| 1 | # regcomp.sym |
| 2 | # |
| 3 | # File has two sections, divided by a line of dashes '-'. |
| 4 | # |
| 5 | # Empty rows after #-comment are removed from input are ignored |
| 6 | # |
| 7 | # First section is for regops, second section is for regmatch-states |
| 8 | # |
| 9 | # Note that the order in this file is important. |
| 10 | # |
| 11 | # Format for first section: |
| 12 | # NAME \s+ TYPE, arg-description [num-args] [flags] [longjump-len] ; DESCRIPTION |
| 13 | # flag <S> means is REGNODE_SIMPLE; flag <V> means is REGNODE_VARIES |
| 14 | # |
| 15 | # |
| 16 | # run perl regen.pl after editing this file |
| 17 | # Also update perlredebguts.pod |
| 18 | |
| 19 | |
| 20 | |
| 21 | #* Exit points |
| 22 | |
| 23 | END END, no ; End of program. |
| 24 | SUCCEED END, no ; Return from a subroutine, basically. |
| 25 | |
| 26 | #* Anchors: |
| 27 | |
| 28 | BOL BOL, no ; Match "" at beginning of line. |
| 29 | MBOL BOL, no ; Same, assuming multiline. |
| 30 | SBOL BOL, no ; Same, assuming singleline. |
| 31 | EOS EOL, no ; Match "" at end of string. |
| 32 | EOL EOL, no ; Match "" at end of line. |
| 33 | MEOL EOL, no ; Same, assuming multiline. |
| 34 | SEOL EOL, no ; Same, assuming singleline. |
| 35 | # The regops that have varieties that vary depending on the character set regex |
| 36 | # modifiers have to ordered thusly: /d, /l, /u, /a, /aa. This is because code |
| 37 | # in regcomp.c uses the enum value of the modifier as an offset from the /d |
| 38 | # version. The complements must come after the non-complements. |
| 39 | # BOUND, ALNUM, SPACE, DIGIT, and their complements are affected, as well as |
| 40 | # EXACTF. |
| 41 | BOUND BOUND, no ; Match "" at any word boundary using native charset semantics for non-utf8 |
| 42 | BOUNDL BOUND, no ; Match "" at any locale word boundary |
| 43 | BOUNDU BOUND, no ; Match "" at any word boundary using Unicode semantics |
| 44 | BOUNDA BOUND, no ; Match "" at any word boundary using ASCII semantics |
| 45 | # All NBOUND nodes are required by code in regexec.c to be greater than all BOUND ones |
| 46 | NBOUND NBOUND, no ; Match "" at any word non-boundary using native charset semantics for non-utf8 |
| 47 | NBOUNDL NBOUND, no ; Match "" at any locale word non-boundary |
| 48 | NBOUNDU NBOUND, no ; Match "" at any word non-boundary using Unicode semantics |
| 49 | NBOUNDA NBOUND, no ; Match "" at any word non-boundary using ASCII semantics |
| 50 | GPOS GPOS, no ; Matches where last m//g left off. |
| 51 | |
| 52 | #* [Special] alternatives: |
| 53 | |
| 54 | REG_ANY REG_ANY, no 0 S ; Match any one character (except newline). |
| 55 | SANY REG_ANY, no 0 S ; Match any one character. |
| 56 | CANY REG_ANY, no 0 S ; Match any one byte. |
| 57 | ANYOF ANYOF, sv 0 S ; Match character in (or not in) this class, single char match only |
| 58 | |
| 59 | # Order (within each group) of the below is important. See ordering comment |
| 60 | # above. The PLACEHOLDERn ones are wasting a value. Right now, we have plenty |
| 61 | # to spare, but these would be obvious candidates if ever we ran out of node |
| 62 | # types in a U8. |
| 63 | ALNUM ALNUM, no 0 S ; Match any alphanumeric character using native charset semantics for non-utf8 |
| 64 | ALNUML ALNUM, no 0 S ; Match any alphanumeric char in locale |
| 65 | ALNUMU ALNUM, no 0 S ; Match any alphanumeric char using Unicode semantics |
| 66 | ALNUMA ALNUM, no 0 S ; Match [A-Za-z_0-9] |
| 67 | NALNUM NALNUM, no 0 S ; Match any non-alphanumeric character using native charset semantics for non-utf8 |
| 68 | NALNUML NALNUM, no 0 S ; Match any non-alphanumeric char in locale |
| 69 | NALNUMU NALNUM, no 0 S ; Match any non-alphanumeric char using Unicode semantics |
