.\" -*- mode: troff; coding: utf-8 -*- .\" Automatically generated by Pod::Man 5.01 (Pod::Simple 3.43) .\" .\" Standard preamble: .\" ======================================================================== .de Sp \" Vertical space (when we can't use .PP) .if t .sp .5v .if n .sp .. .de Vb \" Begin verbatim text .ft CW .nf .ne \\$1 .. .de Ve \" End verbatim text .ft R .fi .. .\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>. .ie n \{\ . ds C` "" . ds C' "" 'br\} .el\{\ . ds C` . ds C' 'br\} .\" .\" Escape single quotes in literal strings from groff's Unicode transform. .ie \n(.g .ds Aq \(aq .el .ds Aq ' .\" .\" If the F register is >0, we'll generate index entries on stderr for .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index .\" entries marked with X<> in POD. Of course, you'll have to process the .\" output yourself in some meaningful fashion. .\" .\" Avoid warning from groff about undefined register 'F'. .de IX .. .nr rF 0 .if \n(.g .if rF .nr rF 1 .if (\n(rF:(\n(.g==0)) \{\ . if \nF \{\ . de IX . tm Index:\\$1\t\\n%\t"\\$2" .. . if !\nF==2 \{\ . nr % 0 . nr F 2 . \} . \} .\} .rr rF .\" ======================================================================== .\" .IX Title "PERLREGUTS 1" .TH PERLREGUTS 1 2024-04-17 "perl v5.38.2" "Perl Programmers Reference Guide" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l .nh .SH NAME perlreguts \- Description of the Perl regular expression engine. .SH DESCRIPTION .IX Header "DESCRIPTION" This document is an attempt to shine some light on the guts of the regex engine and how it works. The regex engine represents a significant chunk of the perl codebase, but is relatively poorly understood. This document is a meagre attempt at addressing this situation. It is derived from the author's experience, comments in the source code, other papers on the regex engine, feedback on the perl5\-porters mail list, and no doubt other places as well. .PP \&\fBNOTICE!\fR It should be clearly understood that the behavior and structures discussed in this represents the state of the engine as the author understood it at the time of writing. It is \fBNOT\fR an API definition, it is purely an internals guide for those who want to hack the regex engine, or understand how the regex engine works. Readers of this document are expected to understand perl's regex syntax and its usage in detail. If you want to learn about the basics of Perl's regular expressions, see perlre. And if you want to replace the regex engine with your own, see perlreapi. .SH OVERVIEW .IX Header "OVERVIEW" .SS "A quick note on terms" .IX Subsection "A quick note on terms" There is some debate as to whether to say "regexp" or "regex". In this document we will use the term "regex" unless there is a special reason not to, in which case we will explain why. .PP When speaking about regexes we need to distinguish between their source code form and their internal form. In this document we will use the term "pattern" when we speak of their textual, source code form, and the term "program" when we speak of their internal representation. These correspond to the terms \fIS\-regex\fR and \fIB\-regex\fR that Mark Jason Dominus employs in his paper on "Rx" ([1] in "REFERENCES"). .SS "What is a regular expression engine?" .IX Subsection "What is a regular expression engine?" A regular expression engine is a program that takes a set of constraints specified in a mini-language, and then applies those constraints to a target string, and determines whether or not the string satisfies the constraints. See perlre for a full definition of the language. .PP In less grandiose terms, the first part of the job is to turn a pattern into something the computer can efficiently use to find the matching point in the string, and the second part is performing the search itself. .PP To do this we need to produce a program by parsing the text. We then need to execute the program to find the point in the string that matches. And we need to do the whole thing efficiently. .SS "Structure of a Regexp Program" .IX Subsection "Structure of a Regexp Program" \fIHigh Level\fR .IX Subsection "High Level" .PP Although it is a bit confusing and some people object to the terminology, it is worth taking a look at a comment that has been in \fIregexp.h\fR for years: .PP \&\fIThis is essentially a linear encoding of a nondeterministic finite-state machine (aka syntax charts or "railroad normal form" in parsing technology).\fR .PP The term "railroad normal form" is a bit esoteric, with "syntax diagram/charts", or "railroad diagram/charts" being more common terms. Nevertheless it provides a useful mental image of a regex program: each node can be thought of as a unit of track, with a single entry and in most cases a single exit point (there are pieces of track that fork, but statistically not many), and the whole forms a layout with a single entry and single exit point. The matching process can be thought of as a car that moves along the track, with the particular route through the system being determined by the character read at each possible connector point. A car can fall off the track at any point but it may only proceed as long as it matches the track. .PP Thus the pattern \f(CW\*(C`/foo(?:\ew+|\ed+|\es+)bar/\*(C'\fR can be thought of as the following chart: .PP .Vb 10 \& [start] \& | \& \& | \& +\-\-\-\-\-+\-\-\-\-\-+ \& | | | \& <\ew+> <\ed+> <\es+> \& | | | \& +\-\-\-\-\-+\-\-\-\-\-+ \& | \& \& | \& [end] .Ve .PP The truth of the matter is that perl's regular expressions these days are much more complex than this kind of structure, but visualising it this way can help when