Commit | Line | Data |
---|---|---|
1850c8f9 KW |
1 | /* op_reg_common.h |
2 | * | |
3 | * Definitions common to by op.h and regexp.h | |
4 | * | |
2eee27d7 | 5 | * Copyright (C) 2010, 2011 by Larry Wall and others |
1850c8f9 KW |
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 | |
bb62883e | 13 | * shift form so that ext/B/Makefile.PL will pick them up. |
5b126c84 KW |
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 */ | |
da494bb7 | 21 | /* Make sure to update ext/re/re.pm when changing this! */ |
93136d17 KW |
22 | #ifndef RXf_PMf_STD_PMMOD_SHIFT /* Only expand #include of this file once */ |
23 | ||
5b126c84 KW |
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 */ | |
1e215989 | 29 | /* Make sure to update lib/re.pm when changing these! */ |
dbc200c5 | 30 | /* Make sure you keep the pure PMf_ versions below in sync */ |
9cba692b KW |
31 | #define RXf_PMf_MULTILINE (1U << (RXf_PMf_STD_PMMOD_SHIFT+0)) /* /m */ |
32 | #define RXf_PMf_SINGLELINE (1U << (RXf_PMf_STD_PMMOD_SHIFT+1)) /* /s */ | |
33 | #define RXf_PMf_FOLD (1U << (RXf_PMf_STD_PMMOD_SHIFT+2)) /* /i */ | |
34 | #define RXf_PMf_EXTENDED (1U << (RXf_PMf_STD_PMMOD_SHIFT+3)) /* /x */ | |
35 | #define RXf_PMf_EXTENDED_MORE (1U << (RXf_PMf_STD_PMMOD_SHIFT+4)) /* /xx */ | |
36 | #define RXf_PMf_KEEPCOPY (1U << (RXf_PMf_STD_PMMOD_SHIFT+5)) /* /p */ | |
a62b1201 KW |
37 | |
38 | /* The character set for the regex is stored in a field of more than one bit | |
39 | * using an enum, for reasons of compactness and to ensure that the options are | |
40 | * mutually exclusive */ | |
693fefec | 41 | /* Make sure to update ext/re/re.pm and regcomp.sym (as these are used as |
ca143fe8 | 42 | * offsets for various node types, like POSIXD vs POSIXL, etc) when changing |
693fefec | 43 | * this! */ |
a62b1201 KW |
44 | typedef enum { |
45 | REGEX_DEPENDS_CHARSET = 0, | |
46 | REGEX_LOCALE_CHARSET, | |
cfaf538b | 47 | REGEX_UNICODE_CHARSET, |
df7a8460 KW |
48 | REGEX_ASCII_RESTRICTED_CHARSET, |
49 | REGEX_ASCII_MORE_RESTRICTED_CHARSET | |
a62b1201 KW |
50 | } regex_charset; |
51 | ||
334afb3e | 52 | #define _RXf_PMf_CHARSET_SHIFT ((RXf_PMf_STD_PMMOD_SHIFT)+6) |
7641d82c | 53 | #define RXf_PMf_CHARSET (7U << (_RXf_PMf_CHARSET_SHIFT)) /* 3 bits */ |
a62b1201 | 54 | |
759ba3f2 KW |
55 | /* Manually decorate these functions here with gcc-style attributes just to |
56 | * avoid making the regex_charset typedef global, which it would need to be for | |
57 | * proto.h to understand it */ | |
a62b1201 KW |
58 | PERL_STATIC_INLINE void |
59 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
60 | __attribute__nonnull__(1); | |
61 | ||
62 | PERL_STATIC_INLINE void | |
63 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
64 | { | |
65 | /* Sets the character set portion of 'flags' to 'cs', which is a member of | |
66 | * the above enum */ | |
67 | ||
68 | *flags &= ~RXf_PMf_CHARSET; | |
69 | *flags |= (cs << _RXf_PMf_CHARSET_SHIFT); | |
70 | } | |
71 | ||
72 | PERL_STATIC_INLINE regex_charset | |
73 | get_regex_charset(const U32 flags) | |
74 | __attribute__warn_unused_result__; | |
75 | ||
76 | PERL_STATIC_INLINE regex_charset | |
