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 | ||
33be4c61 | 26 | /* The bits need to be ordered so that the msixn are contiguous starting at bit |
5b126c84 KW |
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 */ | |
e3b64d84 KW |
36 | #define RXf_PMf_NOCAPTURE (1U << (RXf_PMf_STD_PMMOD_SHIFT+5)) /* /n */ |
37 | ||
38 | #define RXf_PMf_KEEPCOPY (1U << (RXf_PMf_STD_PMMOD_SHIFT+6)) /* /p */ | |
a62b1201 KW |
39 | |
40 | /* The character set for the regex is stored in a field of more than one bit | |
41 | * using an enum, for reasons of compactness and to ensure that the options are | |
42 | * mutually exclusive */ | |
693fefec | 43 | /* Make sure to update ext/re/re.pm and regcomp.sym (as these are used as |
ca143fe8 | 44 | * offsets for various node types, like POSIXD vs POSIXL, etc) when changing |
693fefec | 45 | * this! */ |
a62b1201 KW |
46 | typedef enum { |
47 | REGEX_DEPENDS_CHARSET = 0, | |
48 | REGEX_LOCALE_CHARSET, | |
cfaf538b | 49 | REGEX_UNICODE_CHARSET, |
df7a8460 KW |
50 | REGEX_ASCII_RESTRICTED_CHARSET, |
51 | REGEX_ASCII_MORE_RESTRICTED_CHARSET | |
a62b1201 KW |
52 | } regex_charset; |
53 | ||
e3b64d84 | 54 | #define _RXf_PMf_CHARSET_SHIFT ((RXf_PMf_STD_PMMOD_SHIFT)+7) |
7641d82c | 55 | #define RXf_PMf_CHARSET (7U << (_RXf_PMf_CHARSET_SHIFT)) /* 3 bits */ |
a62b1201 | 56 | |
759ba3f2 KW |
57 | /* Manually decorate these functions here with gcc-style attributes just to |
58 | * avoid making the regex_charset typedef global, which it would need to be for | |
59 | * proto.h to understand it */ | |
a62b1201 KW |
60 | PERL_STATIC_INLINE void |
61 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
62 | __attribute__nonnull__(1); | |
63 | ||
64 | PERL_STATIC_INLINE void | |
65 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
66 | { | |
67 | /* Sets the character set portion of 'flags' to 'cs', which is a member of | |
68 | * the above enum */ | |
69 | ||
70 | *flags &= ~RXf_PMf_CHARSET; | |
71 | *flags |= (cs << _RXf_PMf_CHARSET_SHIFT); | |
72 | } | |
73 | ||
74 | PERL_STATIC_INLINE regex_charset | |
75 | get_regex_charset(const U32 flags) | |
76 | __attribute__warn_unused_result__; | |
77 | ||
78 | PERL_STATIC_INLINE regex_charset | |
79 | get_regex_charset(const U32 flags) | |
80 | { | |
81 | /* Returns the enum corresponding to the character set in 'flags' */ | |
82 | ||
bfba585a | 83 | return (regex_charset) ((flags & RXf_PMf_CHARSET) >> _RXf_PMf_CHARSET_SHIFT); |
a62b1201 | 84 | } |
1850c8f9 | 85 | |
d262c0c7 KW |
86 | #define RXf_PMf_STRICT (1U<<(RXf_PMf_STD_PMMOD_SHIFT+10)) |
87 | ||
88 | #define _RXf_PMf_SHIFT_COMPILETIME (RXf_PMf_STD_PMMOD_SHIFT+11) | |
89 | ||
dbc200c5 YO |
90 | |
91 | /* | |
92 | Set in Perl_pmruntime if op_flags & OPf_SPECIAL, i.e. split. Will | |
93 | be used by regex engines to check whether they should set | |
94 | RXf_SKIPWHITE | |
95 | */ | |
d262c0c7 | 96 | #define RXf_PMf_SPLIT (1U<<(RXf_PMf_STD_PMMOD_SHIFT+11)) |
dbc200c5 | 97 | |
3214c85f KW |
98 | /* Next available bit after the above. Name begins with '_' so won't be |
99 | * exported by B */ | |
d262c0c7 | 100 | #define _RXf_PMf_SHIFT_NEXT (RXf_PMf_STD_PMMOD_SHIFT+12) |
bb1c6009 | 101 | |
5b126c84 KW |
102 | /* Mask of the above bits. These need to be transferred from op_pmflags to |
103 | * re->extflags during compilation */ | |
d262c0c7 | 104 | #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_NOCAPTURE|RXf_PMf_CHARSET|RXf_PMf_STRICT) |
9a7d4f5f | 105 | #define RXf_PMf_FLAGCOPYMASK (RXf_PMf_COMPILETIME|RXf_PMf_SPLIT) |
bb1c6009 | 106 | |
ac892e4a DM |
107 | /* Temporary to get Jenkins happy again |
108 | * See thread starting at http://nntp.perl.org/group/perl.perl5.porters/220710 | |
109 | */ | |
110 | #if 0 | |
13f27704 KW |
111 | /* Exclude win32 because it can't cope with I32_MAX definition */ |
112 | #ifndef WIN32 | |
113 | # if RXf_PMf_COMPILETIME > I32_MAX | |
114 | # error RXf_PMf_COMPILETIME wont fit in arg2 field of eval node | |
115 | # endif | |
db703679 | 116 | #endif |
59e279a1 | 117 | #endif |
db703679 | 118 | |
bb62883e KW |
119 | /* These copies need to be numerical or ext/B/Makefile.PL won't think they are |
120 | * constants */ | |
7641d82c KW |
121 | #define PMf_MULTILINE (1U<<0) |
122 | #define PMf_SINGLELINE (1U<<1) | |
123 | #define PMf_FOLD (1U<<2) | |
124 | #define PMf_EXTENDED (1U<<3) | |
125 | #define PMf_EXTENDED_MORE (1U<<4) | |
e3b64d84 KW |
126 | #define PMf_NOCAPTURE (1U<<5) |
127 | #define PMf_KEEPCOPY (1U<<6) | |
128 | #define PMf_CHARSET (7U<<7) | |
d262c0c7 KW |
129 | #define PMf_STRICT (1U<<10) |
130 | #define PMf_SPLIT (1U<<11) | |
bb1c6009 | 131 | |
d262c0c7 | 132 | #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 || PMf_NOCAPTURE != RXf_PMf_NOCAPTURE || PMf_STRICT != RXf_PMf_STRICT |
bb1c6009 KW |
133 | # error RXf_PMf defines are wrong |
134 | #endif | |
5b126c84 | 135 | |
5b126c84 KW |
136 | /* Error check that haven't left something out of this. This isn't done |
137 | * directly in the #define because doing so confuses regcomp.pl. | |
138 | * (2**n - 1) is n 1 bits, so the below gets the contiguous bits between the | |
139 | * beginning and ending shifts */ | |
d223e1ea KW |
140 | #if RXf_PMf_COMPILETIME != ((nBIT_MASK(_RXf_PMf_SHIFT_COMPILETIME)) \ |
141 | & (~(nBIT_MASK( RXf_PMf_STD_PMMOD_SHIFT)))) | |
5b126c84 KW |
142 | # error RXf_PMf_COMPILETIME is invalid |
143 | #endif | |
93136d17 KW |
144 | |
145 | #endif /* Include only once */ |