Commit | Line | Data |
---|---|---|
03363afd YO |
1 | # regcomp.sym |
2 | # | |
3 | # File has two sections, divided by a line of dashes '-'. | |
4 | # | |
65aa4ca7 FC |
5 | # Lines beginning with # are ignored, except for those that start with #* |
6 | # which are included in pod/perldebguts.pod. # within a line may be part | |
7 | # of a description. | |
03363afd | 8 | # |
486ec47a | 9 | # First section is for regops, second section is for regmatch-states |
03363afd | 10 | # |
3dab1dad YO |
11 | # Note that the order in this file is important. |
12 | # | |
03363afd | 13 | # Format for first section: |
e21ef692 KW |
14 | # NAME \s+ TYPE, arg-description [struct regnode suffix] [flags] [longjump] ; DESCRIPTION |
15 | # arg-description is currently unused | |
16 | # suffix is appended to 'struct_regnode_' giving which one to use. If empty, | |
17 | # it means plain 'struct regnode'. If the regnode is a string one, this | |
18 | # should instead refer to the base regnode, without the char[1] element | |
19 | # of the structure | |
46167d76 | 20 | # flag <S> means is REGNODE_SIMPLE; flag <V> means is REGNODE_VARIES; <.> is |
e21ef692 KW |
21 | # a placeholder |
22 | # longjump is 1 if the (first) argument holds the next offset (instead of the | |
23 | # usual 'next_offset' field | |
3dab1dad | 24 | # |
c476f425 | 25 | # run perl regen.pl after editing this file |
3dab1dad | 26 | |
e21ef692 KW |
27 | # +- suffix of which struct regnode to use e.g., |
28 | # | +- flags (S or V) struct regnode_1 | |
29 | # un- | | +- longjmp (0, blank, or 1) blank means 0 | |
30 | # Name Type used | | | ; comment | |
31 | # -------------------------------------------------------------------------- | |
32 | # IFMATCH BRANCHJ, off 1 . 1 ; Succeeds if the following matches. | |
33 | # UNLESSM BRANCHJ, off 1 . 1 ; Fails if the following matches. | |
34 | # SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE. | |
35 | # IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher. | |
36 | # GROUPP GROUPP, num 1 ; Whether the group matched. | |
03363afd YO |
37 | |
38 | ||
381b57db | 39 | #* Exit points |
1de06328 | 40 | |
f8abb37e NC |
41 | END END, no ; End of program. |
42 | SUCCEED END, no ; Return from a subroutine, basically. | |
d09b2d29 | 43 | |
d3d47aac | 44 | #* Line Start Anchors: |
1645b83c | 45 | #Note flags field for SBOL indicates if it is a /^/ or a /\A/ |
d3d47aac YO |
46 | SBOL BOL, no ; Match "" at beginning of line: /^/, /\A/ |
47 | MBOL BOL, no ; Same, assuming multiline: /^/m | |
48 | ||
49 | #* Line End Anchors: | |
50 | SEOL EOL, no ; Match "" at end of line: /$/ | |
51 | MEOL EOL, no ; Same, assuming multiline: /$/m | |
52 | EOS EOL, no ; Match "" at end of string: /\z/ | |
53 | ||
54 | #* Match Start Anchors: | |
55 | GPOS GPOS, no ; Matches where last m//g left off. | |
56 | ||
57 | #* Word Boundary Opcodes: | |
693fefec KW |
58 | # The regops that have varieties that vary depending on the character set regex |
59 | # modifiers have to ordered thusly: /d, /l, /u, /a, /aa. This is because code | |
60 | # in regcomp.c uses the enum value of the modifier as an offset from the /d | |
61 | # version. The complements must come after the non-complements. | |
3018b823 | 62 | # BOUND, POSIX and their complements are affected, as well as EXACTF. |
c440a570 KW |
63 | BOUND BOUND, no ; Like BOUNDA for non-utf8, otherwise match "" between any Unicode \w\W or \W\w |
64 | BOUNDL BOUND, no ; Like BOUND/BOUNDU, but \w and \W are defined by current locale | |
912b808c | 65 | BOUNDU BOUND, no ; Match "" at any boundary of a given type using /u rules. |
