3 # File has two sections, divided by a line of dashes '-'.
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
9 # First section is for regops, second section is for regmatch-states
11 # Note that the order in this file is important.
13 # Format for first section:
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
20 # flag <S> means is REGNODE_SIMPLE; flag <V> means is REGNODE_VARIES; <.> is
22 # longjump is 1 if the (first) argument holds the next offset (instead of the
23 # usual 'next_offset' field
25 # run perl regen.pl after editing this file
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.
41 END END, no ; End of program.
42 SUCCEED END, no ; Return from a subroutine, basically.
44 #* Line Start Anchors:
45 #Note flags field for SBOL indicates if it is a /^/ or a /\A/
46 SBOL BOL, no ; Match "" at beginning of line: /^/, /\A/
47 MBOL BOL, no ; Same, assuming multiline: /^/m
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/
54 #* Match Start Anchors:
55 GPOS GPOS, no ; Matches where last m//g left off.
57 #* Word Boundary Opcodes:
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.
62 # BOUND, POSIX and their complements are affected, as well as EXACTF.
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
65 BOUNDU BOUND, no ; Match "" at any boundary of a given type using /u rules.
66 BOUNDA BOUND, no ; Match "" at any boundary between \w\W or \W\w, where \w is [_a-zA-Z0-9]
67 # All NBOUND nodes are required by code in regexec.c to be greater than all BOUND ones
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
70 NBOUNDU NBOUND, no ; Match "" at any non-boundary of a given type using using /u rules.
71 NBOUNDA NBOUND, no ; Match "" betweeen any \w\w or \W\W, where \w is [_a-zA-Z0-9]
73 #* [Special] alternatives:
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.
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
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
83 ANYOFHb ANYOF, sv 1 S ; Like ANYOFH, but all matches share the same UTF-8 start byte, given in the flags field
84 ANYOFHr ANYOF, sv 1 S ; Like ANYOFH, but the flags field contains packed bounds for all matchable UTF-8 start bytes.
86 ANYOFM ANYOFM byte 1 S ; Like ANYOF, but matches an invariant byte as determined by the mask and arg
87 NANYOFM ANYOFM byte 1 S ; complement of ANYOFM
89 #* POSIX Character Classes:
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
94 POSIXA POSIXD, none 0 S ; Some [[:class:]] under /a; the FLAGS field gives which one
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:]]
98 NPOSIXA NPOSIXD, none 0 S ; complement of POSIXA, [[:^class:]]
99 # End of order is important
101 CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence
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.
114 BRANCH BRANCH, node 0 V ; Match this alternative, or the next...
117 # NOTE: the relative ordering of these types is important do not change it
119 EXACT EXACT, str ; Match this string (flags field is the length).
120 EXACTL EXACT, str ; Like EXACT, but /l is in effect (used so locale-related warnings can be checked for).
121 EXACTF EXACT, str ; Like EXACT, but match using /id rules; (string not UTF-8, not guaranteed to be folded).
122 EXACTFL EXACT, str ; Like EXACT, but match using /il rules; (string not likely to be folded).
123 EXACTFU EXACT, str ; Like EXACT, but match using /iu rules; (string folded).
124 EXACTFAA EXACT, str ; Like EXACT, but match using /iaa rules; (string folded iff pattern is UTF8; folded length <= unfolded).
126 # End of important relative ordering.
128 EXACTFUP EXACT, str ; Like EXACT, but match using /iu rules; (string not UTF-8, not guaranteed to be folded; and it is Problematic).
129 # In order for a non-UTF-8 EXACTFAA to think the pattern is pre-folded when
130 # matching a UTF-8 target string, there would have to be something like an
131 # EXACTFAA_MICRO which would not be considered pre-folded for UTF-8 targets,
132 # since the fold of the MICRO SIGN would not be done, and would be
133 # representable in the UTF-8 target string.
135 EXACTFLU8 EXACT, str ; Like EXACTFU, but use /il, UTF-8, (string is folded, and everything in it is above 255.
136 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).
139 EXACT_ONLY8 EXACT, str ; Like EXACT, but only UTF-8 encoded targets can match
140 EXACTFU_ONLY8 EXACT, str ; Like EXACTFU, but only UTF-8 encoded targets can match
141 # One could add EXACTFAA8 and something that has the same effect for /l,
142 # but these would be extremely uncommon
144 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).
148 NOTHING NOTHING, no ; Match empty string.
149 #*A variant of above which delimits a group, thus stops optimizations
150 TAIL NOTHING, no ; Match empty string. Can jump here from outside.
154 #* STAR,PLUS '?', and complex '*' and '+', are implemented as
155 #* circular BRANCH structures. Simple cases
156 #* (one character per match) are implemented with STAR
157 #* and PLUS for speed and to minimize recursive plunges.
