diff options
author | Alexander Barkov <bar@mnogosearch.org> | 2013-12-02 14:39:08 +0400 |
---|---|---|
committer | Alexander Barkov <bar@mnogosearch.org> | 2013-12-02 14:39:08 +0400 |
commit | 5bb01fa1ace1dcfe87c9c1eae3cd30a55c9de032 (patch) | |
tree | 0e88f2fe808ebab43f6f6c1c4bd314693525455c /pcre | |
parent | d25d7ec589cb83acd00ae2c7251dd851ff3cc1a7 (diff) | |
download | mariadb-git-5bb01fa1ace1dcfe87c9c1eae3cd30a55c9de032.tar.gz |
MDEV-5357 REGEXP word boundaries don't work
Applied a patch from Philip Hazel implementing the non-standard
syntax for word boundaries in PCRE, for compatibility with the
old Henry Spencer's regex library.
Diffstat (limited to 'pcre')
-rw-r--r-- | pcre/pcre_compile.c | 35 | ||||
-rw-r--r-- | pcre/pcre_internal.h | 2 |
2 files changed, 37 insertions, 0 deletions
diff --git a/pcre/pcre_compile.c b/pcre/pcre_compile.c index 0ebb3f168f1..a307372fbfe 100644 --- a/pcre/pcre_compile.c +++ b/pcre/pcre_compile.c @@ -253,6 +253,19 @@ static const verbitem verbs[] = { static const int verbcount = sizeof(verbs)/sizeof(verbitem); +/* Substitutes for [[:<:]] and [[:>:]], which mean start and end of word in +another regex library. */ + +static const pcre_uchar sub_start_of_word[] = { + CHAR_BACKSLASH, CHAR_b, CHAR_LEFT_PARENTHESIS, CHAR_QUESTION_MARK, + CHAR_EQUALS_SIGN, CHAR_BACKSLASH, CHAR_w, CHAR_RIGHT_PARENTHESIS, '\0' }; + +static const pcre_uchar sub_end_of_word[] = { + CHAR_BACKSLASH, CHAR_b, CHAR_LEFT_PARENTHESIS, CHAR_QUESTION_MARK, + CHAR_LESS_THAN_SIGN, CHAR_EQUALS_SIGN, CHAR_BACKSLASH, CHAR_w, + CHAR_RIGHT_PARENTHESIS, '\0' }; + + /* Tables of names of POSIX character classes and their lengths. The names are now all in a single string, to reduce the number of relocations when a shared library is dynamically loaded. The list of lengths is terminated by a zero @@ -4036,8 +4049,30 @@ for (;; ptr++) goto FAILED; } goto NORMAL_CHAR; + + /* In another (POSIX) regex library, the ugly syntax [[:<:]] and [[:>:]] is + used for "start of word" and "end of word". As these are otherwise illegal + sequences, we don't break anything by recognizing them. They are replaced + by \b(?=\w) and \b(?<=\w) respectively. Sequences like [a[:<:]] are + erroneous and are handled by the normal code below. */ case CHAR_LEFT_SQUARE_BRACKET: + if (STRNCMP_UC_C8(ptr+1, STRING_WEIRD_STARTWORD, 6) == 0) + { + nestptr = ptr + 7; + ptr = sub_start_of_word - 1; + continue; + } + + if (STRNCMP_UC_C8(ptr+1, STRING_WEIRD_ENDWORD, 6) == 0) + { + nestptr = ptr + 7; + ptr = sub_end_of_word - 1; + continue; + } + + /* Handle a real character class. */ + previous = code; /* PCRE supports POSIX class stuff inside a class. Perl gives an error if diff --git a/pcre/pcre_internal.h b/pcre/pcre_internal.h index 307069ca9d6..cd6ef3ed83e 100644 --- a/pcre/pcre_internal.h +++ b/pcre/pcre_internal.h @@ -1794,6 +1794,8 @@ only. */ #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E +#define STRING_WEIRD_STARTWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET +#define STRING_WEIRD_ENDWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS |