivity are to be applied.
diff --git a/doc/html/boost_regex/syntax/basic_syntax.html b/doc/html/boost_regex/syntax/basic_syntax.html
index 87d92ff1..72f216b8 100644
--- a/doc/html/boost_regex/syntax/basic_syntax.html
+++ b/doc/html/boost_regex/syntax/basic_syntax.html
@@ -3,7 +3,7 @@
POSIX Basic Regular Expression Syntax
@@ -45,7 +45,7 @@
.[\*^$
@@ -73,7 +73,7 @@
@@ -85,7 +85,7 @@
of an expression, or the last character of a sub-expression.
@@ -155,7 +155,7 @@ aaaa
to.
@@ -173,7 +173,7 @@ aaaa
aaabba
@@ -220,7 +220,7 @@ aaaa
range a-c.
character class names .
@@ -284,7 +284,7 @@ aaaa
[[: digit :] a - c [. NUL .]].
@@ -299,7 +299,7 @@ aaaa
will match either a literal '\' or a '^'.
@@ -333,7 +333,7 @@ aaaa
As its name suggests, this behavior is consistent with the Unix utility grep.
@@ -613,7 +613,7 @@ aaaa
leftmost-longest rule.
@@ -627,7 +627,7 @@ aaaa
options modify how the case and locale sensitivity are to be applied.
diff --git a/doc/html/boost_regex/syntax/character_classes.html b/doc/html/boost_regex/syntax/character_classes.html
index 730370e1..8611af01 100644
--- a/doc/html/boost_regex/syntax/character_classes.html
+++ b/doc/html/boost_regex/syntax/character_classes.html
@@ -3,7 +3,7 @@
Character Class Names
Character classes that are supported by Unicode Regular Expressions
Character Classes that are Always Supported
Collating Names
Digraphs
Named Unicode Characters
POSIX Symbolic Names
The Leftmost Longest Rule
Perl Regular Expression Syntax
@@ -45,7 +45,7 @@
boost :: regex e2 ( my_expression , boost :: regex :: perl | boost :: regex :: icase );
.[{()\*+?|^$
@@ -75,7 +75,7 @@
@@ -85,7 +85,7 @@
A '$' character shall match the end of a line.
@@ -197,7 +197,7 @@
operator to be applied to.
+ By default when a repeated patten does not match then the engine will backtrack
+ until a match is found. However, this behaviour can sometime be undesireable
+ so there are also "pocessive" repeats: these match as much as possible
+ and do not then allow backtracking if the rest of the expression fails to
+ match.
+
+
+ *+
+
+ ++
+
+ ?+
+
+ { n ,}+
+
+ { n , m }+
@@ -248,7 +280,7 @@
aaabba
@@ -277,7 +309,7 @@
(?: abc )??
@@ -290,7 +322,7 @@
A bracket expression may contain any combination of the following:
@@ -298,7 +330,7 @@
or 'c'.
@@ -320,7 +352,7 @@
range a - c
character class names .
not
@@ -391,7 +423,7 @@
[[: digit :] a - c [. NUL .]]
@@ -584,7 +616,7 @@
@@ -676,6 +708,30 @@
+
+
+ \ h
+
+
+
+ Horizontal whitespace
+
+
+
+
+
+
+ \ v
+
+
+
+ Vertical whitespace
+
+
+
+
\ D
+
+
+
+ \ H
+
+
+
+ Not Horizontal whitespace
+
+
+
+
+
+
+ \ V
+
+
+
+ Not Vertical whitespace
+
+
+
\ p { digit }
@@ -868,7 +948,7 @@
Matches only when not at a word boundary.
@@ -893,7 +973,7 @@
to the regular expression \ n *\ z
@@ -918,7 +998,7 @@
\*+ aaa
@@ -929,7 +1009,7 @@
combining characters.
(?
@@ -956,7 +1036,7 @@
are ignored.
@@ -971,7 +1051,7 @@
applies the specified modifiers to pattern only.
@@ -1003,7 +1083,7 @@
could be used to validate the password.
@@ -1017,7 +1097,7 @@
(pattern must be of fixed length).