c440a570 | 66 | BOUNDA BOUND, no ; Match "" at any boundary between \w\W or \W\w, where \w is [_a-zA-Z0-9] |
693fefec | 67 | # All NBOUND nodes are required by code in regexec.c to be greater than all BOUND ones |
c440a570 KW |
68 | NBOUND NBOUND, no ; Like NBOUNDA for non-utf8, otherwise match "" between any Unicode \w\w or \W\W |
69 | NBOUNDL NBOUND, no ; Like NBOUND/NBOUNDU, but \w and \W are defined by current locale | |
912b808c | 70 | NBOUNDU NBOUND, no ; Match "" at any non-boundary of a given type using using /u rules. |
c440a570 | 71 | NBOUNDA NBOUND, no ; Match "" betweeen any \w\w or \W\W, where \w is [_a-zA-Z0-9] |
d09b2d29 | 72 | |
381b57db | 73 | #* [Special] alternatives: |
f9ef50a7 NC |
74 | REG_ANY REG_ANY, no 0 S ; Match any one character (except newline). |
75 | SANY REG_ANY, no 0 S ; Match any one character. | |
46fc0c43 KW |
76 | ANYOF ANYOF, sv charclass S ; Match character in (or not in) this class, single char match only |
77 | ANYOFD ANYOF, sv charclass S ; Like ANYOF, but /d is in effect | |
78 | ANYOFL ANYOF, sv charclass S ; Like ANYOF, but /l is in effect | |
79 | ANYOFPOSIXL ANYOF, sv charclass_posixl S ; Like ANYOFL, but matches [[:posix:]] classes | |
f6eaa562 KW |
80 | |
81 | # Must be sequential | |
29a889ef | 82 | ANYOFH ANYOF, sv 1 S ; Like ANYOF, but only has "High" matches, none in the bitmap; the flags field contains the lowest matchable UTF-8 start byte |
a5bc0742 | 83 | ANYOFHb ANYOF, sv 1 S ; Like ANYOFH, but all matches share the same UTF-8 start byte, given in the flags field |
3146c00a | 84 | ANYOFHr ANYOF, sv 1 S ; Like ANYOFH, but the flags field contains packed bounds for all matchable UTF-8 start bytes. |
f6eaa562 | 85 | |
67a1b5f9 | 86 | ANYOFM ANYOFM byte 1 S ; Like ANYOF, but matches an invariant byte as determined by the mask and arg |
3db0bccc | 87 | NANYOFM ANYOFM byte 1 S ; complement of ANYOFM |
693fefec | 88 | |
d3d47aac | 89 | #* POSIX Character Classes: |
3018b823 KW |
90 | # Order of the below is important. See ordering comment above. |
91 | POSIXD POSIXD, none 0 S ; Some [[:class:]] under /d; the FLAGS field gives which one | |
92 | POSIXL POSIXD, none 0 S ; Some [[:class:]] under /l; the FLAGS field gives which one | |
93 | POSIXU POSIXD, none 0 S ; Some [[:class:]] under /u; the FLAGS field gives which one | |
3615ea58 | 94 | POSIXA POSIXD, none 0 S ; Some [[:class:]] under /a; the FLAGS field gives which one |
3018b823 KW |
95 | NPOSIXD NPOSIXD, none 0 S ; complement of POSIXD, [[:^class:]] |
96 | NPOSIXL NPOSIXD, none 0 S ; complement of POSIXL, [[:^class:]] | |
97 | NPOSIXU NPOSIXD, none 0 S ; complement of POSIXU, [[:^class:]] | |
9e84774b | 98 | NPOSIXA NPOSIXD, none 0 S ; complement of POSIXA, [[:^class:]] |
3018b823 | 99 | # End of order is important |
693fefec | 100 | |
2448cf39 | 101 | CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence |
d09b2d29 | 102 | |
381b57db | 103 | #* Alternation |
1de06328 | 104 | |
65aa4ca7 FC |
105 | #* BRANCH The set of branches constituting a single choice are |
106 | #* hooked together with their "next" pointers, since | |
107 | #* precedence prevents anything being concatenated to | |
108 | #* any individual branch. The "next" pointer of the last | |
109 | #* BRANCH in a choice points to the thing following the | |
110 | #* whole choice. This is also where the final "next" | |
111 | #* pointer of each individual branch points; each branch | |
112 | #* starts with the operand node of a BRANCH node. | |
113 | #* | |
f9ef50a7 | 114 | BRANCH BRANCH, node 0 V ; Match this alternative, or the next... |
d09b2d29 | 115 | |
