xref: /openbsd/gnu/usr.bin/perl/op_reg_common.h (revision e0680481)
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/Makefile.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 msixn 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      (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_NOCAPTURE      (1U << (RXf_PMf_STD_PMMOD_SHIFT+5))    /* /n */
37 
38 #define RXf_PMf_KEEPCOPY       (1U << (RXf_PMf_STD_PMMOD_SHIFT+6))    /* /p */
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 */
43 /* Make sure to update ext/re/re.pm and regcomp.sym (as these are used as
44  * offsets for various node types, like POSIXD vs POSIXL, etc) when changing
45  * this! */
46 typedef enum {
47     REGEX_DEPENDS_CHARSET = 0,
48     REGEX_LOCALE_CHARSET,
49     REGEX_UNICODE_CHARSET,
50     REGEX_ASCII_RESTRICTED_CHARSET,
51     REGEX_ASCII_MORE_RESTRICTED_CHARSET
52 } regex_charset;
53 
54 #define _RXf_PMf_CHARSET_SHIFT ((RXf_PMf_STD_PMMOD_SHIFT)+7)
55 #define RXf_PMf_CHARSET (7U << (_RXf_PMf_CHARSET_SHIFT)) /* 3 bits */
56 
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 */
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
set_regex_charset(U32 * const flags,const regex_charset cs)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
get_regex_charset(const U32 flags)79 get_regex_charset(const U32 flags)
80 {
81     /* Returns the enum corresponding to the character set in 'flags' */
82 
83     return (regex_charset) ((flags & RXf_PMf_CHARSET) >> _RXf_PMf_CHARSET_SHIFT);
84 }
85 
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 
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 */
96 #define RXf_PMf_SPLIT (1U<<(RXf_PMf_STD_PMMOD_SHIFT+11))
97 
98 /* Next available bit after the above.  Name begins with '_' so won't be
99  * exported by B */
100 #define _RXf_PMf_SHIFT_NEXT (RXf_PMf_STD_PMMOD_SHIFT+12)
101 
102 /* Mask of the above bits.  These need to be transferred from op_pmflags to
103  * re->extflags during compilation */
104 #define RXf_PMf_COMPILETIME \
105     ( RXf_PMf_MULTILINE     \
106     | RXf_PMf_SINGLELINE    \
107     | RXf_PMf_FOLD          \
108     | RXf_PMf_EXTENDED      \
109     | RXf_PMf_EXTENDED_MORE \
110     | RXf_PMf_KEEPCOPY      \
111     | RXf_PMf_NOCAPTURE     \
112     | RXf_PMf_CHARSET       \
113     | RXf_PMf_STRICT )
114 
115 #define RXf_PMf_FLAGCOPYMASK    \
116     ( RXf_PMf_COMPILETIME       \
117     | RXf_PMf_SPLIT )
118 
119 /* Temporary to get Jenkins happy again
120  * See thread starting at http://nntp.perl.org/group/perl.perl5.porters/220710
121  */
122 #if 0
123     /* Exclude win32 because it can't cope with I32_MAX definition */
124 #ifndef WIN32
125 #   if RXf_PMf_COMPILETIME > I32_MAX
126 #     error RXf_PMf_COMPILETIME wont fit in arg2 field of eval node
127 #   endif
128 #endif
129 #endif
130 
131 /* These copies need to be numerical or ext/B/Makefile.PL won't think they are
132  * constants */
133 #define PMf_MULTILINE     (1U<<0)
134 #define PMf_SINGLELINE    (1U<<1)
135 #define PMf_FOLD          (1U<<2)
136 #define PMf_EXTENDED      (1U<<3)
137 #define PMf_EXTENDED_MORE (1U<<4)
138 #define PMf_NOCAPTURE     (1U<<5)
139 #define PMf_KEEPCOPY      (1U<<6)
140 #define PMf_CHARSET       (7U<<7)
141 #define PMf_STRICT        (1U<<10)
142 #define PMf_SPLIT         (1U<<11)
143 
144 #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
145 #   error RXf_PMf defines are wrong
146 #endif
147 
148 /*  Error check that haven't left something out of this.  This isn't done
149  *  directly in the #define because doing so confuses regcomp.pl.
150  *  (2**n - 1) is n 1 bits, so the below gets the contiguous bits between the
151  *  beginning and ending shifts */
152 #if RXf_PMf_COMPILETIME != ((nBIT_MASK(_RXf_PMf_SHIFT_COMPILETIME)) \
153                         & (~(nBIT_MASK( RXf_PMf_STD_PMMOD_SHIFT))))
154 #   error RXf_PMf_COMPILETIME is invalid
155 #endif
156 
157 #endif /* Include only once */
158