| 70 | NALNUMA NALNUM, no 0 S ; Match [^A-Za-z_0-9] |
| 71 | SPACE SPACE, no 0 S ; Match any whitespace character using native charset semantics for non-utf8 |
| 72 | SPACEL SPACE, no 0 S ; Match any whitespace char in locale |
| 73 | SPACEU SPACE, no 0 S ; Match any whitespace char using Unicode semantics |
| 74 | SPACEA SPACE, no 0 S ; Match [ \t\n\f\r] |
| 75 | NSPACE NSPACE, no 0 S ; Match any non-whitespace character using native charset semantics for non-utf8 |
| 76 | NSPACEL NSPACE, no 0 S ; Match any non-whitespace char in locale |
| 77 | NSPACEU NSPACE, no 0 S ; Match any non-whitespace char using Unicode semantics |
| 78 | NSPACEA NSPACE, no 0 S ; Match [^ \t\n\f\r] |
| 79 | DIGIT DIGIT, no 0 S ; Match any numeric character using native charset semantics for non-utf8 |
| 80 | DIGITL DIGIT, no 0 S ; Match any numeric character in locale |
| 81 | PLACEHOLDER1 NOTHING, no ; placeholder for missing DIGITU |
| 82 | DIGITA DIGIT, no 0 S ; Match [0-9] |
| 83 | NDIGIT NDIGIT, no 0 S ; Match any non-numeric character using native charset semantics for non-utf8 |
| 84 | NDIGITL NDIGIT, no 0 S ; Match any non-numeric character in locale |
| 85 | PLACEHOLDER2 NOTHING, no ; placeholder for missing NDIGITU |
| 86 | NDIGITA NDIGIT, no 0 S ; Match [^0-9] |
| 87 | |
| 88 | POSIXD POSIXD, none 0 S ; currently unused except as a placeholder |
| 89 | POSIXL POSIXD, none 0 S ; currently unused except as a placeholder |
| 90 | POSIXU POSIXD, none 0 S ; currently unused except as a placeholder |
| 91 | POSIXA POSIXD, none 0 S ; Some [[:class:]] under /a; the FLAGS field gives which one |
| 92 | NPOSIXD POSIXD, none 0 S ; currently unused except as a placeholder |
| 93 | NPOSIXL POSIXD, none 0 S ; currently unused except as a placeholder |
| 94 | NPOSIXU POSIXD, none 0 S ; currently unused except as a placeholder |
| 95 | NPOSIXA POSIXD, none 0 S ; complement of POSIXA, [[:^class:]] |
| 96 | # End of order is important (within groups) |
| 97 | |
| 98 | CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence |
| 99 | |
| 100 | #* Alternation |
| 101 | |
| 102 | # BRANCH The set of branches constituting a single choice are hooked |
| 103 | # together with their "next" pointers, since precedence prevents |
| 104 | # anything being concatenated to any individual branch. The |
| 105 | # "next" pointer of the last BRANCH in a choice points to the |
| 106 | # thing following the whole choice. This is also where the |
| 107 | # final "next" pointer of each individual branch points; each |
| 108 | # branch starts with the operand node of a BRANCH node. |
| 109 | # |
| 110 | BRANCH BRANCH, node 0 V ; Match this alternative, or the next... |
| 111 | |
| 112 | #*Back pointer |
| 113 | |
| 114 | # BACK Normal "next" pointers all implicitly point forward; BACK |
| 115 | # exists to make loop structures possible. |
| 116 | # not used |
| 117 | BACK BACK, no 0 V ; Match "", "next" ptr points backward. |
| 118 | |
| 119 | #*Literals - NOTE the relative ordering of these types is important do not change it |
| 120 | |
| 121 | EXACT EXACT, str ; Match this string (preceded by length). |
| 122 | EXACTF EXACT, str ; Match this non-UTF-8 string (not guaranteed to be folded) using /id rules (w/len). |
| 123 | EXACTFL EXACT, str ; Match this string (not guaranteed to be folded) using /il rules (w/len). |
| 124 | EXACTFU EXACT, str ; Match this string (folded iff in UTF-8, length in folding doesn't change if not in UTF-8) using /iu rules (w/len). |
| 125 | EXACTFA EXACT, str ; Match this string (not guaranteed to be folded) using /iaa rules (w/len). |
| 126 | EXACTFU_SS EXACT, str ; Match this string (folded iff in UTF-8, length in folding may change even if not in UTF-8) using /iu rules (w/len). |
