This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
regcomp.c: Properly dereference a ptr
[perl5.git] / regcomp.sym
CommitLineData
03363afd
YO
1# regcomp.sym
2#
3# File has two sections, divided by a line of dashes '-'.
4#
65aa4ca7
FC
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
7# of a description.
03363afd 8#
486ec47a 9# First section is for regops, second section is for regmatch-states
03363afd 10#
3dab1dad
YO
11# Note that the order in this file is important.
12#
03363afd 13# Format for first section:
519ebf3d
KW
14# NAME \s+ TYPE, arg-description [num-args] [flags] [longjump-len] ; DESCRIPTION
15# flag <S> means is REGNODE_SIMPLE; flag <V> means is REGNODE_VARIES
03363afd 16#
3dab1dad 17#
c476f425 18# run perl regen.pl after editing this file
3dab1dad 19
03363afd
YO
20
21
381b57db 22#* Exit points
1de06328 23
f8abb37e
NC
24END END, no ; End of program.
25SUCCEED END, no ; Return from a subroutine, basically.
d09b2d29 26
d3d47aac 27#* Line Start Anchors:
1645b83c 28#Note flags field for SBOL indicates if it is a /^/ or a /\A/
d3d47aac
YO
29SBOL BOL, no ; Match "" at beginning of line: /^/, /\A/
30MBOL BOL, no ; Same, assuming multiline: /^/m
31
32#* Line End Anchors:
33SEOL EOL, no ; Match "" at end of line: /$/
34MEOL EOL, no ; Same, assuming multiline: /$/m
35EOS EOL, no ; Match "" at end of string: /\z/
36
37#* Match Start Anchors:
38GPOS GPOS, no ; Matches where last m//g left off.
39
40#* Word Boundary Opcodes:
693fefec
KW
41# The regops that have varieties that vary depending on the character set regex
42# modifiers have to ordered thusly: /d, /l, /u, /a, /aa. This is because code
43# in regcomp.c uses the enum value of the modifier as an offset from the /d
44# version. The complements must come after the non-complements.
3018b823 45# BOUND, POSIX and their complements are affected, as well as EXACTF.
850b7ec9 46BOUND BOUND, no ; Match "" at any word boundary using native charset rules for non-utf8
f2284805 47BOUNDL BOUND, no ; Match "" at any locale word boundary
850b7ec9
KW
48BOUNDU BOUND, no ; Match "" at any word boundary using Unicode rules
49BOUNDA BOUND, no ; Match "" at any word boundary using ASCII rules
693fefec 50# All NBOUND nodes are required by code in regexec.c to be greater than all BOUND ones
850b7ec9 51NBOUND NBOUND, no ; Match "" at any word non-boundary using native charset rules for non-utf8
f2284805 52NBOUNDL NBOUND, no ; Match "" at any locale word non-boundary
850b7ec9
KW
53NBOUNDU NBOUND, no ; Match "" at any word non-boundary using Unicode rules
54NBOUNDA NBOUND, no ; Match "" at any word non-boundary using ASCII rules
d09b2d29 55
381b57db 56#* [Special] alternatives:
f9ef50a7
NC
57REG_ANY REG_ANY, no 0 S ; Match any one character (except newline).
58SANY REG_ANY, no 0 S ; Match any one character.
59CANY REG_ANY, no 0 S ; Match any one byte.
975a06f7 60ANYOF ANYOF, sv 1 S ; Match character in (or not in) this class, single char match only
693fefec 61
d3d47aac 62#* POSIX Character Classes:
3018b823
KW
63# Order of the below is important. See ordering comment above.
64POSIXD POSIXD, none 0 S ; Some [[:class:]] under /d; the FLAGS field gives which one
65POSIXL POSIXD, none 0 S ; Some [[:class:]] under /l; the FLAGS field gives which one
66POSIXU POSIXD, none 0 S ; Some [[:class:]] under /u; the FLAGS field gives which one
3615ea58 67POSIXA POSIXD, none 0 S ; Some [[:class:]] under /a; the FLAGS field gives which one
3018b823
KW
68NPOSIXD NPOSIXD, none 0 S ; complement of POSIXD, [[:^class:]]
69NPOSIXL NPOSIXD, none 0 S ; complement of POSIXL, [[:^class:]]
70NPOSIXU NPOSIXD, none 0 S ; complement of POSIXU, [[:^class:]]
9e84774b 71NPOSIXA NPOSIXD, none 0 S ; complement of POSIXA, [[:^class:]]
3018b823 72# End of order is important
693fefec 73
2448cf39 74CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence
d09b2d29 75
381b57db 76#* Alternation
1de06328 77
65aa4ca7
FC
78#* BRANCH The set of branches constituting a single choice are
79#* hooked together with their "next" pointers, since
80#* precedence prevents anything being concatenated to
81#* any individual branch. The "next" pointer of the last
82#* BRANCH in a choice points to the thing following the
83#* whole choice. This is also where the final "next"
84#* pointer of each individual branch points; each branch
85#* starts with the operand node of a BRANCH node.