65aa4ca7 FC |
116 | #*Literals |
117 | # NOTE: the relative ordering of these types is important do not change it | |
1de06328 | 118 | |
3ace85ea | 119 | EXACT EXACT, str ; Match this string (flags field is the length). |
ae06e581 KW |
120 | |
121 | #* In a long string node, the U32 argument is the length, and is | |
122 | #* immediately followed by the string. | |
123 | LEXACT EXACT, len:str 1; Match this long string (preceded by length; flags unused). | |
b24abbc8 | 124 | EXACTL EXACT, str ; Like EXACT, but /l is in effect (used so locale-related warnings can be checked for). |
a2f213ef KW |
125 | EXACTF EXACT, str ; Like EXACT, but match using /id rules; (string not UTF-8, not guaranteed to be folded). |
126 | EXACTFL EXACT, str ; Like EXACT, but match using /il rules; (string not likely to be folded). | |
127 | EXACTFU EXACT, str ; Like EXACT, but match using /iu rules; (string folded). | |
128 | EXACTFAA EXACT, str ; Like EXACT, but match using /iaa rules; (string folded iff pattern is UTF8; folded length <= unfolded). | |
2f306ab9 KW |
129 | |
130 | # End of important relative ordering. | |
131 | ||
a2f213ef | 132 | EXACTFUP EXACT, str ; Like EXACT, but match using /iu rules; (string not UTF-8, not guaranteed to be folded; and it is Problematic). |
aa419ff3 KW |
133 | # In order for a non-UTF-8 EXACTFAA to think the pattern is pre-folded when |
134 | # matching a UTF-8 target string, there would have to be something like an | |
135 | # EXACTFAA_MICRO which would not be considered pre-folded for UTF-8 targets, | |
136 | # since the fold of the MICRO SIGN would not be done, and would be | |
137 | # representable in the UTF-8 target string. | |
138 | ||
a2f213ef KW |
139 | EXACTFLU8 EXACT, str ; Like EXACTFU, but use /il, UTF-8, (string is folded, and everything in it is above 255. |
140 | EXACTFAA_NO_TRIE EXACT, str ; Like EXACT, but match using /iaa rules (string not UTF-8, not guaranteed to be folded, not currently trie-able). | |
d09b2d29 | 141 | |
4f4c2c24 | 142 | |
3f2416ae KW |
143 | EXACT_REQ8 EXACT, str ; Like EXACT, but only UTF-8 encoded targets can match |
144 | LEXACT_REQ8 EXACT, len:str 1 ; Like LEXACT, but only UTF-8 encoded targets can match | |
145 | EXACTFU_REQ8 EXACT, str ; Like EXACTFU, but only UTF-8 encoded targets can match | |
efec1f81 | 146 | # One could add EXACTFAA8 and something that has the same effect for /l, |
a9f8c7ac | 147 | # but these would be extremely uncommon |
f6b4b99d | 148 | |
95fb0a6e | 149 | EXACTFU_S_EDGE EXACT, str ; /di rules, but nothing in it precludes /ui, except begins and/or ends with [Ss]; (string not UTF-8; compile-time only). |
8a100c91 | 150 | |
381b57db | 151 | #*Do nothing types |
1de06328 | 152 | |
f8abb37e | 153 | NOTHING NOTHING, no ; Match empty string. |
65aa4ca7 | 154 | #*A variant of above which delimits a group, thus stops optimizations |
f8abb37e | 155 | TAIL NOTHING, no ; Match empty string. Can jump here from outside. |
d09b2d29 | 156 | |
381b57db | 157 | #*Loops |
1de06328 | 158 | |
65aa4ca7 | 159 | #* STAR,PLUS '?', and complex '*' and '+', are implemented as |
62e6ef33 | 160 | #* circular BRANCH structures. Simple cases |
65aa4ca7 FC |
161 | #* (one character per match) are implemented with STAR |
162 | #* and PLUS for speed and to minimize recursive plunges. | |
163 | #* | |
f9ef50a7 NC |
164 | STAR STAR, node 0 V ; Match this (simple) thing 0 or more times. |
165 | PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times. | |
d09b2d29 | 166 | |
f9ef50a7 NC |
167 | CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times. |
168 | CURLYN CURLY, no 2 V ; Capture next-after-this simple thing | |