| 127 | EXACTFU_TRICKYFOLD EXACT, str ; Match this folded UTF-8 string using /iu rules |
| 128 | |
| 129 | #*Do nothing types |
| 130 | |
| 131 | NOTHING NOTHING, no ; Match empty string. |
| 132 | # A variant of above which delimits a group, thus stops optimizations |
| 133 | TAIL NOTHING, no ; Match empty string. Can jump here from outside. |
| 134 | |
| 135 | #*Loops |
| 136 | |
| 137 | # STAR,PLUS '?', and complex '*' and '+', are implemented as circular |
| 138 | # BRANCH structures using BACK. Simple cases (one character |
| 139 | # per match) are implemented with STAR and PLUS for speed |
| 140 | # and to minimize recursive plunges. |
| 141 | # |
| 142 | STAR STAR, node 0 V ; Match this (simple) thing 0 or more times. |
| 143 | PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times. |
| 144 | |
| 145 | CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times. |
| 146 | CURLYN CURLY, no 2 V ; Capture next-after-this simple thing |
| 147 | CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times. |
| 148 | CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times. |
| 149 | |
| 150 | # This terminator creates a loop structure for CURLYX |
| 151 | WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches. |
| 152 | |
| 153 | #*Buffer related |
| 154 | |
| 155 | # OPEN,CLOSE,GROUPP ...are numbered at compile time. |
| 156 | OPEN OPEN, num 1 ; Mark this point in input as start of #n. |
| 157 | CLOSE CLOSE, num 1 ; Analogous to OPEN. |
| 158 | |
| 159 | REF REF, num 1 V ; Match some already matched string |
| 160 | REFF REF, num 1 V ; Match already matched string, folded using native charset semantics for non-utf8 |
| 161 | REFFL REF, num 1 V ; Match already matched string, folded in loc. |
| 162 | # N?REFF[AU] could have been implemented using the FLAGS field of the |
| 163 | # regnode, but by having a separate node type, we can use the existing switch |
| 164 | # statement to avoid some tests |
| 165 | REFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8 |
| 166 | REFFA REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII |
| 167 | |
| 168 | #*Named references. Code in regcomp.c assumes that these all are after the numbered references |
| 169 | NREF REF, no-sv 1 V ; Match some already matched string |
| 170 | NREFF REF, no-sv 1 V ; Match already matched string, folded using native charset semantics for non-utf8 |
| 171 | NREFFL REF, no-sv 1 V ; Match already matched string, folded in loc. |
| 172 | NREFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8 |
| 173 | NREFFA REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII |
| 174 | |
| 175 | IFMATCH BRANCHJ, off 1 . 2 ; Succeeds if the following matches. |
| 176 | UNLESSM BRANCHJ, off 1 . 2 ; Fails if the following matches. |
| 177 | SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE. |
| 178 | IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher . |
| 179 | GROUPP GROUPP, num 1 ; Whether the group matched. |
| 180 | |
| 181 | #*Support for long RE |
| 182 | |
| 183 | LONGJMP LONGJMP, off 1 . 1 ; Jump far away. |
| 184 | BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset. |
| 185 | |
| 186 | #*The heavy worker |
| 187 | |
| 188 | EVAL EVAL, evl 1 ; Execute some Perl code. |
| 189 | |
| 190 | #*Modifiers |
| 191 | |
| 192 | MINMOD MINMOD, no ; Next operator is not greedy. |
| 193 | LOGICAL LOGICAL, no ; Next opcode should set the flag only. |
| 194 | |
| 195 | # This is not used yet |
| 196 | RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens. |
| 197 | |
| 198 | #*Trie Related |
| 199 | |
| 200 | # Behave the same as A|LIST|OF|WORDS would. The '..C' variants have |
| 201 | # inline charclass data (ascii only), the 'C' store it in the structure. |