@@ -1280,7 +1360,7 @@
JavaScript JScript perl
@@ -1293,7 +1373,7 @@
sensitivity are to be applied.
no_mod_s
diff --git a/doc/html/boost_regex/unicode.html b/doc/html/boost_regex/unicode.html
index 85678400..00b14229 100644
--- a/doc/html/boost_regex/unicode.html
+++ b/doc/html/boost_regex/unicode.html
@@ -3,7 +3,7 @@
Unicode and Boost.Regex
@@ -56,7 +56,7 @@
diff --git a/doc/html/index.html b/doc/html/index.html
index 9b623b67..365165f2 100644
--- a/doc/html/index.html
+++ b/doc/html/index.html
@@ -3,7 +3,7 @@
Boost.Regex
Copyright © 1998 -2007 John Maddock
-Last revised: December 23, 2008 at 17:35:37 GMT
Last revised: April 23, 2009 at 09:45:24 GMT
diff --git a/doc/syntax_perl.qbk b/doc/syntax_perl.qbk
index 9a76276d..b3b4056f 100644
--- a/doc/syntax_perl.qbk
+++ b/doc/syntax_perl.qbk
@@ -143,6 +143,23 @@ consume as little input as possible while still producing a match.
`{n,m}?` Matches the previous atom between n and m times, while
consuming as little input as possible.
+[h4 Pocessive repeats]
+
+By default when a repeated patten does not match then the engine will backtrack until
+a match is found. However, this behaviour can sometime be undesireable so there are
+also "pocessive" repeats: these match as much as possible and do not then allow
+backtracking if the rest of the expression fails to match.
+
+`*+` Matches the previous atom zero or more times, while giving nothing back.
+
+`++` Matches the previous atom one or more times, while giving nothing back.
+
+`?+` Matches the previous atom zero or one times, while giving nothing back.
+
+`{n,}+` Matches the previous atom n or more times, while giving nothing back.
+
+`{n,m}+` Matches the previous atom between n and m times, while giving nothing back.
+
[h4 Back references]
An escape character followed by a digit /n/, where /n/ is in the range 1-9,
@@ -296,11 +313,15 @@ The following are supported by default:
[[`\s`][`[[:space:]]`]]
[[`\u`][`[[:upper:]]`]]
[[`\w`][`[[:word:]]`]]
+[[`\h`][Horizontal whitespace]]
+[[`\v`][Vertical whitespace]]
[[`\D`][`[^[:digit:]]`]]
[[`\L`][`[^[:lower:]]`]]
[[`\S`][`[^[:space:]]`]]
[[`\U`][`[^[:upper:]]`]]
[[`\W`][`[^[:word:]]`]]
+[[`\H`][Not Horizontal whitespace]]
+[[`\V`][Not Vertical whitespace]]
]
[h5 Character Properties]
diff --git a/include/boost/regex/icu.hpp b/include/boost/regex/icu.hpp
index 2cf37e5c..24715572 100644
--- a/include/boost/regex/icu.hpp
+++ b/include/boost/regex/icu.hpp
@@ -184,7 +184,9 @@ private:
offset_underscore = U_CHAR_CATEGORY_COUNT+3,
offset_unicode = U_CHAR_CATEGORY_COUNT+4,
offset_any = U_CHAR_CATEGORY_COUNT+5,
- offset_ascii = U_CHAR_CATEGORY_COUNT+6
+ offset_ascii = U_CHAR_CATEGORY_COUNT+6,
+ offset_horizontal = U_CHAR_CATEGORY_COUNT+7,
+ offset_vertical = U_CHAR_CATEGORY_COUNT+8
};
//
@@ -197,6 +199,8 @@ private:
static const char_class_type mask_unicode;
static const char_class_type mask_any;
static const char_class_type mask_ascii;
+ static const char_class_type mask_horizontal;
+ static const char_class_type mask_vertical;
static char_class_type lookup_icu_mask(const ::UChar32* p1, const ::UChar32* p2);
diff --git a/include/boost/regex/v4/basic_regex_parser.hpp b/include/boost/regex/v4/basic_regex_parser.hpp
index 9cc4abe8..065a2d80 100644
--- a/include/boost/regex/v4/basic_regex_parser.hpp
+++ b/include/boost/regex/v4/basic_regex_parser.hpp
@@ -610,6 +610,7 @@ bool basic_regex_parser
::parse_extended_escape()
// fall through:
case regex_constants::escape_type_class:
{
+escape_type_class_jump:
typedef typename traits::char_class_type mask_type;
mask_type m = this->m_traits.lookup_classname(m_position, m_position+1);
if(m != 0)
@@ -720,6 +721,10 @@ bool basic_regex_parser::parse_extended_escape()
}
fail(regex_constants::error_ctype, m_position - m_base);
}
+ case regex_constants::escape_type_control_v:
+ if(0 == (this->flags() & (regbase::main_option_type | regbase::no_perl_ex)))
+ goto escape_type_class_jump;
+ // fallthrough:
default:
this->append_literal(unescape_character());
break;
@@ -747,6 +752,7 @@ template
bool basic_regex_parser::parse_repeat(std::size_t low, std::size_t high)
{
bool greedy = true;
+ bool pocessive = false;
std::size_t insert_point;
//
// when we get to here we may have a non-greedy ? mark still to come:
@@ -758,12 +764,19 @@ bool basic_regex_parser::parse_repeat(std::size_t low, std::size_
)
)
{
- // OK we have a perl regex, check for a '?':
+ // OK we have a perl or emacs regex, check for a '?':
if(this->m_traits.syntax_type(*m_position) == regex_constants::syntax_question)
{
greedy = false;
++m_position;
}
+ // for perl regexes only check for pocessive ++ repeats.
