3 # File has two sections, divided by a line of dashes '-'.
5 # Empty rows after #-comment are removed from input are ignored
7 # First section is for regops, second section is for regmatch-states
9 # Note that the order in this file is important.
11 # Format for first section:
12 # NAME \s+ TYPE, arg-description [flags] [num-args] [longjump-len] ; DESCRIPTION
15 # run perl regen.pl after editing this file
16 # Also update perlredebguts.pod
22 END END, no ; End of program.
23 SUCCEED END, no ; Return from a subroutine, basically.
27 BOL BOL, no ; Match "" at beginning of line.
28 MBOL BOL, no ; Same, assuming multiline.
29 SBOL BOL, no ; Same, assuming singleline.
30 EOS EOL, no ; Match "" at end of string.
31 EOL EOL, no ; Match "" at end of line.
32 MEOL EOL, no ; Same, assuming multiline.
33 SEOL EOL, no ; Same, assuming singleline.
34 BOUND BOUND, no ; Match "" at any word boundary using native charset semantics for non-utf8
35 BOUNDL BOUND, no ; Match "" at any locale word boundary
36 BOUNDU BOUND, no ; Match "" at any word boundary using Unicode semantics
37 BOUNDA BOUND, no ; Match "" at any word boundary using ASCII semantics
38 # All NBOUND nodes are required by a line regexec.c to be greater than all BOUND ones
39 NBOUND NBOUND, no ; Match "" at any word non-boundary using native charset semantics for non-utf8
40 NBOUNDL NBOUND, no ; Match "" at any locale word non-boundary
41 NBOUNDU NBOUND, no ; Match "" at any word non-boundary using Unicode semantics
42 NBOUNDA NBOUND, no ; Match "" at any word non-boundary using ASCII semantics
43 GPOS GPOS, no ; Matches where last m//g left off.
45 #* [Special] alternatives:
47 REG_ANY REG_ANY, no 0 S ; Match any one character (except newline).
48 SANY REG_ANY, no 0 S ; Match any one character.
49 CANY REG_ANY, no 0 S ; Match any one byte.
50 ANYOF ANYOF, sv 0 S ; Match character in (or not in) this class, single char match only
51 ANYOFV ANYOF, sv 0 V ; Match character in (or not in) this class, can match-multiple chars
52 ALNUM ALNUM, no 0 S ; Match any alphanumeric character using native charset semantics for non-utf8
53 ALNUML ALNUM, no 0 S ; Match any alphanumeric char in locale
54 ALNUMU ALNUM, no 0 S ; Match any alphanumeric char using Unicode semantics
55 ALNUMA ALNUM, no 0 S ; Match [A-Za-z_0-9]
56 NALNUM NALNUM, no 0 S ; Match any non-alphanumeric character using native charset semantics for non-utf8
57 NALNUML NALNUM, no 0 S ; Match any non-alphanumeric char in locale
58 NALNUMU NALNUM, no 0 S ; Match any non-alphanumeric char using Unicode semantics
59 NALNUMA NALNUM, no 0 S ; Match [^A-Za-z_0-9]
60 SPACE SPACE, no 0 S ; Match any whitespace character using native charset semantics for non-utf8
61 SPACEL SPACE, no 0 S ; Match any whitespace char in locale
62 SPACEU SPACE, no 0 S ; Match any whitespace char using Unicode semantics
63 SPACEA SPACE, no 0 S ; Match [ \t\n\f\r]
64 NSPACE NSPACE, no 0 S ; Match any non-whitespace character using native charset semantics for non-utf8
65 NSPACEL NSPACE, no 0 S ; Match any non-whitespace char in locale
66 NSPACEU NSPACE, no 0 S ; Match any non-whitespace char using Unicode semantics
67 NSPACEA NSPACE, no 0 S ; Match [^ \t\n\f\r]
68 DIGIT DIGIT, no 0 S ; Match any numeric character using native charset semantics for non-utf8
69 DIGITL DIGIT, no 0 S ; Match any numeric character in locale
70 DIGITA DIGIT, no 0 S ; Match [0-9]
71 NDIGIT NDIGIT, no 0 S ; Match any non-numeric character using native charset semantics for non-utf8
72 NDIGITL NDIGIT, no 0 S ; Match any non-numeric character in locale
73 NDIGITA NDIGIT, no 0 S ; Match [^0-9]
74 CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence
78 # BRANCH The set of branches constituting a single choice are hooked
79 # together with their "next" pointers, since precedence prevents
80 # anything being concatenated to any individual branch. The
81 # "next" pointer of the last BRANCH in a choice points to the
82 # thing following the whole choice. This is also where the
83 # final "next" pointer of each individual branch points; each
84 # branch starts with the operand node of a BRANCH node.
86 BRANCH BRANCH, node 0 V ; Match this alternative, or the next...
90 # BACK Normal "next" pointers all implicitly point forward; BACK
91 # exists to make loop structures possible.
93 BACK BACK, no 0 V ; Match "", "next" ptr points backward.
97 EXACT EXACT, str ; Match this string (preceded by length).
98 EXACTF EXACT, str ; Match this string, folded, native charset semantics for non-utf8 (prec. by length).
99 EXACTFL EXACT, str ; Match this string, folded in locale (w/len).
100 EXACTFU EXACT, str ; Match this string, folded, Unicode semantics for non-utf8 (prec. by length).
101 EXACTFA EXACT, str ; Match this string, folded, Unicode semantics for non-utf8, but no ASCII-range character matches outside ASCII (prec. by length),.