| 202 | # NOTE: the relative order of the TRIE-like regops is significant |
| 203 | |
| 204 | TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type |
| 205 | TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data |
| 206 | |
| 207 | # For start classes, contains an added fail table. |
| 208 | AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type |
| 209 | AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data |
| 210 | |
| 211 | #*Regex Subroutines |
| 212 | GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2 |
| 213 | GOSTART GOSTART, no ; recurse to start of pattern |
| 214 | |
| 215 | #*Special conditionals |
| 216 | NGROUPP NGROUPP, no-sv 1 ; Whether the group matched. |
| 217 | INSUBP INSUBP, num 1 ; Whether we are in a specific recurse. |
| 218 | DEFINEP DEFINEP, none 1 ; Never execute directly. |
| 219 | |
| 220 | #*Backtracking Verbs |
| 221 | ENDLIKE ENDLIKE, none ; Used only for the type field of verbs |
| 222 | OPFAIL ENDLIKE, none ; Same as (?!) |
| 223 | ACCEPT ENDLIKE, parno 1 ; Accepts the current matched string. |
| 224 | |
| 225 | |
| 226 | #*Verbs With Arguments |
| 227 | VERB VERB, no-sv 1 ; Used only for the type field of verbs |
| 228 | PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this |
| 229 | MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut. |
| 230 | SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying |
| 231 | COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this |
| 232 | CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group |
| 233 | |
| 234 | #*Control what to keep in $&. |
| 235 | KEEPS KEEPS, no ; $& begins here. |
| 236 | |
| 237 | #*New charclass like patterns |
| 238 | LNBREAK LNBREAK, none ; generic newline pattern |
| 239 | |
| 240 | # regcomp.c expects the node number of the complement to be one greater than |
| 241 | # the non-complement |
| 242 | VERTWS VERTWS, none 0 S ; vertical whitespace (Perl 6) |
| 243 | NVERTWS NVERTWS, none 0 S ; not vertical whitespace (Perl 6) |
| 244 | HORIZWS HORIZWS, none 0 S ; horizontal whitespace (Perl 6) |
| 245 | NHORIZWS NHORIZWS, none 0 S ; not horizontal whitespace (Perl 6) |
| 246 | |
| 247 | # NEW STUFF SOMEWHERE ABOVE THIS LINE |
| 248 | |
| 249 | ################################################################################ |
| 250 | |
| 251 | #*SPECIAL REGOPS |
| 252 | |
| 253 | # This is not really a node, but an optimized away piece of a "long" node. |
| 254 | # To simplify debugging output, we mark it as if it were a node |
| 255 | OPTIMIZED NOTHING, off ; Placeholder for dump. |
| 256 | |
| 257 | # Special opcode with the property that no opcode in a compiled program |
| 258 | # will ever be of this type. Thus it can be used as a flag value that |
| 259 | # no other opcode has been seen. END is used similarly, in that an END |
| 260 | # node cant be optimized. So END implies "unoptimizable" and PSEUDO mean |
| 261 | # "not seen anything to optimize yet". |
| 262 | PSEUDO PSEUDO, off ; Pseudo opcode for internal use. |
| 263 | |
| 264 | ------------------------------------------------------------------------------- |
| 265 | # Format for second section: |
| 266 | # REGOP \t typelist [ \t typelist] [# Comment] |
| 267 | # typelist= namelist |
| 268 | # = namelist:FAIL |
| 269 | # = name:count |
| 270 | |
| 271 | # Anything below is a state |
| 272 | # |
| 273 | # |
| 274 | TRIE next:FAIL |
| 275 | EVAL AB:FAIL |
| 276 | CURLYX end:FAIL |
| 277 | WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL |
| 278 | BRANCH next:FAIL |
| 279 | CURLYM A,B:FAIL |
| 280 | IFMATCH A:FAIL |
| 281 | CURLY B_min_known,B_min,B_max:FAIL |
| 282 | COMMIT next:FAIL |
| 283 | MARKPOINT next:FAIL |
| 284 | SKIP next:FAIL |
| 285 | CUTGROUP next:FAIL |
| 286 | KEEPS next:FAIL |