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 [num-args] [flags] [longjump] ; DESCRIPTION
15 # flag <S> means is REGNODE_SIMPLE; flag <V> means is REGNODE_VARIES; <.> is
17 # longjump is 1 if the (first) argument holds the next offset.
20 # run perl regen.pl after editing this file
26 END END, no ; End of program.
27 SUCCEED END, no ; Return from a subroutine, basically.
29 #* Line Start Anchors:
30 #Note flags field for SBOL indicates if it is a /^/ or a /\A/
31 SBOL BOL, no ; Match "" at beginning of line: /^/, /\A/
32 MBOL BOL, no ; Same, assuming multiline: /^/m
35 SEOL EOL, no ; Match "" at end of line: /$/
36 MEOL EOL, no ; Same, assuming multiline: /$/m
37 EOS EOL, no ; Match "" at end of string: /\z/
39 #* Match Start Anchors:
40 GPOS GPOS, no ; Matches where last m//g left off.
42 #* Word Boundary Opcodes:
43 # The regops that have varieties that vary depending on the character set regex
44 # modifiers have to ordered thusly: /d, /l, /u, /a, /aa. This is because code
45 # in regcomp.c uses the enum value of the modifier as an offset from the /d
46 # version. The complements must come after the non-complements.
47 # BOUND, POSIX and their complements are affected, as well as EXACTF.
48 BOUND BOUND, no ; Like BOUNDA for non-utf8, otherwise match "" between any Unicode \w\W or \W\w
49 BOUNDL BOUND, no ; Like BOUND/BOUNDU, but \w and \W are defined by current locale
50 BOUNDU BOUND, no ; Match "" at any boundary of a given type using /u rules.
51 BOUNDA BOUND, no ; Match "" at any boundary between \w\W or \W\w, where \w is [_a-zA-Z0-9]
52 # All NBOUND nodes are required by code in regexec.c to be greater than all BOUND ones
53 NBOUND NBOUND, no ; Like NBOUNDA for non-utf8, otherwise match "" between any Unicode \w\w or \W\W
54 NBOUNDL NBOUND, no ; Like NBOUND/NBOUNDU, but \w and \W are defined by current locale
55 NBOUNDU NBOUND, no ; Match "" at any non-boundary of a given type using using /u rules.
56 NBOUNDA NBOUND, no ; Match "" betweeen any \w\w or \W\W, where \w is [_a-zA-Z0-9]
58 #* [Special] alternatives:
59 REG_ANY REG_ANY, no 0 S ; Match any one character (except newline).
60 SANY REG_ANY, no 0 S ; Match any one character.
61 ANYOF ANYOF, sv charclass S ; Match character in (or not in) this class, single char match only
62 ANYOFD ANYOF, sv charclass S ; Like ANYOF, but /d is in effect
63 ANYOFL ANYOF, sv charclass S ; Like ANYOF, but /l is in effect
64 ANYOFPOSIXL ANYOF, sv charclass_posixl S ; Like ANYOFL, but matches [[:posix:]] classes
67 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
68 ANYOFHb ANYOF, sv 1 S ; Like ANYOFH, but all matches share the same UTF-8 start byte, given in the flags field
69 ANYOFHr ANYOF, sv 1 S ; Like ANYOFH, but the flags field contains packed bounds for all matchable UTF-8 start bytes.
71 ANYOFM ANYOFM byte 1 S ; Like ANYOF, but matches an invariant byte as determined by the mask and arg
72 NANYOFM ANYOFM byte 1 S ; complement of ANYOFM
74 #* POSIX Character Classes:
75 # Order of the below is important. See ordering comment above.