+ if((0 == (this->flags() & regbase::main_option_type))
+ && (this->m_traits.syntax_type(*m_position) == regex_constants::syntax_plus))
+ {
+ pocessive = true;
+ ++m_position;
+ }
}
if(0 == this->m_last_state)
{
@@ -832,6 +845,20 @@ bool basic_regex_parser::parse_repeat(std::size_t low, std::size_
// now fill in the alt jump for the repeat:
rep = static_cast(this->getaddress(rep_off));
rep->alt.i = this->m_pdata->m_data.size() - rep_off;
+ //
+ // If the repeat is pocessive then bracket the repeat with a (?>...)
+ // independent sub-expression construct:
+ //
+ if(pocessive)
+ {
+ re_brace* pb = static_cast(this->insert_state(insert_point, syntax_element_startmark, sizeof(re_brace)));
+ pb->index = -3;
+ re_jump* jmp = static_cast(this->insert_state(insert_point + sizeof(re_brace), syntax_element_jump, sizeof(re_jump)));
+ this->m_pdata->m_data.align();
+ jmp->alt.i = this->m_pdata->m_data.size() - this->getoffset(jmp);
+ pb = static_cast(this->append_state(syntax_element_endmark, sizeof(re_brace)));
+ pb->index = -3;
+ }
return true;
}
diff --git a/include/boost/regex/v4/cpp_regex_traits.hpp b/include/boost/regex/v4/cpp_regex_traits.hpp
index 89fe49d8..7ce3ed30 100644
--- a/include/boost/regex/v4/cpp_regex_traits.hpp
+++ b/include/boost/regex/v4/cpp_regex_traits.hpp
@@ -394,7 +394,9 @@ enum
char_class_graph=char_class_alnum|char_class_punct,
char_class_blank=1<<9,
char_class_word=1<<10,
- char_class_unicode=1<<11
+ char_class_unicode=1<<11,
+ char_class_horizontal_space=1<<12,
+ char_class_vertical_space=1<<13
};
#endif
@@ -413,6 +415,8 @@ public:
BOOST_STATIC_CONSTANT(char_class_type, mask_blank = 1u << 24);
BOOST_STATIC_CONSTANT(char_class_type, mask_word = 1u << 25);
BOOST_STATIC_CONSTANT(char_class_type, mask_unicode = 1u << 26);
+ BOOST_STATIC_CONSTANT(char_class_type, mask_horizontal = 1u << 27);
+ BOOST_STATIC_CONSTANT(char_class_type, mask_vertical = 1u << 28);
#endif
typedef std::basic_string string_type;
@@ -477,6 +481,10 @@ template
typename cpp_regex_traits_implementation::char_class_type const cpp_regex_traits_implementation::mask_word;
template
typename cpp_regex_traits_implementation::char_class_type const cpp_regex_traits_implementation::mask_unicode;
+template
+typename cpp_regex_traits_implementation::char_class_type const cpp_regex_traits_implementation::mask_vertical;
+template
+typename cpp_regex_traits_implementation::char_class_type const cpp_regex_traits_implementation::mask_horizontal;
#endif
#endif
@@ -688,18 +696,20 @@ void cpp_regex_traits_implementation::init()
// Custom class names:
//
#ifndef BOOST_REGEX_BUGGY_CTYPE_FACET
- static const char_class_type masks[14] =
+ static const char_class_type masks[16] =
{
std::ctype::alnum,
std::ctype::alpha,
std::ctype::cntrl,
std::ctype::digit,
std::ctype::graph,
+ cpp_regex_traits_implementation::mask_horizontal,
std::ctype::lower,
std::ctype::print,
std::ctype::punct,
std::ctype::space,
std::ctype::upper,
+ cpp_regex_traits_implementation::mask_vertical,
std::ctype::xdigit,
cpp_regex_traits_implementation::mask_blank,
cpp_regex_traits_implementation::mask_word,
@@ -713,11 +723,13 @@ void cpp_regex_traits_implementation::init()
::boost::re_detail::char_class_cntrl,
::boost::re_detail::char_class_digit,
::boost::re_detail::char_class_graph,
+ ::boost::re_detail::char_class_horizontal_space,
::boost::re_detail::char_class_lower,
::boost::re_detail::char_class_print,
::boost::re_detail::char_class_punct,
::boost::re_detail::char_class_space,
::boost::re_detail::char_class_upper,