86#*
f9ef50a7 87BRANCH BRANCH, node 0 V ; Match this alternative, or the next...
d09b2d29 88
381b57db 89#*Back pointer
1de06328 90
65aa4ca7
FC
91#* BACK Normal "next" pointers all implicitly point forward;
92#* BACK exists to make loop structures possible.
93#* not used
f9ef50a7 94BACK BACK, no 0 V ; Match "", "next" ptr points backward.
d09b2d29 95
65aa4ca7
FC
96#*Literals
97# NOTE: the relative ordering of these types is important do not change it
1de06328 98
f8abb37e 99EXACT EXACT, str ; Match this string (preceded by length).
a36afff1
KW
100EXACTF EXACT, str ; Match this non-UTF-8 string (not guaranteed to be folded) using /id rules (w/len).
101EXACTFL EXACT, str ; Match this string (not guaranteed to be folded) using /il rules (w/len).
3c760661 102EXACTFU 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).
8c1182fd 103EXACTFA EXACT, str ; Match this string (not guaranteed to be folded) using /iaa rules (w/len).
3c760661 104EXACTFU_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).
098b07d5 105EXACTFA_NO_TRIE EXACT, str ; Match this string (which is not trie-able; not guaranteed to be folded) using /iaa rules (w/len).
d09b2d29 106
381b57db 107#*Do nothing types
1de06328 108
f8abb37e 109NOTHING NOTHING, no ; Match empty string.
65aa4ca7 110#*A variant of above which delimits a group, thus stops optimizations
f8abb37e 111TAIL NOTHING, no ; Match empty string. Can jump here from outside.
d09b2d29 112
381b57db 113#*Loops
1de06328 114
65aa4ca7
FC
115#* STAR,PLUS '?', and complex '*' and '+', are implemented as
116#* circular BRANCH structures using BACK. Simple cases
117#* (one character per match) are implemented with STAR
118#* and PLUS for speed and to minimize recursive plunges.
119#*
f9ef50a7
NC
120STAR STAR, node 0 V ; Match this (simple) thing 0 or more times.
121PLUS PLUS, node 0 V ; Match this (simple) thing 1 or more times.
d09b2d29 122
f9ef50a7
NC
123CURLY CURLY, sv 2 V ; Match this simple thing {n,m} times.
124CURLYN CURLY, no 2 V ; Capture next-after-this simple thing
125CURLYM CURLY, no 2 V ; Capture this medium-complex thing {n,m} times.
126CURLYX CURLY, sv 2 V ; Match this complex thing {n,m} times.
d09b2d29 127
65aa4ca7 128#*This terminator creates a loop structure for CURLYX
f9ef50a7 129WHILEM WHILEM, no 0 V ; Do curly processing and see if rest matches.
d09b2d29 130
381b57db 131#*Buffer related
1de06328 132
65aa4ca7 133#*OPEN,CLOSE,GROUPP ...are numbered at compile time.
f8abb37e
NC
134OPEN OPEN, num 1 ; Mark this point in input as start of #n.
135CLOSE CLOSE, num 1 ; Analogous to OPEN.
d09b2d29 136
f9ef50a7 137REF REF, num 1 V ; Match some already matched string
850b7ec9 138REFF REF, num 1 V ; Match already matched string, folded using native charset rules for non-utf8
f9ef50a7 139REFFL REF, num 1 V ; Match already matched string, folded in loc.
781aab5c 140# N?REFF[AU] could have been implemented using the FLAGS field of the
01f98ec2
KW
141# regnode, but by having a separate node type, we can use the existing switch
142# statement to avoid some tests
850b7ec9
KW
143REFFU REF, num 1 V ; Match already matched string, folded using unicode rules for non-utf8
144REFFA REF, num 1 V ; Match already matched string, folded using unicode rules for non-utf8, no mixing ASCII, non-ASCII
d09b2d29 145
65aa4ca7
FC
146#*Named references. Code in regcomp.c assumes that these all are after
147#*the numbered references
01f98ec2 148NREF REF, no-sv 1 V ; Match some already matched string
850b7ec9 149NREFF REF, no-sv 1 V ; Match already matched string, folded using native charset rules for non-utf8
01f98ec2 150NREFFL REF, no-sv 1 V ; Match already matched string, folded in loc.
850b7ec9
KW
151NREFFU REF, num 1 V ; Match already matched string, folded using unicode rules for non-utf8
152NREFFA REF, num 1 V ; Match already matched string, folded using unicode rules for non-utf8, no mixing ASCII, non-ASCII
1de06328 153
d3d47aac
YO
154#*Support for long RE
155LONGJMP LONGJMP, off 1 . 1 ; Jump far away.
156BRANCHJ BRANCHJ, off 1 V 1 ; BRANCH with long offset.
157
158#*Special Case Regops
f9ef50a7
NC
159IFMATCH BRANCHJ, off 1 . 2 ; Succeeds if the following matches.
160UNLESSM BRANCHJ, off 1 . 2 ; Fails if the following matches.
161SUSPEND BRANCHJ, off 1 V 1 ; "Independent" sub-RE.
