| 1 | /* op_reg_common.h |
| 2 | * |
| 3 | * Definitions common to by op.h and regexp.h |
| 4 | * |
| 5 | * Copyright (C) 2010, 2011 by Larry Wall and others |
| 6 | * |
| 7 | * You may distribute under the terms of either the GNU General Public |
| 8 | * License or the Artistic License, as specified in the README file. |
| 9 | * |
| 10 | */ |
| 11 | |
| 12 | /* These defines are used in both op.h and regexp.h The definitions use the |
| 13 | * shift form so that ext/B/defsubs_h.PL will pick them up. |
| 14 | * |
| 15 | * Data structures used in the two headers have common fields, and in fact one |
| 16 | * is copied onto the other. This makes it easy to keep them in sync */ |
| 17 | |
| 18 | /* This tells where the first of these bits is. Setting it to 0 saved cycles |
| 19 | * and memory. I (khw) think the code will work if changed back, but haven't |
| 20 | * tested it */ |
| 21 | /* Make sure to update ext/re/re.pm when changing this! */ |
| 22 | #ifndef RXf_PMf_STD_PMMOD_SHIFT /* Only expand #include of this file once */ |
| 23 | |
| 24 | #define RXf_PMf_STD_PMMOD_SHIFT 0 |
| 25 | |
| 26 | /* The bits need to be ordered so that the msix are contiguous starting at bit |
| 27 | * RXf_PMf_STD_PMMOD_SHIFT, followed by the p. See STD_PAT_MODS and |
| 28 | * INT_PAT_MODS in regexp.h for the reason contiguity is needed */ |
| 29 | /* Make sure to update lib/re.pm when changing these! */ |
| 30 | /* Make sure you keep the pure PMf_ versions below in sync */ |
| 31 | #define RXf_PMf_MULTILINE (1 << (RXf_PMf_STD_PMMOD_SHIFT+0)) /* /m */ |
| 32 | #define RXf_PMf_SINGLELINE (1 << (RXf_PMf_STD_PMMOD_SHIFT+1)) /* /s */ |
| 33 | #define RXf_PMf_FOLD (1 << (RXf_PMf_STD_PMMOD_SHIFT+2)) /* /i */ |
| 34 | #define RXf_PMf_EXTENDED (1 << (RXf_PMf_STD_PMMOD_SHIFT+3)) /* /x */ |
| 35 | #define RXf_PMf_KEEPCOPY (1 << (RXf_PMf_STD_PMMOD_SHIFT+4)) /* /p */ |
| 36 | |
| 37 | /* The character set for the regex is stored in a field of more than one bit |
| 38 | * using an enum, for reasons of compactness and to ensure that the options are |
| 39 | * mutually exclusive */ |
| 40 | /* Make sure to update ext/re/re.pm and regcomp.sym (as these are used as |
| 41 | * offsets for various node types, like POSIXD vs POSIXL, etc) when changing |
| 42 | * this! */ |
| 43 | typedef enum { |
| 44 | REGEX_DEPENDS_CHARSET = 0, |
| 45 | REGEX_LOCALE_CHARSET, |
| 46 | REGEX_UNICODE_CHARSET, |
| 47 | REGEX_ASCII_RESTRICTED_CHARSET, |
| 48 | REGEX_ASCII_MORE_RESTRICTED_CHARSET |
| 49 | } regex_charset; |
| 50 | |
| 51 | #define _RXf_PMf_CHARSET_SHIFT ((RXf_PMf_STD_PMMOD_SHIFT)+5) |
| 52 | #define RXf_PMf_CHARSET (7 << (_RXf_PMf_CHARSET_SHIFT)) /* 3 bits */ |
| 53 | |
| 54 | /* Manually decorate these functions here with gcc-style attributes just to |
| 55 | * avoid making the regex_charset typedef global, which it would need to be for |
| 56 | * proto.h to understand it */ |
| 57 | PERL_STATIC_INLINE void |
| 58 | set_regex_charset(U32 * const flags, const regex_charset cs) |
| 59 | __attribute__nonnull__(1); |
| 60 | |
| 61 | PERL_STATIC_INLINE void |
