diff options
Diffstat (limited to 'pcre_study.c')
| -rw-r--r-- | pcre_study.c | 86 |
1 files changed, 43 insertions, 43 deletions
diff --git a/pcre_study.c b/pcre_study.c index e2e64e7..e473fdd 100644 --- a/pcre_study.c +++ b/pcre_study.c @@ -441,7 +441,7 @@ for (;;) * Set a bit and maybe its alternate case * *************************************************/ -/* Given a character, set its first byte's bit in the table, and also the +/* Given a character, set its first byte's bit in the table, and also the corresponding bit for the other version of a letter if we are caseless. In UTF-8 mode, for characters greater than 127, we can only do the caseless thing when Unicode property support is available. @@ -451,7 +451,7 @@ Arguments: p points to the character caseless the caseless flag cd the block with char table pointers - utf8 TRUE for UTF-8 mode + utf8 TRUE for UTF-8 mode Returns: pointer after the character */ @@ -471,15 +471,15 @@ if (utf8 && c > 127) #ifdef SUPPORT_UCP if (caseless) { - uschar buff[8]; + uschar buff[8]; c = UCD_OTHERCASE(c); - (void)_pcre_ord2utf8(c, buff); - SET_BIT(buff[0]); - } -#endif + (void)_pcre_ord2utf8(c, buff); + SET_BIT(buff[0]); + } +#endif return p; } -#endif +#endif /* Not UTF-8 mode, or character is less than 127. */ @@ -666,40 +666,40 @@ do (void)set_table_bit(start_bits, tcode + 1, caseless, cd, utf8); try_next = FALSE; break; - - /* Special spacing and line-terminating items. These recognize specific - lists of characters. The difference between VSPACE and ANYNL is that the - latter can match the two-character CRLF sequence, but that is not - relevant for finding the first character, so their code here is + + /* Special spacing and line-terminating items. These recognize specific + lists of characters. The difference between VSPACE and ANYNL is that the + latter can match the two-character CRLF sequence, but that is not + relevant for finding the first character, so their code here is identical. */ - + case OP_HSPACE: SET_BIT(0x09); SET_BIT(0x20); SET_BIT(0xA0); if (utf8) - { + { SET_BIT(0xE1); /* For U+1680, U+180E */ SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */ - SET_BIT(0xE3); /* For U+3000 */ + SET_BIT(0xE3); /* For U+3000 */ } try_next = FALSE; - break; + break; - case OP_ANYNL: + case OP_ANYNL: case OP_VSPACE: - SET_BIT(0x0A); - SET_BIT(0x0B); - SET_BIT(0x0C); - SET_BIT(0x0D); - SET_BIT(0x85); - if (utf8) SET_BIT(0xE2); /* For U+2028, U+2029 */ + SET_BIT(0x0A); + SET_BIT(0x0B); + SET_BIT(0x0C); + SET_BIT(0x0D); + SET_BIT(0x85); + if (utf8) SET_BIT(0xE2); /* For U+2028, U+2029 */ try_next = FALSE; - break; + break; - /* Single character types set the bits and stop. Note that if PCRE_UCP - is set, we do not see these op codes because \d etc are converted to - properties. Therefore, these apply in the case when only ASCII characters + /* Single character types set the bits and stop. Note that if PCRE_UCP + is set, we do not see these op codes because \d etc are converted to + properties. Therefore, these apply in the case when only ASCII characters are recognized to match the types. */ case OP_NOT_DIGIT: @@ -757,7 +757,7 @@ do case OP_TYPEPLUS: case OP_TYPEMINPLUS: - case OP_TYPEPOSPLUS: + case OP_TYPEPOSPLUS: tcode++; break; @@ -785,29 +785,29 @@ do case OP_ANY: case OP_ALLANY: return SSB_FAIL; - + case OP_HSPACE: SET_BIT(0x09); SET_BIT(0x20); SET_BIT(0xA0); if (utf8) - { + { SET_BIT(0xE1); /* For U+1680, U+180E */ SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */ - SET_BIT(0xE3); /* For U+3000 */ + SET_BIT(0xE3); /* For U+3000 */ } - break; - - case OP_ANYNL: + break; + + case OP_ANYNL: case OP_VSPACE: - SET_BIT(0x0A); - SET_BIT(0x0B); - SET_BIT(0x0C); - SET_BIT(0x0D); - SET_BIT(0x85); - if (utf8) SET_BIT(0xE2); /* For U+2028, U+2029 */ - break; - + SET_BIT(0x0A); + SET_BIT(0x0B); + SET_BIT(0x0C); + SET_BIT(0x0D); + SET_BIT(0x85); + if (utf8) SET_BIT(0xE2); /* For U+2028, U+2029 */ + break; + case OP_NOT_DIGIT: for (c = 0; c < 32; c++) start_bits[c] |= ~cd->cbits[c+cbit_digit]; |