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
21 END END, no ; End of program.
22 SUCCEED END, no ; Return from a subroutine, basically.
26 BOL BOL, no ; Match "" at beginning of line.
27 MBOL BOL, no ; Same, assuming multiline.
28 SBOL BOL, no ; Same, assuming singleline.
29 EOS EOL, no ; Match "" at end of string.
30 EOL EOL, no ; Match "" at end of line.
31 MEOL EOL, no ; Same, assuming multiline.
32 SEOL EOL, no ; Same, assuming singleline.
33 BOUND BOUND, no ; Match "" at any word boundary using native charset semantics for non-utf8
34 BOUNDL BOUND, no ; Match "" at any locale word boundary
35 BOUNDU BOUND, no ; Match "" at any word boundary using Unicode semantics
36 BOUNDA BOUND, no ; Match "" at any word boundary using ASCII semantics
37 # All NBOUND nodes are required by a line regexec.c to be greater than all BOUND ones
38 NBOUND NBOUND, no ; Match "" at any word non-boundary using native charset semantics for non-utf8
39 NBOUNDL NBOUND, no ; Match "" at any locale word non-boundary
40 NBOUNDU NBOUND, no ; Match "" at any word non-boundary using Unicode semantics
41 NBOUNDA NBOUND, no ; Match "" at any word non-boundary using ASCII semantics
42 GPOS GPOS, no ; Matches where last m//g left off.
44 #* [Special] alternatives:
46 REG_ANY REG_ANY, no 0 S ; Match any one character (except newline).
47 SANY REG_ANY, no 0 S ; Match any one character.
48 CANY REG_ANY, no 0 S ; Match any one byte.
49 ANYOF ANYOF, sv 0 S ; Match character in (or not in) this class, single char match only
50 ANYOFV ANYOF, sv 0 V ; Match character in (or not in) this class, can match-multiple chars
51 ALNUM ALNUM, no 0 S ; Match any alphanumeric character using native charset semantics for non-utf8
52 ALNUML ALNUM, no 0 S ; Match any alphanumeric char in locale
53 ALNUMU ALNUM, no 0 S ; Match any alphanumeric char using Unicode semantics
54 ALNUMA ALNUM, no 0 S ; Match [A-Za-z_0-9]
55 NALNUM NALNUM, no 0 S ; Match any non-alphanumeric character using native charset semantics for non-utf8
56 NALNUML NALNUM, no 0 S ; Match any non-alphanumeric char in locale
57 NALNUMU NALNUM, no 0 S ; Match any non-alphanumeric char using Unicode semantics
58 NALNUMA NALNUM, no 0 S ; Match [^A-Za-z_0-9]
59 SPACE SPACE, no 0 S ; Match any whitespace character using native charset semantics for non-utf8
60 SPACEL SPACE, no 0 S ; Match any whitespace char in locale
61 SPACEU SPACE, no 0 S ; Match any whitespace char using Unicode semantics
62 SPACEA SPACE, no 0 S ; Match [ \t\n\f\r]
63 NSPACE NSPACE, no 0 S ; Match any non-whitespace character using native charset semantics for non-utf8
64 NSPACEL NSPACE, no 0 S ; Match any non-whitespace char in locale
65 NSPACEU NSPACE, no 0 S ; Match any non-whitespace char using Unicode semantics
66 NSPACEA NSPACE, no 0 S ; Match [^ \t\n\f\r]
67 DIGIT DIGIT, no 0 S ; Match any numeric character using native charset semantics for non-utf8
68 DIGITL DIGIT, no 0 S ; Match any numeric character in locale
69 DIGITA DIGIT, no 0 S ; Match [0-9]
70 NDIGIT NDIGIT, no 0 S ; Match any non-numeric character using native charset semantics for non-utf8
71 NDIGITL NDIGIT, no 0 S ; Match any non-numeric character in locale
72 NDIGITA NDIGIT, no 0 S ; Match [^0-9]
73 CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence
77 # BRANCH The set of branches constituting a single choice are hooked
78 # together with their "next" pointers, since precedence prevents
79 # anything being concatenated to any individual branch. The
80 # "next" pointer of the last BRANCH in a choice points to the
81 # thing following the whole choice. This is also where the
82 # final "next" pointer of each individual branch points; each
83 # branch starts with the operand node of a BRANCH node.
85 BRANCH BRANCH, node 0 V ; Match this alternative, or the next...
89 # BACK Normal "next" pointers all implicitly point forward; BACK
90 # exists to make loop structures possible.
92 BACK BACK, no 0 V ; Match "", "next" ptr points backward.
96 EXACT EXACT, str ; Match this string (preceded by length).
97 EXACTF EXACT, str ; Match this string, folded, native charset semantics for non-utf8 (prec. by length).
98 EXACTFL EXACT, str ; Match this string, folded in locale (w/len).
99 EXACTFU EXACT, str ; Match this string, folded, Unicode semantics for non-utf8 (prec. by length).
103 NOTHING NOTHING, no ; Match empty string.
104 # A variant of above which delimits a group, thus stops optimizations
105 TAIL NOTHING, no ; Match empty string. Can jump here from outside.
109 # STAR,PLUS '?', and complex '*' and '+', are implemented as circular
110 # BRANCH structures using BACK. Simple cases (one character
111 # per match) are implemented with STAR and PLUS for speed
112 # and to minimize recursive plunges.