65aa4ca7 162IFTHEN BRANCHJ, off 1 V 1 ; Switch, should be preceded by switcher.
f8abb37e 163GROUPP GROUPP, num 1 ; Whether the group matched.
d09b2d29 164
d09b2d29 165
381b57db 166#*The heavy worker
1de06328 167
13f27704 168EVAL EVAL, evl/flags 2L ; Execute some Perl code.
d09b2d29 169
381b57db 170#*Modifiers
1de06328 171
f8abb37e
NC
172MINMOD MINMOD, no ; Next operator is not greedy.
173LOGICAL LOGICAL, no ; Next opcode should set the flag only.
d09b2d29 174
65aa4ca7 175#*This is not used yet
f9ef50a7 176RENUM BRANCHJ, off 1 . 1 ; Group with independently numbered parens.
d09b2d29 177
381b57db 178#*Trie Related
1de06328 179
65aa4ca7
FC
180#* Behave the same as A|LIST|OF|WORDS would. The '..C' variants
181#* have inline charclass data (ascii only), the 'C' store it in the
182#* structure.
486ec47a 183# NOTE: the relative order of the TRIE-like regops is significant
ce5e9471 184
7986cb47 185TRIE TRIE, trie 1 ; Match many EXACT(F[ALU]?)? at once. flags==type
f8abb37e 186TRIEC TRIE,trie charclass ; Same as TRIE, but with embedded charclass data
3dab1dad 187
1de06328 188# For start classes, contains an added fail table.
f8abb37e
NC
189AHOCORASICK TRIE, trie 1 ; Aho Corasick stclass. flags==type
190AHOCORASICKC TRIE,trie charclass ; Same as AHOCORASICK, but with embedded charclass data
1de06328 191
381b57db 192#*Regex Subroutines
f8abb37e
NC
193GOSUB GOSUB, num/ofs 2L ; recurse to paren arg1 at (signed) ofs arg2
194GOSTART GOSTART, no ; recurse to start of pattern
03363afd 195
381b57db 196#*Special conditionals
f8abb37e
NC
197NGROUPP NGROUPP, no-sv 1 ; Whether the group matched.
198INSUBP INSUBP, num 1 ; Whether we are in a specific recurse.
199DEFINEP DEFINEP, none 1 ; Never execute directly.
0a4db386 200
486ec47a 201#*Backtracking Verbs
f8abb37e
NC
202ENDLIKE ENDLIKE, none ; Used only for the type field of verbs
203OPFAIL ENDLIKE, none ; Same as (?!)
204ACCEPT ENDLIKE, parno 1 ; Accepts the current matched string.
5d458dd8
YO
205
206
207#*Verbs With Arguments
f8abb37e
NC
208VERB VERB, no-sv 1 ; Used only for the type field of verbs
209PRUNE VERB, no-sv 1 ; Pattern fails at this startpoint if no-backtracking through this
210MARKPOINT VERB, no-sv 1 ; Push the current location for rollback by cut.
211SKIP VERB, no-sv 1 ; On failure skip forward (to the mark) before retrying
212COMMIT VERB, no-sv 1 ; Pattern fails outright if backtracking through this
213CUTGROUP VERB, no-sv 1 ; On failure go to the next alternation in the group
e2e6a0f1 214
ee9b8eae 215#*Control what to keep in $&.
f8abb37e 216KEEPS KEEPS, no ; $& begins here.
ee9b8eae 217
e1d1eefb 218#*New charclass like patterns
f8abb37e 219LNBREAK LNBREAK, none ; generic newline pattern
3172e3fd 220
381b57db 221# NEW STUFF SOMEWHERE ABOVE THIS LINE
1de06328 222
03363afd
YO
223################################################################################
224
7f69552c 225#*SPECIAL REGOPS
1de06328 226
65aa4ca7
FC
227#* This is not really a node, but an optimized away piece of a "long"
228#* node. To simplify debugging output, we mark it as if it were a node
f8abb37e 229OPTIMIZED NOTHING, off ; Placeholder for dump.
1de06328 230
65aa4ca7
FC
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
235#* mean "not seen anything to optimize yet".
f8abb37e 236PSEUDO PSEUDO, off ; Pseudo opcode for internal use.
1de06328 237
03363afd
YO
238-------------------------------------------------------------------------------
239# Format for second section:
65aa4ca7 240# REGOP \t typelist [ \t typelist]
03363afd
YO
241# typelist= namelist
242# = namelist:FAIL
243# = name:count
244
245# Anything below is a state
246#
247#
f8abb37e
NC
248TRIE next:FAIL
249EVAL AB:FAIL
250CURLYX end:FAIL
251WHILEM A_pre,A_min,A_max,B_min,B_max:FAIL
252BRANCH next:FAIL
253CURLYM A,B:FAIL
254IFMATCH A:FAIL
255CURLY B_min_known,B_min,B_max:FAIL
256COMMIT next:FAIL
257MARKPOINT next:FAIL
258SKIP next:FAIL
259CUTGROUP next:FAIL
260KEEPS next:FAIL