+ ::boost::re_detail::char_class_vertical_space,
::boost::re_detail::char_class_xdigit,
::boost::re_detail::char_class_blank,
::boost::re_detail::char_class_word,
@@ -744,7 +756,7 @@ typename cpp_regex_traits_implementation::char_class_type
cpp_regex_traits_implementation::lookup_classname_imp(const charT* p1, const charT* p2) const
{
#ifndef BOOST_REGEX_BUGGY_CTYPE_FACET
- static const char_class_type masks[20] =
+ static const char_class_type masks[22] =
{
0,
std::ctype::alnum,
@@ -754,6 +766,7 @@ typename cpp_regex_traits_implementation::char_class_type
std::ctype::digit,
std::ctype::digit,
std::ctype::graph,
+ cpp_regex_traits_implementation::mask_horizontal,
std::ctype::lower,
std::ctype::lower,
std::ctype::print,
@@ -763,12 +776,13 @@ typename cpp_regex_traits_implementation::char_class_type
std::ctype::upper,
cpp_regex_traits_implementation::mask_unicode,
std::ctype::upper,
+ cpp_regex_traits_implementation::mask_vertical,
std::ctype::alnum | cpp_regex_traits_implementation::mask_word,
std::ctype::alnum | cpp_regex_traits_implementation::mask_word,
std::ctype::xdigit,
};
#else
- static const char_class_type masks[20] =
+ static const char_class_type masks[22] =
{
0,
::boost::re_detail::char_class_alnum,
@@ -778,6 +792,7 @@ typename cpp_regex_traits_implementation::char_class_type
::boost::re_detail::char_class_digit,
::boost::re_detail::char_class_digit,
::boost::re_detail::char_class_graph,
+ ::boost::re_detail::char_class_horizontal_space,
::boost::re_detail::char_class_lower,
::boost::re_detail::char_class_lower,
::boost::re_detail::char_class_print,
@@ -787,6 +802,7 @@ typename cpp_regex_traits_implementation::char_class_type
::boost::re_detail::char_class_upper,
::boost::re_detail::char_class_unicode,
::boost::re_detail::char_class_upper,
+ ::boost::re_detail::char_class_vertical_space,
::boost::re_detail::char_class_alnum | ::boost::re_detail::char_class_word,
::boost::re_detail::char_class_alnum | ::boost::re_detail::char_class_word,
::boost::re_detail::char_class_xdigit,
@@ -820,7 +836,9 @@ bool cpp_regex_traits_implementation::isctype(const charT c, char_class_t
|| ((mask & ::boost::re_detail::char_class_xdigit) && (m_pctype->is(std::ctype::xdigit, c)))
|| ((mask & ::boost::re_detail::char_class_blank) && (m_pctype->is(std::ctype::space, c)) && !::boost::re_detail::is_separator(c))
|| ((mask & ::boost::re_detail::char_class_word) && (c == '_'))
- || ((mask & ::boost::re_detail::char_class_unicode) && ::boost::re_detail::is_extended(c));
+ || ((mask & ::boost::re_detail::char_class_unicode) && ::boost::re_detail::is_extended(c))
+ || ((mask & ::boost::re_detail::char_class_vertical) && (is_separator(c) || (c == '\v')))
+ || ((mask & ::boost::re_detail::char_class_horizontal) && m_pctype->is(std::ctype::space, c) && !(is_separator(c) || (c == '\v')));
}
#endif
@@ -930,6 +948,12 @@ public:
&& m_pimpl->m_pctype->is(std::ctype::space, c)
&& !re_detail::is_separator(c))
return true;
+ else if((f & re_detail::cpp_regex_traits_implementation::mask_vertical)
+ && (::boost::re_detail::is_separator(c) || (c == '\v')))
+ return true;
+ else if((f & re_detail::cpp_regex_traits_implementation::mask_horizontal)
+ && this->isctype(c, std::ctype::space) && !this->isctype(c, re_detail::cpp_regex_traits_implementation::mask_vertical))
+ return true;
return false;
#else
return m_pimpl->isctype(c, f);
diff --git a/include/boost/regex/v4/regex_traits_defaults.hpp b/include/boost/regex/v4/regex_traits_defaults.hpp
index 42428dd8..8f33d2c8 100644