169 | CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times. | |
170 | CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times. | |
d09b2d29 | 171 | |
65aa4ca7 | 172 | #*This terminator creates a loop structure for CURLYX |
f9ef50a7 | 173 | WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches. |
d09b2d29 | 174 | |
381b57db | 175 | #*Buffer related |
1de06328 | 176 | |
65aa4ca7 | 177 | #*OPEN,CLOSE,GROUPP ...are numbered at compile time. |
f8abb37e | 178 | OPEN OPEN, num 1 ; Mark this point in input as start of #n. |
48e88d42 | 179 | CLOSE CLOSE, num 1 ; Close corresponding OPEN of #n. |
07093db4 KW |
180 | SROPEN SROPEN, none ; Same as OPEN, but for script run |
181 | SRCLOSE SRCLOSE, none ; Close preceding SROPEN | |
d09b2d29 | 182 | |
f9ef50a7 | 183 | REF REF, num 1 V ; Match some already matched string |
912b808c KW |
184 | REFF REF, num 1 V ; Match already matched string, using /di rules. |
185 | REFFL REF, num 1 V ; Match already matched string, using /li rules. | |
781aab5c | 186 | # N?REFF[AU] could have been implemented using the FLAGS field of the |
01f98ec2 KW |
187 | # regnode, but by having a separate node type, we can use the existing switch |
188 | # statement to avoid some tests | |
912b808c KW |
189 | REFFU REF, num 1 V ; Match already matched string, usng /ui. |
190 | REFFA REF, num 1 V ; Match already matched string, using /aai rules. | |
d09b2d29 | 191 | |
65aa4ca7 FC |
192 | #*Named references. Code in regcomp.c assumes that these all are after |
193 | #*the numbered references | |
016b7209 KW |
194 | REFN REF, no-sv 1 V ; Match some already matched string |
195 | REFFN REF, no-sv 1 V ; Match already matched string, using /di rules. | |
196 | REFFLN REF, no-sv 1 V ; Match already matched string, using /li rules. | |
197 | REFFUN REF, num 1 V ; Match already matched string, using /ui rules. | |
198 | REFFAN REF, num 1 V ; Match already matched string, using /aai rules. | |
1de06328 | 199 | |
d3d47aac YO |
200 | #*Support for long RE |
201 | LONGJMP LONGJMP, off 1 . 1 ; Jump far away. | |
202 | BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset. | |
203 | ||
204 | #*Special Case Regops | |
2abbd513 KW |
205 | IFMATCH BRANCHJ, off 1 . 1 ; Succeeds if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current |
206 | UNLESSM BRANCHJ, off 1 . 1 ; Fails if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current | |
f9ef50a7 | 207 | SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE. |
65aa4ca7 | 208 | IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher. |
f8abb37e | 209 | GROUPP GROUPP, num 1 ; Whether the group matched. |
d09b2d29 | 210 | |
d09b2d29 | 211 | |
381b57db | 212 | #*The heavy worker |
1de06328 | 213 | |
13f27704 | 214 | EVAL EVAL, evl/flags 2L ; Execute some Perl code. |
d09b2d29 | 215 | |
381b57db | 216 | #*Modifiers |
1de06328 | 217 | |
f8abb37e NC |
218 | MINMOD MINMOD, no ; Next operator is not greedy. |
219 | LOGICAL LOGICAL, no ; Next opcode should set the flag only. | |
d09b2d29 | 220 | |
65aa4ca7 | 221 | #*This is not used yet |
f9ef50a7 | 222 | RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens. |
d09b2d29 | 223 | |
381b57db | 224 | #*Trie Related |
1de06328 | 225 | |
65aa4ca7 FC |
226 | #* Behave the same as A|LIST|OF|WORDS would. The '..C' variants |
227 | #* have inline charclass data (ascii only), the 'C' store it in the | |
228 | #* structure. | |
486ec47a | 229 | # NOTE: the relative order of the TRIE-like regops is significant |
ce5e9471 | 230 | |
7986cb47 | 231 | TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type |