105 NOTHING NOTHING, no ; Match empty string.
106 # A variant of above which delimits a group, thus stops optimizations
107 TAIL NOTHING, no ; Match empty string. Can jump here from outside.
111 # STAR,PLUS '?', and complex '*' and '+', are implemented as circular
112 # BRANCH structures using BACK. Simple cases (one character
113 # per match) are implemented with STAR and PLUS for speed
114 # and to minimize recursive plunges.
116 STAR STAR, node 0 V ; Match this (simple) thing 0 or more times.
117 PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times.
119 CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times.
120 CURLYN CURLY, no 2 V ; Capture next-after-this simple thing
121 CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times.
122 CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times.
124 # This terminator creates a loop structure for CURLYX
125 WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches.
129 # OPEN,CLOSE,GROUPP ...are numbered at compile time.
130 OPEN OPEN, num 1 ; Mark this point in input as start of #n.
131 CLOSE CLOSE, num 1 ; Analogous to OPEN.
133 REF REF, num 1 V ; Match some already matched string
134 REFF REF, num 1 V ; Match already matched string, folded using native charset semantics for non-utf8
135 REFFL REF, num 1 V ; Match already matched string, folded in loc.
136 # N?REFF[AU] could have been implemented using the FLAGS field of the
137 # regnode, but by having a separate node type, we can use the existing switch
138 # statement to avoid some tests
139 REFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8
140 REFFA REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII
142 #*Named references. Code in regcomp.c assumes that these all are after the numbered references
143 NREF REF, no-sv 1 V ; Match some already matched string
144 NREFF REF, no-sv 1 V ; Match already matched string, folded using native charset semantics for non-utf8
145 NREFFL REF, no-sv 1 V ; Match already matched string, folded in loc.
146 NREFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8
147 NREFFA REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII
149 IFMATCH BRANCHJ, off 1 . 2 ; Succeeds if the following matches.
150 UNLESSM BRANCHJ, off 1 . 2 ; Fails if the following matches.
151 SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE.
152 IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher .
153 GROUPP GROUPP, num 1 ; Whether the group matched.
155 #*Support for long RE
157 LONGJMP LONGJMP, off 1 . 1 ; Jump far away.
158 BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset.
162 EVAL EVAL, evl 1 ; Execute some Perl code.
166 MINMOD MINMOD, no ; Next operator is not greedy.
167 LOGICAL LOGICAL, no ; Next opcode should set the flag only.
169 # This is not used yet
170 RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens.
174 # Behave the same as A|LIST|OF|WORDS would. The '..C' variants have
175 # inline charclass data (ascii only), the 'C' store it in the structure.
176 # NOTE: the relative order of the TRIE-like regops is significant
178 TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type
179 TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data
181 # For start classes, contains an added fail table.
182 AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type
183 AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data
186 GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2
187 GOSTART GOSTART, no ; recurse to start of pattern
189 #*Special conditionals
190 NGROUPP NGROUPP, no-sv 1 ; Whether the group matched.
191 INSUBP INSUBP, num 1 ; Whether we are in a specific recurse.
192 DEFINEP DEFINEP, none 1 ; Never execute directly.
195 ENDLIKE ENDLIKE, none ; Used only for the type field of verbs
196 OPFAIL ENDLIKE, none ; Same as (?!)
197 ACCEPT ENDLIKE, parno 1 ; Accepts the current matched string.
200 #*Verbs With Arguments
201 VERB VERB, no-sv 1 ; Used only for the type field of verbs
202 PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this
203 MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut.
204 SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying
205 COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this
206 CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group
208 #*Control what to keep in $&.
209 KEEPS KEEPS, no ; $& begins here.
211 #*New charclass like patterns
212 LNBREAK LNBREAK, none ; generic newline pattern
213 VERTWS VERTWS, none 0 S ; vertical whitespace (Perl 6)
214 NVERTWS NVERTWS, none 0 S ; not vertical whitespace (Perl 6)
215 HORIZWS HORIZWS, none 0 S ; horizontal whitespace (Perl 6)
216 NHORIZWS NHORIZWS, none 0 S ; not horizontal whitespace (Perl 6)
218 FOLDCHAR FOLDCHAR, codepoint 1 ; codepoint with tricky case folding properties.
221 # NEW STUFF SOMEWHERE ABOVE THIS LINE
223 ################################################################################
227 # This is not really a node, but an optimized away piece of a "long" node.
228 # To simplify debugging output, we mark it as if it were a node
229 OPTIMIZED NOTHING, off ; Placeholder for dump.
231 # Special opcode with the property that no opcode in a compiled program
232 # will ever be of this type. Thus it can be used as a flag value that
233 # no other opcode has been seen. END is used similarly, in that an END
234 # node cant be optimized. So END implies "unoptimizable" and PSEUDO mean
235 # "not seen anything to optimize yet".
236 PSEUDO PSEUDO, off ; Pseudo opcode for internal use.
238 -------------------------------------------------------------------------------
239 # Format for second section:
240 # REGOP \t typelist [ \t typelist] [# Comment]
245 # Anything below is a state
251 WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL
255 CURLY B_min_known,B_min,B_max:FAIL