--- a/include/boost/regex/v4/regex_traits_defaults.hpp
+++ b/include/boost/regex/v4/regex_traits_defaults.hpp
@@ -159,7 +159,7 @@ struct character_pointer_range
template
int get_default_class_id(const charT* p1, const charT* p2)
{
- static const charT data[72] = {
+ static const charT data[73] = {
'a', 'l', 'n', 'u', 'm',
'a', 'l', 'p', 'h', 'a',
'b', 'l', 'a', 'n', 'k',
@@ -172,11 +172,12 @@ int get_default_class_id(const charT* p1, const charT* p2)
's', 'p', 'a', 'c', 'e',
'u', 'n', 'i', 'c', 'o', 'd', 'e',
'u', 'p', 'p', 'e', 'r',
+ 'v',
'w', 'o', 'r', 'd',
'x', 'd', 'i', 'g', 'i', 't',
};
- static const character_pointer_range ranges[19] =
+ static const character_pointer_range ranges[21] =
{
{data+0, data+5,}, // alnum
{data+5, data+10,}, // alpha
@@ -185,6 +186,7 @@ int get_default_class_id(const charT* p1, const charT* p2)
{data+20, data+21,}, // d
{data+20, data+25,}, // digit
{data+25, data+30,}, // graph
+ {data+29, data+30,}, // h
{data+30, data+31,}, // l
{data+30, data+35,}, // lower
{data+35, data+40,}, // print
@@ -194,9 +196,10 @@ int get_default_class_id(const charT* p1, const charT* p2)
{data+57, data+58,}, // u
{data+50, data+57,}, // unicode
{data+57, data+62,}, // upper
- {data+62, data+63,}, // w
- {data+62, data+66,}, // word
- {data+66, data+72,}, // xdigit
+ {data+62, data+63,}, // v
+ {data+63, data+64,}, // w
+ {data+63, data+67,}, // word
+ {data+67, data+73,}, // xdigit
};
static const character_pointer_range* ranges_begin = ranges;
static const character_pointer_range* ranges_end = ranges + (sizeof(ranges)/sizeof(ranges[0]));
diff --git a/include/boost/regex/v4/w32_regex_traits.hpp b/include/boost/regex/v4/w32_regex_traits.hpp
index 21a9694a..d5562072 100644
--- a/include/boost/regex/v4/w32_regex_traits.hpp
+++ b/include/boost/regex/v4/w32_regex_traits.hpp
@@ -294,6 +294,8 @@ public:
typedef typename w32_regex_traits::char_class_type char_class_type;
BOOST_STATIC_CONSTANT(char_class_type, mask_word = 0x0400); // must be C1_DEFINED << 1
BOOST_STATIC_CONSTANT(char_class_type, mask_unicode = 0x0800); // must be C1_DEFINED << 2
+ BOOST_STATIC_CONSTANT(char_class_type, mask_horizontal = 0x1000); // must be C1_DEFINED << 3
+ BOOST_STATIC_CONSTANT(char_class_type, mask_vertical = 0x2000); // must be C1_DEFINED << 4
BOOST_STATIC_CONSTANT(char_class_type, mask_base = 0x3ff); // all the masks used by the CT_CTYPE1 group
typedef std::basic_string string_type;
@@ -510,7 +512,7 @@ template
typename w32_regex_traits_implementation::char_class_type
w32_regex_traits_implementation::lookup_classname_imp(const charT* p1, const charT* p2) const
{
- static const char_class_type masks[20] =
+ static const char_class_type masks[22] =
{
0,
0x0104u, // C1_ALPHA | C1_DIGIT
@@ -520,6 +522,7 @@ typename w32_regex_traits_implementation::char_class_type
0x0004u, // C1_DIGIT
0x0004u, // C1_DIGIT
(~(0x0020u|0x0008u|0x0040) & 0x01ffu) | 0x0400u, // not C1_CNTRL or C1_SPACE or C1_BLANK
+ w32_regex_traits_implementation::mask_horizontal,
0x0002u, // C1_LOWER
0x0002u, // C1_LOWER
(~0x0020u & 0x01ffu) | 0x0400, // not C1_CNTRL
@@ -529,6 +532,7 @@ typename w32_regex_traits_implementation::char_class_type
0x0001u, // C1_UPPER
w32_regex_traits_implementation::mask_unicode,
0x0001u, // C1_UPPER
+ w32_regex_traits_implementation::mask_vertical,
0x0104u | w32_regex_traits_implementation::mask_word,
0x0104u | w32_regex_traits_implementation::mask_word,
0x0080u, // C1_XDIGIT
@@ -628,6 +632,12 @@ public:
return true;
else if((f & re_detail::w32_regex_traits_implementation::mask_word) && (c == '_'))