76 POSIXD POSIXD, none 0 S ; Some [[:class:]] under /d; the FLAGS field gives which one
77 POSIXL POSIXD, none 0 S ; Some [[:class:]] under /l; the FLAGS field gives which one
78 POSIXU POSIXD, none 0 S ; Some [[:class:]] under /u; the FLAGS field gives which one
79 POSIXA POSIXD, none 0 S ; Some [[:class:]] under /a; the FLAGS field gives which one
80 NPOSIXD NPOSIXD, none 0 S ; complement of POSIXD, [[:^class:]]
81 NPOSIXL NPOSIXD, none 0 S ; complement of POSIXL, [[:^class:]]
82 NPOSIXU NPOSIXD, none 0 S ; complement of POSIXU, [[:^class:]]
83 NPOSIXA NPOSIXD, none 0 S ; complement of POSIXA, [[:^class:]]
84 # End of order is important
86 CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence
90 #* BRANCH The set of branches constituting a single choice are
91 #* hooked together with their "next" pointers, since
92 #* precedence prevents anything being concatenated to
93 #* any individual branch. The "next" pointer of the last
94 #* BRANCH in a choice points to the thing following the
95 #* whole choice. This is also where the final "next"
96 #* pointer of each individual branch points; each branch
97 #* starts with the operand node of a BRANCH node.
99 BRANCH BRANCH, node 0 V ; Match this alternative, or the next...
102 # NOTE: the relative ordering of these types is important do not change it
104 EXACT EXACT, str ; Match this string (preceded by length).
105 EXACTL EXACT, str ; Like EXACT, but /l is in effect (used so locale-related warnings can be checked for).
106 EXACTF EXACT, str ; Match this string using /id rules (w/len); (string not UTF-8, not guaranteed to be folded).
107 EXACTFL EXACT, str ; Match this string using /il rules (w/len); (string not guaranteed to be folded).
108 EXACTFU EXACT, str ; Match this string using /iu rules (w/len); (string folded iff in UTF-8; non-UTF8 folded length <= unfolded).
109 EXACTFAA EXACT, str ; Match this string using /iaa rules (w/len) (string folded iff in UTF-8; non-UTF8 folded length <= unfolded).
111 # End of important relative ordering.
113 EXACTFUP EXACT, str ; Match this string using /iu rules (w/len); (string not UTF-8, not guaranteed to be folded; and its Problematic).
114 # In order for a non-UTF-8 EXACTFAA to think the pattern is pre-folded when
115 # matching a UTF-8 target string, there would have to be something like an
116 # EXACTFAA_MICRO which would not be considered pre-folded for UTF-8 targets,
117 # since the fold of the MICRO SIGN would not be done, and would be
118 # representable in the UTF-8 target string.
120 EXACTFLU8 EXACT, str ; Like EXACTFU, but use /il, UTF-8, folded, and everything in it is above 255.
121 EXACTFAA_NO_TRIE EXACT, str ; Match this string using /iaa rules (w/len) (string not UTF-8, not guaranteed to be folded, not currently trie-able).
124 EXACT_ONLY8 EXACT, str ; Like EXACT, but only UTF-8 encoded targets can match
125 EXACTFU_ONLY8 EXACT, str ; Like EXACTFU, but only UTF-8 encoded targets can match
126 # One could add EXACTFAA8 and something that has the same effect for /l,
127 # but these would be extremely uncommon
129 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).
133 NOTHING NOTHING, no ; Match empty string.
134 #*A variant of above which delimits a group, thus stops optimizations
135 TAIL NOTHING, no ; Match empty string. Can jump here from outside.
139 #* STAR,PLUS '?', and complex '*' and '+', are implemented as
140 #* circular BRANCH structures. Simple cases
141 #* (one character per match) are implemented with STAR
142 #* and PLUS for speed and to minimize recursive plunges.
144 STAR STAR, node 0 V ; Match this (simple) thing 0 or more times.
145 PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times.
147 CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times.
148 CURLYN CURLY, no 2 V ; Capture next-after-this simple thing
149 CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times.
150 CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times.
152 #*This terminator creates a loop structure for CURLYX
153 WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches.
157 #*OPEN,CLOSE,GROUPP ...are numbered at compile time.