159 STAR STAR, node 0 V ; Match this (simple) thing 0 or more times.
160 PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times.
162 CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times.
163 CURLYN CURLY, no 2 V ; Capture next-after-this simple thing
164 CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times.
165 CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times.
167 #*This terminator creates a loop structure for CURLYX
168 WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches.
172 #*OPEN,CLOSE,GROUPP ...are numbered at compile time.
173 OPEN OPEN, num 1 ; Mark this point in input as start of #n.
174 CLOSE CLOSE, num 1 ; Close corresponding OPEN of #n.
175 SROPEN SROPEN, none ; Same as OPEN, but for script run
176 SRCLOSE SRCLOSE, none ; Close preceding SROPEN
178 REF REF, num 1 V ; Match some already matched string
179 REFF REF, num 1 V ; Match already matched string, using /di rules.
180 REFFL REF, num 1 V ; Match already matched string, using /li rules.
181 # N?REFF[AU] could have been implemented using the FLAGS field of the
182 # regnode, but by having a separate node type, we can use the existing switch
183 # statement to avoid some tests
184 REFFU REF, num 1 V ; Match already matched string, usng /ui.
185 REFFA REF, num 1 V ; Match already matched string, using /aai rules.
187 #*Named references. Code in regcomp.c assumes that these all are after
188 #*the numbered references
189 REFN REF, no-sv 1 V ; Match some already matched string
190 REFFN REF, no-sv 1 V ; Match already matched string, using /di rules.
191 REFFLN REF, no-sv 1 V ; Match already matched string, using /li rules.
192 REFFUN REF, num 1 V ; Match already matched string, using /ui rules.
193 REFFAN REF, num 1 V ; Match already matched string, using /aai rules.
195 #*Support for long RE
196 LONGJMP LONGJMP, off 1 . 1 ; Jump far away.
197 BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset.
199 #*Special Case Regops
200 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
201 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
202 SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE.
203 IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher.
204 GROUPP GROUPP, num 1 ; Whether the group matched.
209 EVAL EVAL, evl/flags 2L ; Execute some Perl code.
213 MINMOD MINMOD, no ; Next operator is not greedy.
214 LOGICAL LOGICAL, no ; Next opcode should set the flag only.
216 #*This is not used yet
217 RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens.
221 #* Behave the same as A|LIST|OF|WORDS would. The '..C' variants
222 #* have inline charclass data (ascii only), the 'C' store it in the
224 # NOTE: the relative order of the TRIE-like regops is significant
226 TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type
227 TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data
229 # For start classes, contains an added fail table.
230 AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type
231 AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data
234 GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2
236 #*Special conditionals
237 GROUPPN GROUPPN, no-sv 1 ; Whether the group matched.
238 INSUBP INSUBP, num 1 ; Whether we are in a specific recurse.
239 DEFINEP DEFINEP, none 1 ; Never execute directly.
242 ENDLIKE ENDLIKE, none ; Used only for the type field of verbs
243 OPFAIL ENDLIKE, no-sv 1 ; Same as (?!), but with verb arg
244 ACCEPT ENDLIKE, no-sv/num 2L ; Accepts the current matched string, with verbar
246 #*Verbs With Arguments
247 VERB VERB, no-sv 1 ; Used only for the type field of verbs
248 PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this
249 MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut.
250 SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying
251 COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this
252 CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group
254 #*Control what to keep in $&.
255 KEEPS KEEPS, no ; $& begins here.
257 #*New charclass like patterns
258 LNBREAK LNBREAK, none ; generic newline pattern
260 # NEW STUFF SOMEWHERE ABOVE THIS LINE
262 ################################################################################
266 #* This is not really a node, but an optimized away piece of a "long"
267 #* node. To simplify debugging output, we mark it as if it were a node
268 OPTIMIZED NOTHING, off ; Placeholder for dump.
270 #* Special opcode with the property that no opcode in a compiled program
271 #* will ever be of this type. Thus it can be used as a flag value that
272 #* no other opcode has been seen. END is used similarly, in that an END
273 #* node cant be optimized. So END implies "unoptimizable" and PSEUDO
274 #* mean "not seen anything to optimize yet".
275 PSEUDO PSEUDO, off ; Pseudo opcode for internal use.
277 -------------------------------------------------------------------------------
278 # Format for second section:
279 # REGOP \t typelist [ \t typelist]
284 # Anything below is a state
288 EVAL B,postponed_AB:FAIL
290 WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL
294 CURLY B_min,B_max:FAIL