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 | |
5b126c84 KW |
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 */ | |
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! */ |
5b126c84 KW |
30 | #define RXf_PMf_MULTILINE (1 << (RXf_PMf_STD_PMMOD_SHIFT+0)) /* /m */ |
31 | #define RXf_PMf_SINGLELINE (1 << (RXf_PMf_STD_PMMOD_SHIFT+1)) /* /s */ | |
32 | #define RXf_PMf_FOLD (1 << (RXf_PMf_STD_PMMOD_SHIFT+2)) /* /i */ | |
33 | #define RXf_PMf_EXTENDED (1 << (RXf_PMf_STD_PMMOD_SHIFT+3)) /* /x */ | |
34 | #define RXf_PMf_KEEPCOPY (1 << (RXf_PMf_STD_PMMOD_SHIFT+4)) /* /p */ | |
a62b1201 KW |
35 | |
36 | /* The character set for the regex is stored in a field of more than one bit | |
37 | * using an enum, for reasons of compactness and to ensure that the options are | |
38 | * mutually exclusive */ | |
693fefec KW |
39 | /* Make sure to update ext/re/re.pm and regcomp.sym (as these are used as |
40 | * offsets for various node types, like SPACE vs SPACEL, etc) when changing | |
41 | * this! */ | |
a62b1201 KW |
42 | typedef enum { |
43 | REGEX_DEPENDS_CHARSET = 0, | |
44 | REGEX_LOCALE_CHARSET, | |
cfaf538b | 45 | REGEX_UNICODE_CHARSET, |
df7a8460 KW |
46 | REGEX_ASCII_RESTRICTED_CHARSET, |
47 | REGEX_ASCII_MORE_RESTRICTED_CHARSET | |
a62b1201 KW |
48 | } regex_charset; |
49 | ||
50 | #define _RXf_PMf_CHARSET_SHIFT ((RXf_PMf_STD_PMMOD_SHIFT)+5) | |
df7a8460 | 51 | #define RXf_PMf_CHARSET (7 << (_RXf_PMf_CHARSET_SHIFT)) /* 3 bits */ |
a62b1201 | 52 | |
759ba3f2 KW |
53 | /* Manually decorate these functions here with gcc-style attributes just to |
54 | * avoid making the regex_charset typedef global, which it would need to be for | |
55 | * proto.h to understand it */ | |
a62b1201 KW |
56 | PERL_STATIC_INLINE void |
57 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
58 | __attribute__nonnull__(1); | |
59 | ||
60 | PERL_STATIC_INLINE void | |
61 | set_regex_charset(U32 * const flags, const regex_charset cs) | |
62 | { | |
63 | /* Sets the character set portion of 'flags' to 'cs', which is a member of | |
64 | * the above enum */ | |
65 | ||
66 | *flags &= ~RXf_PMf_CHARSET; | |
67 | *flags |= (cs << _RXf_PMf_CHARSET_SHIFT); | |
68 | } | |
69 | ||
70 | PERL_STATIC_INLINE regex_charset | |
71 | get_regex_charset(const U32 flags) | |
72 | __attribute__warn_unused_result__; | |
73 | ||
74 | PERL_STATIC_INLINE regex_charset | |
75 | get_regex_charset(const U32 flags) | |
76 | { | |
77 | /* Returns the enum corresponding to the character set in 'flags' */ | |
78 | ||
bfba585a | 79 | return (regex_charset) ((flags & RXf_PMf_CHARSET) >> _RXf_PMf_CHARSET_SHIFT); |
a62b1201 | 80 | } |
1850c8f9 | 81 | |
3214c85f KW |
82 | /* Next available bit after the above. Name begins with '_' so won't be |
83 | * exported by B */ | |
df7a8460 | 84 | #define _RXf_PMf_SHIFT_NEXT (RXf_PMf_STD_PMMOD_SHIFT+8) |
bb1c6009 | 85 | |
5b126c84 KW |
86 | /* Mask of the above bits. These need to be transferred from op_pmflags to |
87 | * re->extflags during compilation */ | |
a62b1201 | 88 | #define RXf_PMf_COMPILETIME (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_CHARSET|RXf_PMf_FOLD|RXf_PMf_EXTENDED|RXf_PMf_KEEPCOPY) |
bb1c6009 | 89 | |
db703679 DM |
90 | #if RXf_PMf_COMPILETIME > 255 |
91 | # error RXf_PMf_COMPILETIME wont fit in U8 flags field of eval node | |
92 | #endif | |
93 | ||
bb1c6009 KW |
94 | /* These copies need to be numerical or defsubs_h.PL won't know about them. */ |
95 | #define PMf_MULTILINE 1<<0 | |
96 | #define PMf_SINGLELINE 1<<1 | |
97 | #define PMf_FOLD 1<<2 | |
98 | #define PMf_EXTENDED 1<<3 | |
99 | #define PMf_KEEPCOPY 1<<4 | |
bb1c6009 | 100 | |
a62b1201 | 101 | #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 |
bb1c6009 KW |
102 | # error RXf_PMf defines are wrong |
103 | #endif | |
5b126c84 KW |
104 | |
105 | #define PMf_COMPILETIME RXf_PMf_COMPILETIME | |
106 | ||
107 | /* Error check that haven't left something out of this. This isn't done | |
108 | * directly in the #define because doing so confuses regcomp.pl. | |
109 | * (2**n - 1) is n 1 bits, so the below gets the contiguous bits between the | |
110 | * beginning and ending shifts */ | |
111 | #if RXf_PMf_COMPILETIME != (((1 << (_RXf_PMf_SHIFT_NEXT))-1) \ | |
112 | & (~((1 << RXf_PMf_STD_PMMOD_SHIFT)-1))) | |
113 | # error RXf_PMf_COMPILETIME is invalid | |
114 | #endif | |
93136d17 KW |
115 | |
116 | #endif /* Include only once */ |