return true;
+ else if((f & re_detail::w32_regex_traits_implementation::mask_vertical)
+ && (::boost::re_detail::is_separator(c) || (c == '\v')))
+ return true;
+ else if((f & re_detail::w32_regex_traits_implementation::mask_horizontal)
+ && this->isctype(c, 0x0008u) && !this->isctype(c, re_detail::w32_regex_traits_implementation::mask_vertical))
+ return true;
return false;
}
int toi(const charT*& p1, const charT* p2, int radix)const
diff --git a/src/c_regex_traits.cpp b/src/c_regex_traits.cpp
index da960eb0..6466bc42 100644
--- a/src/c_regex_traits.cpp
+++ b/src/c_regex_traits.cpp
@@ -122,7 +122,9 @@ enum
char_class_graph=char_class_alnum|char_class_punct,
char_class_blank=1<<9,
char_class_word=1<<10,
- char_class_unicode=1<<11
+ char_class_unicode=1<<11,
+ char_class_horizontal=1<<12,
+ char_class_vertical=1<<13
};
c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::lookup_classname(const char* p1, const char* p2)
@@ -137,6 +139,7 @@ c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::loo
char_class_digit,
char_class_digit,
char_class_graph,
+ char_class_horizontal,
char_class_lower,
char_class_lower,
char_class_print,
@@ -146,6 +149,7 @@ c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::loo
char_class_upper,
char_class_unicode,
char_class_upper,
+ char_class_vertical,
char_class_alnum | char_class_word,
char_class_alnum | char_class_word,
char_class_xdigit,
@@ -176,7 +180,9 @@ bool BOOST_REGEX_CALL c_regex_traits::isctype(char c, char_class_type mask
|| ((mask & char_class_punct) && (std::ispunct)(static_cast(c)))
|| ((mask & char_class_xdigit) && (std::isxdigit)(static_cast(c)))
|| ((mask & char_class_blank) && (std::isspace)(static_cast(c)) && !::boost::re_detail::is_separator(c))
- || ((mask & char_class_word) && (c == '_'));
+ || ((mask & char_class_word) && (c == '_'))
+ || ((mask & char_class_vertical) && (::boost::re_detail::is_separator(c) || (c == '\v')))
+ || ((mask & char_class_horizontal) && (std::isspace)(static_cast(c)) && !::boost::re_detail::is_separator(c) && (c != '\v'));
}
c_regex_traits::string_type BOOST_REGEX_CALL c_regex_traits::lookup_collatename(const char* p1, const char* p2)
diff --git a/src/icu.cpp b/src/icu.cpp
index e06c3176..a815e915 100644
--- a/src/icu.cpp
+++ b/src/icu.cpp
@@ -101,6 +101,8 @@ const icu_regex_traits::char_class_type icu_regex_traits::mask_underscore = icu_
const icu_regex_traits::char_class_type icu_regex_traits::mask_unicode = icu_regex_traits::char_class_type(1) << offset_unicode;
const icu_regex_traits::char_class_type icu_regex_traits::mask_any = icu_regex_traits::char_class_type(1) << offset_any;
const icu_regex_traits::char_class_type icu_regex_traits::mask_ascii = icu_regex_traits::char_class_type(1) << offset_ascii;
+const icu_regex_traits::char_class_type icu_regex_traits::mask_horizontal = icu_regex_traits::char_class_type(1) << offset_horizontal;
+const icu_regex_traits::char_class_type icu_regex_traits::mask_vertical = icu_regex_traits::char_class_type(1) << offset_vertical;
icu_regex_traits::char_class_type icu_regex_traits::lookup_icu_mask(const ::UChar32* p1, const ::UChar32* p2)
{
@@ -370,6 +372,7 @@ icu_regex_traits::char_class_type icu_regex_traits::lookup_classname(const char_
U_GC_ND_MASK,
U_GC_ND_MASK,
(0x3FFFFFFFu) & ~(U_GC_CC_MASK | U_GC_CF_MASK | U_GC_CS_MASK | U_GC_CN_MASK | U_GC_Z_MASK),
+ mask_horizontal,
U_GC_LL_MASK,
U_GC_LL_MASK,
~(U_GC_C_MASK),
@@ -379,6 +382,7 @@ icu_regex_traits::char_class_type icu_regex_traits::lookup_classname(const char_