| 62 | set_regex_charset(U32 * const flags, const regex_charset cs) |
| 63 | { |
| 64 | /* Sets the character set portion of 'flags' to 'cs', which is a member of |
| 65 | * the above enum */ |
| 66 | |
| 67 | *flags &= ~RXf_PMf_CHARSET; |
| 68 | *flags |= (cs << _RXf_PMf_CHARSET_SHIFT); |
| 69 | } |
| 70 | |
| 71 | PERL_STATIC_INLINE regex_charset |
| 72 | get_regex_charset(const U32 flags) |
| 73 | __attribute__warn_unused_result__; |
| 74 | |
| 75 | PERL_STATIC_INLINE regex_charset |
| 76 | get_regex_charset(const U32 flags) |
| 77 | { |
| 78 | /* Returns the enum corresponding to the character set in 'flags' */ |
| 79 | |
| 80 | return (regex_charset) ((flags & RXf_PMf_CHARSET) >> _RXf_PMf_CHARSET_SHIFT); |
| 81 | } |
| 82 | |
| 83 | #define _RXf_PMf_SHIFT_COMPILETIME (RXf_PMf_STD_PMMOD_SHIFT+8) |
| 84 | |
| 85 | /* |
| 86 | Set in Perl_pmruntime if op_flags & OPf_SPECIAL, i.e. split. Will |
| 87 | be used by regex engines to check whether they should set |
| 88 | RXf_SKIPWHITE |
| 89 | */ |
| 90 | #define RXf_PMf_SPLIT (1<<(RXf_PMf_STD_PMMOD_SHIFT+8)) |
| 91 | |
| 92 | /* Next available bit after the above. Name begins with '_' so won't be |
| 93 | * exported by B */ |
| 94 | #define _RXf_PMf_SHIFT_NEXT (RXf_PMf_STD_PMMOD_SHIFT+9) |
| 95 | |
| 96 | /* Mask of the above bits. These need to be transferred from op_pmflags to |
| 97 | * re->extflags during compilation */ |
| 98 | #define RXf_PMf_COMPILETIME (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_FOLD|RXf_PMf_EXTENDED|RXf_PMf_KEEPCOPY|RXf_PMf_CHARSET) |
| 99 | #define RXf_PMf_FLAGCOPYMASK (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_FOLD|RXf_PMf_EXTENDED|RXf_PMf_KEEPCOPY|RXf_PMf_CHARSET|RXf_PMf_SPLIT) |
| 100 | |
| 101 | #if RXf_PMf_COMPILETIME > 255 |
| 102 | # error RXf_PMf_COMPILETIME wont fit in U8 flags field of eval node |
| 103 | #endif |
| 104 | |
| 105 | /* These copies need to be numerical or defsubs_h.PL won't know about them. */ |
| 106 | #define PMf_MULTILINE 1<<0 |
| 107 | #define PMf_SINGLELINE 1<<1 |
| 108 | #define PMf_FOLD 1<<2 |
| 109 | #define PMf_EXTENDED 1<<3 |
| 110 | #define PMf_KEEPCOPY 1<<4 |
| 111 | #define PMf_CHARSET 7<<5 |
| 112 | #define PMf_SPLIT 1<<8 |
| 113 | |
| 114 | #if PMf_MULTILINE != RXf_PMf_MULTILINE || PMf_SINGLELINE != RXf_PMf_SINGLELINE || PMf_FOLD != RXf_PMf_FOLD || PMf_EXTENDED != RXf_PMf_EXTENDED || PMf_KEEPCOPY != RXf_PMf_KEEPCOPY || PMf_SPLIT != RXf_PMf_SPLIT || PMf_CHARSET != RXf_PMf_CHARSET |
| 115 | # error RXf_PMf defines are wrong |
| 116 | #endif |
| 117 | |
| 118 | /* Error check that haven't left something out of this. This isn't done |
| 119 | * directly in the #define because doing so confuses regcomp.pl. |
| 120 | * (2**n - 1) is n 1 bits, so the below gets the contiguous bits between the |
| 121 | * beginning and ending shifts */ |
| 122 | #if RXf_PMf_COMPILETIME != (((1 << (_RXf_PMf_SHIFT_COMPILETIME))-1) \ |
| 123 | & (~((1 << RXf_PMf_STD_PMMOD_SHIFT)-1))) |
| 124 | # error RXf_PMf_COMPILETIME is invalid |
| 125 | #endif |
| 126 | |
| 127 | #endif /* Include only once */ |