77 | get_regex_charset(const U32 flags) | |
78 | { | |
79 | /* Returns the enum corresponding to the character set in 'flags' */ | |
80 | ||
bfba585a | 81 | return (regex_charset) ((flags & RXf_PMf_CHARSET) >> _RXf_PMf_CHARSET_SHIFT); |
a62b1201 | 82 | } |
1850c8f9 | 83 | |
334afb3e | 84 | #define _RXf_PMf_SHIFT_COMPILETIME (RXf_PMf_STD_PMMOD_SHIFT+9) |
dbc200c5 YO |
85 | |
86 | /* | |
87 | Set in Perl_pmruntime if op_flags & OPf_SPECIAL, i.e. split. Will | |
88 | be used by regex engines to check whether they should set | |
89 | RXf_SKIPWHITE | |
90 | */ | |
9cba692b | 91 | #define RXf_PMf_SPLIT (1U<<(RXf_PMf_STD_PMMOD_SHIFT+9)) |
dbc200c5 | 92 | |
3214c85f KW |
93 | /* Next available bit after the above. Name begins with '_' so won't be |
94 | * exported by B */ | |
334afb3e | 95 | #define _RXf_PMf_SHIFT_NEXT (RXf_PMf_STD_PMMOD_SHIFT+10) |
bb1c6009 | 96 | |
5b126c84 KW |
97 | /* Mask of the above bits. These need to be transferred from op_pmflags to |
98 | * re->extflags during compilation */ | |
334afb3e | 99 | #define RXf_PMf_COMPILETIME (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_FOLD|RXf_PMf_EXTENDED|RXf_PMf_EXTENDED_MORE|RXf_PMf_KEEPCOPY|RXf_PMf_CHARSET) |
9a7d4f5f | 100 | #define RXf_PMf_FLAGCOPYMASK (RXf_PMf_COMPILETIME|RXf_PMf_SPLIT) |
bb1c6009 | 101 | |
59e279a1 | 102 | #if 0 /* Temporary to get Jenkins happy again */ |
13f27704 KW |
103 | /* Exclude win32 because it can't cope with I32_MAX definition */ |
104 | #ifndef WIN32 | |
105 | # if RXf_PMf_COMPILETIME > I32_MAX | |
106 | # error RXf_PMf_COMPILETIME wont fit in arg2 field of eval node | |
107 | # endif | |
db703679 | 108 | #endif |
59e279a1 | 109 | #endif |
db703679 | 110 | |
bb62883e KW |
111 | /* These copies need to be numerical or ext/B/Makefile.PL won't think they are |
112 | * constants */ | |
7641d82c KW |
113 | #define PMf_MULTILINE (1U<<0) |
114 | #define PMf_SINGLELINE (1U<<1) | |
115 | #define PMf_FOLD (1U<<2) | |
116 | #define PMf_EXTENDED (1U<<3) | |
117 | #define PMf_EXTENDED_MORE (1U<<4) | |
118 | #define PMf_KEEPCOPY (1U<<5) | |
119 | #define PMf_CHARSET (7U<<6) | |
120 | #define PMf_SPLIT (1U<<9) | |
bb1c6009 | 121 | |
334afb3e | 122 | #if PMf_MULTILINE != RXf_PMf_MULTILINE || PMf_SINGLELINE != RXf_PMf_SINGLELINE || PMf_FOLD != RXf_PMf_FOLD || PMf_EXTENDED != RXf_PMf_EXTENDED || PMf_EXTENDED_MORE != RXf_PMf_EXTENDED_MORE || PMf_KEEPCOPY != RXf_PMf_KEEPCOPY || PMf_SPLIT != RXf_PMf_SPLIT || PMf_CHARSET != RXf_PMf_CHARSET |
bb1c6009 KW |
123 | # error RXf_PMf defines are wrong |
124 | #endif | |
5b126c84 | 125 | |
5b126c84 KW |
126 | /* Error check that haven't left something out of this. This isn't done |
127 | * directly in the #define because doing so confuses regcomp.pl. | |
128 | * (2**n - 1) is n 1 bits, so the below gets the contiguous bits between the | |
129 | * beginning and ending shifts */ | |
dbc200c5 | 130 | #if RXf_PMf_COMPILETIME != (((1 << (_RXf_PMf_SHIFT_COMPILETIME))-1) \ |
5b126c84 KW |
131 | & (~((1 << RXf_PMf_STD_PMMOD_SHIFT)-1))) |
132 | # error RXf_PMf_COMPILETIME is invalid | |
133 | #endif | |
93136d17 KW |
134 | |
135 | #endif /* Include only once */ |