U_GC_LU_MASK,
mask_unicode,
U_GC_LU_MASK,
+ mask_vertical,
char_class_type(U_GC_L_MASK | U_GC_ND_MASK | U_GC_MN_MASK) | mask_underscore,
char_class_type(U_GC_L_MASK | U_GC_ND_MASK | U_GC_MN_MASK) | mask_underscore,
char_class_type(U_GC_ND_MASK) | mask_xdigit,
@@ -487,6 +491,10 @@ bool icu_regex_traits::isctype(char_type c, char_class_type f) const
return true;
if(((f & mask_ascii) != 0) && (c <= 0x7F))
return true;
+ if(((f & mask_vertical) != 0) && (::boost::re_detail::is_separator(c) || (c == static_cast('\v')) || (m == U_GC_ZL_MASK) || (m == U_GC_ZP_MASK)))
+ return true;
+ if(((f & mask_horizontal) != 0) && !::boost::re_detail::is_separator(c) && u_isspace(c) && (c != static_cast('\v')))
+ return true;
return false;
}
diff --git a/src/wc_regex_traits.cpp b/src/wc_regex_traits.cpp
index 3640f292..cff9d185 100644
--- a/src/wc_regex_traits.cpp
+++ b/src/wc_regex_traits.cpp
@@ -161,7 +161,9 @@ enum
char_class_graph=char_class_alnum|char_class_punct,
char_class_blank=1<<9,
char_class_word=1<<10,
- char_class_unicode=1<<11
+ char_class_unicode=1<<11,
+ char_class_horizontal=1<<12,
+ char_class_vertical=1<<13
};
c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::lookup_classname(const wchar_t* p1, const wchar_t* p2)
@@ -176,6 +178,7 @@ c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::char_class_type BOOST_REGEX_CALL c_regex_traits::isctype(wchar_t c, char_class_typ
|| ((mask & char_class_xdigit) && (std::iswxdigit)(c))
|| ((mask & char_class_blank) && (std::iswspace)(c) && !::boost::re_detail::is_separator(c))
|| ((mask & char_class_word) && (c == '_'))
- || ((mask & char_class_unicode) && (c & ~static_cast(0xff)));
+ || ((mask & char_class_unicode) && (c & ~static_cast(0xff)))
+ || ((mask & char_class_vertical) && (::boost::re_detail::is_separator(c) || (c == L'\v')))
+ || ((mask & char_class_horizontal) && (std::iswspace)(c) && !::boost::re_detail::is_separator(c) && (c != L'\v'));
}
c_regex_traits::string_type BOOST_REGEX_CALL c_regex_traits::lookup_collatename(const wchar_t* p1, const wchar_t* p2)
diff --git a/test/regress/main.cpp b/test/regress/main.cpp
index ffbe5efe..73692208 100644
--- a/test/regress/main.cpp
+++ b/test/regress/main.cpp
@@ -40,6 +40,7 @@ int error_count = 0;
void run_tests()
{
+ RUN_TESTS(test_pocessive_repeats);
RUN_TESTS(basic_tests);
RUN_TESTS(test_simple_repeats);
RUN_TESTS(test_alt);
diff --git a/test/regress/test.hpp b/test/regress/test.hpp
index 046b2d56..e812aa94 100644
--- a/test/regress/test.hpp
+++ b/test/regress/test.hpp
@@ -258,6 +258,6 @@ void test_emacs();
void test_operators();
void test_overloads();
void test_unicode();
-
+void test_pocessive_repeats();
#endif
diff --git a/test/regress/test_non_greedy_repeats.cpp b/test/regress/test_non_greedy_repeats.cpp
index 5cf507d2..c9935bd8 100644
--- a/test/regress/test_non_greedy_repeats.cpp
+++ b/test/regress/test_non_greedy_repeats.cpp
@@ -41,6 +41,6 @@ void test_non_greedy_repeats()
TEST_REGEX_SEARCH("xx[/-]{0,2}?(?:[+-][0-9])??\\z", perl, "xx--", match_default, make_array(0, 4, -2, -2));
TEST_INVALID_REGEX("a{1,3}{1}", perl);
TEST_INVALID_REGEX("a**", perl);
- TEST_INVALID_REGEX("a++", perl);
+ //TEST_INVALID_REGEX("a++", perl);
}
diff --git a/test/regress/test_sets.cpp b/test/regress/test_sets.cpp
index 61374819..3cd7a520 100644
--- a/test/regress/test_sets.cpp
+++ b/test/regress/test_sets.cpp
@@ -266,6 +266,10 @@ void test_sets2()
TEST_REGEX_SEARCH("[\\W]+", perl, "AB_ AB", match_default, make_array(3, 6, -2, -2));