158 OPEN OPEN, num 1 ; Mark this point in input as start of #n.
159 CLOSE CLOSE, num 1 ; Close corresponding OPEN of #n.
160 SROPEN SROPEN, none ; Same as OPEN, but for script run
161 SRCLOSE SRCLOSE, none ; Close preceding SROPEN
163 REF REF, num 1 V ; Match some already matched string
164 REFF REF, num 1 V ; Match already matched string, using /di rules.
165 REFFL REF, num 1 V ; Match already matched string, using /li rules.
166 # N?REFF[AU] could have been implemented using the FLAGS field of the
167 # regnode, but by having a separate node type, we can use the existing switch
168 # statement to avoid some tests
169 REFFU REF, num 1 V ; Match already matched string, usng /ui.
170 REFFA REF, num 1 V ; Match already matched string, using /aai rules.
172 #*Named references. Code in regcomp.c assumes that these all are after
173 #*the numbered references
174 REFN REF, no-sv 1 V ; Match some already matched string
175 REFFN REF, no-sv 1 V ; Match already matched string, using /di rules.
176 REFFLN REF, no-sv 1 V ; Match already matched string, using /li rules.
177 REFFUN REF, num 1 V ; Match already matched string, using /ui rules.
178 REFFAN REF, num 1 V ; Match already matched string, using /aai rules.
180 #*Support for long RE
181 LONGJMP LONGJMP, off 1 . 1 ; Jump far away.
182 BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset.
184 #*Special Case Regops
185 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
186 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
187 SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE.
188 IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher.
189 GROUPP GROUPP, num 1 ; Whether the group matched.
194 EVAL EVAL, evl/flags 2L ; Execute some Perl code.
198 MINMOD MINMOD, no ; Next operator is not greedy.
199 LOGICAL LOGICAL, no ; Next opcode should set the flag only.
201 #*This is not used yet
202 RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens.
206 #* Behave the same as A|LIST|OF|WORDS would. The '..C' variants
207 #* have inline charclass data (ascii only), the 'C' store it in the
209 # NOTE: the relative order of the TRIE-like regops is significant
211 TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type
212 TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data
214 # For start classes, contains an added fail table.
215 AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type
216 AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data
219 GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2
221 #*Special conditionals
222 GROUPPN GROUPPN, no-sv 1 ; Whether the group matched.
223 INSUBP INSUBP, num 1 ; Whether we are in a specific recurse.
224 DEFINEP DEFINEP, none 1 ; Never execute directly.
227 ENDLIKE ENDLIKE, none ; Used only for the type field of verbs
228 OPFAIL ENDLIKE, no-sv 1 ; Same as (?!), but with verb arg
229 ACCEPT ENDLIKE, no-sv/num 2L ; Accepts the current matched string, with verbar
231 #*Verbs With Arguments
232 VERB VERB, no-sv 1 ; Used only for the type field of verbs
233 PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this
234 MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut.
235 SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying
236 COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this
237 CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group
239 #*Control what to keep in $&.
240 KEEPS KEEPS, no ; $& begins here.
242 #*New charclass like patterns
243 LNBREAK LNBREAK, none ; generic newline pattern
245 # NEW STUFF SOMEWHERE ABOVE THIS LINE
247 ################################################################################
251 #* This is not really a node, but an optimized away piece of a "long"
252 #* node. To simplify debugging output, we mark it as if it were a node
253 OPTIMIZED NOTHING, off ; Placeholder for dump.
255 #* Special opcode with the property that no opcode in a compiled program
256 #* will ever be of this type. Thus it can be used as a flag value that
257 #* no other opcode has been seen. END is used similarly, in that an END
258 #* node cant be optimized. So END implies "unoptimizable" and PSEUDO
259 #* mean "not seen anything to optimize yet".
260 PSEUDO PSEUDO, off ; Pseudo opcode for internal use.
262 -------------------------------------------------------------------------------
263 # Format for second section:
264 # REGOP \t typelist [ \t typelist]
269 # Anything below is a state
273 EVAL B,postponed_AB:FAIL
275 WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL
279 CURLY B_min,B_max:FAIL