f8abb37e | 232 | TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data |
3dab1dad | 233 | |
1de06328 | 234 | # For start classes, contains an added fail table. |
f8abb37e NC |
235 | AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type |
236 | AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data | |
1de06328 | 237 | |
381b57db | 238 | #*Regex Subroutines |
f8abb37e | 239 | GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2 |
03363afd | 240 | |
381b57db | 241 | #*Special conditionals |
016b7209 | 242 | GROUPPN GROUPPN, no-sv 1 ; Whether the group matched. |
f8abb37e NC |
243 | INSUBP INSUBP, num 1 ; Whether we are in a specific recurse. |
244 | DEFINEP DEFINEP, none 1 ; Never execute directly. | |
0a4db386 | 245 | |
486ec47a | 246 | #*Backtracking Verbs |
f8abb37e | 247 | ENDLIKE ENDLIKE, none ; Used only for the type field of verbs |
fee50582 YO |
248 | OPFAIL ENDLIKE, no-sv 1 ; Same as (?!), but with verb arg |
249 | ACCEPT ENDLIKE, no-sv/num 2L ; Accepts the current matched string, with verbar | |
5d458dd8 YO |
250 | |
251 | #*Verbs With Arguments | |
f8abb37e NC |
252 | VERB VERB, no-sv 1 ; Used only for the type field of verbs |
253 | PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this | |
254 | MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut. | |
255 | SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying | |
256 | COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this | |
257 | CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group | |
e2e6a0f1 | 258 | |
ee9b8eae | 259 | #*Control what to keep in $&. |
f8abb37e | 260 | KEEPS KEEPS, no ; $& begins here. |
ee9b8eae | 261 | |
e1d1eefb | 262 | #*New charclass like patterns |
f8abb37e | 263 | LNBREAK LNBREAK, none ; generic newline pattern |
3172e3fd | 264 | |
381b57db | 265 | # NEW STUFF SOMEWHERE ABOVE THIS LINE |
1de06328 | 266 | |
03363afd YO |
267 | ################################################################################ |
268 | ||
7f69552c | 269 | #*SPECIAL REGOPS |
1de06328 | 270 | |
65aa4ca7 FC |
271 | #* This is not really a node, but an optimized away piece of a "long" |
272 | #* node. To simplify debugging output, we mark it as if it were a node | |
f8abb37e | 273 | OPTIMIZED NOTHING, off ; Placeholder for dump. |
1de06328 | 274 | |
65aa4ca7 FC |
275 | #* Special opcode with the property that no opcode in a compiled program |
276 | #* will ever be of this type. Thus it can be used as a flag value that | |
277 | #* no other opcode has been seen. END is used similarly, in that an END | |
278 | #* node cant be optimized. So END implies "unoptimizable" and PSEUDO | |
279 | #* mean "not seen anything to optimize yet". | |
f8abb37e | 280 | PSEUDO PSEUDO, off ; Pseudo opcode for internal use. |
1de06328 | 281 | |
03363afd YO |
282 | ------------------------------------------------------------------------------- |
283 | # Format for second section: | |
65aa4ca7 | 284 | # REGOP \t typelist [ \t typelist] |
03363afd YO |
285 | # typelist= namelist |
286 | # = namelist:FAIL | |
287 | # = name:count | |
288 | ||
289 | # Anything below is a state | |
290 | # | |
291 | # | |
f8abb37e | 292 | TRIE next:FAIL |
4ee16520 | 293 | EVAL B,postponed_AB:FAIL |
f8abb37e NC |
294 | CURLYX end:FAIL |
295 | WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL | |
296 | BRANCH next:FAIL | |
297 | CURLYM A,B:FAIL | |
298 | IFMATCH A:FAIL | |
21cbe009 | 299 | CURLY B_min,B_max:FAIL |
f8abb37e NC |
300 | COMMIT next:FAIL |
301 | MARKPOINT next:FAIL | |
302 | SKIP next:FAIL | |
303 | CUTGROUP next:FAIL | |
304 | KEEPS next:FAIL |