TEST_REGEX_SEARCH("[[:^word:]]+", perl, "AB_ AB", match_default, make_array(3, 6, -2, -2));
TEST_REGEX_SEARCH("\\W+", perl, "AB_ AB", match_default, make_array(3, 6, -2, -2));
+ TEST_REGEX_SEARCH("\\h+", perl, "\v\f\r\n \t\n", match_default, make_array(4, 6, -2, -2));
+ TEST_REGEX_SEARCH("\\V+", perl, "\v\f\r\n \t\n", match_default, make_array(4, 6, -2, -2));
+ TEST_REGEX_SEARCH("\\H+", perl, " \t\v\f\r\n ", match_default, make_array(2, 6, -2, -2));
+ TEST_REGEX_SEARCH("\\v+", perl, " \t\v\f\r\n ", match_default, make_array(2, 6, -2, -2));
test_sets2c();
}
diff --git a/test/regress/test_simple_repeats.cpp b/test/regress/test_simple_repeats.cpp
index 6811e5f9..690c71eb 100644
--- a/test/regress/test_simple_repeats.cpp
+++ b/test/regress/test_simple_repeats.cpp
@@ -436,3 +436,44 @@ void test_fast_repeats2()
}
+void test_pocessive_repeats()
+{
+ using namespace boost::regex_constants;
+ // and again for sets:
+ TEST_REGEX_SEARCH("^(\\w++|\\s++)*$", perl, "now is the time for all good men to come to the aid of the party", match_default, make_array(0, 64, 59, 64, -2, -2));
+ TEST_REGEX_SEARCH("^(\\w++|\\s++)*$", perl, "this is not a line with only words and spaces!", match_default, make_array(-2, -2));
+ TEST_REGEX_SEARCH("(\\d++)(\\w)", perl, "12345a", match_default, make_array(0, 6, 0, 5, 5, 6, -2, -2));
+ TEST_REGEX_SEARCH("(\\d++)(\\w)", perl, "12345+", match_default, make_array(-2, -2));
+ TEST_REGEX_SEARCH("(\\d++)(\\w)", perl, "12345", match_default, make_array(-2, -2));
+ TEST_REGEX_SEARCH("a++b", perl, "aaab", match_default, make_array(0, 4, -2, -2));
+ TEST_REGEX_SEARCH("(a++b)", perl, "aaab", match_default, make_array(0, 4, 0, 4, -2, -2));
+ TEST_REGEX_SEARCH("([^()]++|\\([^()]*\\))+", perl, "((abc(ade)ufh()()x", match_default, make_array(2, 18, 17, 18, -2, -2));
+ TEST_REGEX_SEARCH("\\(([^()]++|\\([^()]+\\))+\\)", perl, "(abc)", match_default, make_array(0, 5, 1, 4, -2, -2));
+ TEST_REGEX_SEARCH("\\(([^()]++|\\([^()]+\\))+\\)", perl, "(abc(def)xyz)", match_default, make_array(0, 13, 9, 12, -2, -2));
+ TEST_REGEX_SEARCH("\\(([^()]++|\\([^()]+\\))+\\)", perl, "((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa", match_default, make_array(-2, -2));
+ /*
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, "<>", match_default, make_array(0, 2, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, "", match_default, make_array(0, 6, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, " hij>", match_default, make_array(0, 15, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, " hij>", match_default, make_array(5, 10, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, "def>", match_default, make_array(0, 10, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, "", match_default, make_array(4, 6, -2, -2));
+ TEST_REGEX_SEARCH("< (?: (?(R) \\d++ | [^<>]*+) | (?R)) * >", perl|mod_x, "]*+) | (?2)) * >))", perl|mod_x, "<>", match_default, make_array(0, 2, 0, 2, 0, 2, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, "", match_default, make_array(0, 6, 0, 6, 0, 6, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, " hij>", match_default, make_array(0, 15, 0, 15, 0, 15, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, " hij>", match_default, make_array(5, 10, 5, 10, 5, 10, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, "def>", match_default, make_array(0, 10, 0, 10, 0, 10, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, "", match_default, make_array(4, 6, 4, 6, 4, 6, -2, -2));
+ TEST_REGEX_SEARCH("((< (?: (?(R) \d++ | [^<>]*+) | (?2)) * >))", perl|mod_x, "