114 STAR STAR, node 0 V ; Match this (simple) thing 0 or more times.
115 PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times.
117 CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times.
118 CURLYN CURLY, no 2 V ; Capture next-after-this simple thing
119 CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times.
120 CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times.
122 # This terminator creates a loop structure for CURLYX
123 WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches.
127 # OPEN,CLOSE,GROUPP ...are numbered at compile time.
128 OPEN OPEN, num 1 ; Mark this point in input as start of #n.
129 CLOSE CLOSE, num 1 ; Analogous to OPEN.
131 REF REF, num 1 V ; Match some already matched string
132 REFF REF, num 1 V ; Match already matched string, folded using native charset semantics for non-utf8
133 REFFL REF, num 1 V ; Match already matched string, folded in loc.
134 # REFFU and NREFFU could have been implemented using the FLAGS field of the
135 # regnode, but by having a separate node type, we can use the existing switch
136 # statement to avoid some tests
137 REFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8
139 #*Named references. Code in regcomp.c assumes that these all are after the numbered references
140 NREF REF, no-sv 1 V ; Match some already matched string
141 NREFF REF, no-sv 1 V ; Match already matched string, folded using native charset semantics for non-utf8
142 NREFFL REF, no-sv 1 V ; Match already matched string, folded in loc.
143 NREFFU REF, num 1 V ; Match already matched string, folded using unicode semantics for non-utf8
145 IFMATCH BRANCHJ, off 1 . 2 ; Succeeds if the following matches.
146 UNLESSM BRANCHJ, off 1 . 2 ; Fails if the following matches.
147 SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE.
148 IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher .
149 GROUPP GROUPP, num 1 ; Whether the group matched.
151 #*Support for long RE
153 LONGJMP LONGJMP, off 1 . 1 ; Jump far away.
154 BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset.
158 EVAL EVAL, evl 1 ; Execute some Perl code.
162 MINMOD MINMOD, no ; Next operator is not greedy.
163 LOGICAL LOGICAL, no ; Next opcode should set the flag only.
165 # This is not used yet
166 RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens.
170 # Behave the same as A|LIST|OF|WORDS would. The '..C' variants have
171 # inline charclass data (ascii only), the 'C' store it in the structure.
172 # NOTE: the relative order of the TRIE-like regops is significant
174 TRIE TRIE, trie 1 ; Match many EXACT(F[LU]?)? at once. flags==type
175 TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data
177 # For start classes, contains an added fail table.
178 AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type
179 AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data
182 GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2
183 GOSTART GOSTART, no ; recurse to start of pattern
185 #*Special conditionals
186 NGROUPP NGROUPP, no-sv 1 ; Whether the group matched.
187 INSUBP INSUBP, num 1 ; Whether we are in a specific recurse.
188 DEFINEP DEFINEP, none 1 ; Never execute directly.
191 ENDLIKE ENDLIKE, none ; Used only for the type field of verbs
192 OPFAIL ENDLIKE, none ; Same as (?!)
193 ACCEPT ENDLIKE, parno 1 ; Accepts the current matched string.
196 #*Verbs With Arguments
197 VERB VERB, no-sv 1 ; Used only for the type field of verbs
198 PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this
199 MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut.
200 SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying
201 COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this
202 CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group
204 #*Control what to keep in $&.
205 KEEPS KEEPS, no ; $& begins here.
207 #*New charclass like patterns
208 LNBREAK LNBREAK, none ; generic newline pattern
209 VERTWS VERTWS, none 0 S ; vertical whitespace (Perl 6)
210 NVERTWS NVERTWS, none 0 S ; not vertical whitespace (Perl 6)
211 HORIZWS HORIZWS, none 0 S ; horizontal whitespace (Perl 6)
212 NHORIZWS NHORIZWS, none 0 S ; not horizontal whitespace (Perl 6)
214 FOLDCHAR FOLDCHAR, codepoint 1 ; codepoint with tricky case folding properties.
217 # NEW STUFF SOMEWHERE ABOVE THIS LINE
219 ################################################################################
223 # This is not really a node, but an optimized away piece of a "long" node.
224 # To simplify debugging output, we mark it as if it were a node
225 OPTIMIZED NOTHING, off ; Placeholder for dump.
227 # Special opcode with the property that no opcode in a compiled program
228 # will ever be of this type. Thus it can be used as a flag value that
229 # no other opcode has been seen. END is used similarly, in that an END
230 # node cant be optimized. So END implies "unoptimizable" and PSEUDO mean
231 # "not seen anything to optimize yet".
232 PSEUDO PSEUDO, off ; Pseudo opcode for internal use.
234 -------------------------------------------------------------------------------
235 # Format for second section:
236 # REGOP \t typelist [ \t typelist] [# Comment]
241 # Anything below is a state
247 WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL
251 CURLY B_min_known,B_min,B_max:FAIL