trying to get your bearings, and it matches the current implementation pretty closely. .PP To be more precise, we will say that a regex program is an encoding of a graph. Each node in the graph corresponds to part of the original regex pattern, such as a literal string or a branch, and has a pointer to the nodes representing the next component to be matched. Since "node" and "opcode" already have other meanings in the perl source, we will call the nodes in a regex program "regops". .PP The program is represented by an array of \f(CW\*(C`regnode\*(C'\fR structures, one or more of which represent a single regop of the program. Struct \&\f(CW\*(C`regnode\*(C'\fR is the smallest struct needed, and has a field structure which is shared with all the other larger structures. (Outside this document, the term "regnode" is sometimes used to mean "regop", which could be confusing.) .PP The "next" pointers of all regops except \f(CW\*(C`BRANCH\*(C'\fR implement concatenation; a "next" pointer with a \f(CW\*(C`BRANCH\*(C'\fR on both ends of it is connecting two alternatives. [Here we have one of the subtle syntax dependencies: an individual \f(CW\*(C`BRANCH\*(C'\fR (as opposed to a collection of them) is never concatenated with anything because of operator precedence.] .PP The operand of some types of regop is a literal string; for others, it is a regop leading into a sub-program. In particular, the operand of a \f(CW\*(C`BRANCH\*(C'\fR node is the first regop of the branch. .PP \&\fBNOTE\fR: As the railroad metaphor suggests, this is \fBnot\fR a tree structure: the tail of the branch connects to the thing following the set of \f(CW\*(C`BRANCH\*(C'\fRes. It is a like a single line of railway track that splits as it goes into a station or railway yard and rejoins as it comes out the other side. .PP \fIRegops\fR .IX Subsection "Regops" .PP The base structure of a regop is defined in \fIregexp.h\fR as follows: .PP .Vb 5 \& struct regnode { \& U8 flags; /* Various purposes, sometimes overridden */ \& U8 type; /* Opcode value as specified by regnodes.h */ \& U16 next_off; /* Offset in size regnode */ \& }; .Ve .PP Other larger \f(CW\*(C`regnode\*(C'\fR\-like structures are defined in \fIregcomp.h\fR. They are almost like subclasses in that they have the same fields as \&\f(CW\*(C`regnode\*(C'\fR, with possibly additional fields following in the structure, and in some cases the specific meaning (and name) of some of base fields are overridden. The following is a more complete description. .ie n .IP """regnode_1""" 4 .el .IP \f(CWregnode_1\fR 4 .IX Item "regnode_1" .PD 0 .ie n .IP """regnode_2""" 4 .el .IP \f(CWregnode_2\fR 4 .IX Item "regnode_2" .PD \&\f(CW\*(C`regnode_1\*(C'\fR structures have the same header, followed by a single four-byte argument; \f(CW\*(C`regnode_2\*(C'\fR structures contain two two-byte arguments instead: .Sp .Vb 2 \& regnode_1 U32 arg1; \& regnode_2 U16 arg1; U16 arg2; .Ve .ie n .IP """regnode_string""" 4 .el .IP \f(CWregnode_string\fR 4 .IX Item "regnode_string" \&\f(CW\*(C`regnode_string\*(C'\fR structures, used for literal strings, follow the header with a one-byte length and then the string data. Strings are padded on the tail end with zero bytes so that the total length of the node is a multiple of four bytes: .Sp .Vb 2 \& regnode_string char string[1]; \& U8 str_len; /* overrides flags */ .Ve .ie n .IP """regnode_charclass""" 4 .el .IP \f(CWregnode_charclass\fR 4 .IX Item "regnode_charclass" Bracketed character classes are represented by \f(CW\*(C`regnode_charclass\*(C'\fR structures, which have a four-byte argument and then a 32\-byte (256\-bit) bitmap indicating which characters in the Latin1 range are included in the class. .Sp .Vb 2 \& regnode_charclass U32 arg1; \& char bitmap[ANYOF_BITMAP_SIZE]; .Ve .Sp Various flags whose names begin with \f(CW\*(C`ANYOF_\*(C'\fR are used for special situations. Above Latin1 matches and things not known until run-time are stored in "Perl's pprivate structure". .ie n .IP """regnode_charclass_posixl""" 4 .el .IP \f(CWregnode_charclass_posixl\fR 4 .IX Item "regnode_charclass_posixl" There is also a larger form of a char class structure used to represent POSIX char classes under \f(CW\*(C`/l\*(C'\fR matching, called \f(CW\*(C`regnode_charclass_posixl\*(C'\fR which has an additional 32\-bit bitmap indicating which POSIX char classes have been included. .Sp .Vb 3 \& regnode_charclass_posixl U32 arg1; \& char bitmap[ANYOF_BITMAP_SIZE]; \& U32 classflags; .Ve .PP \&\fIregnodes.h\fR defines an array called \f(CW\*(C`PL_regnode_arg_len[]\*(C'\fR which gives the size of each opcode in units of \f(CW\*(C`size regnode\*(C'\fR (4\-byte). A macro is used to calculate the size of an \f(CW\*(C`EXACT\*(C'\fR node based on its \f(CW\*(C`str_len\*(C'\fR field. .PP The regops are defined in \fIregnodes.h\fR which is generated from \&\fIregcomp.sym\fR by \fIregcomp.pl\fR. Currently the maximum possible number of distinct regops is restricted to 256, with about a quarter already used. .PP A set of macros makes accessing the fields easier and more consistent. These include \f(CWOP()\fR, which is used to determine the type of a \f(CW\*(C`regnode\*(C'\fR\-like structure; \f(CWNEXT_OFF()\fR, which is the offset to the next node (more on this later); \f(CWARG()\fR, \f(CWARG1()\fR, \f(CWARG2()\fR, \f(CWARG_SET()\fR, and equivalents for reading and setting the arguments; and \f(CWSTR_LEN()\fR, \&\f(CWSTRING()\fR and \f(CWOPERAND()\fR for manipulating strings and regop bearing types. .PP \fIWhat regnode is next?\fR .IX Subsection "What regnode is next?" .PP There are two distinct concepts of "next regnode" in the regex engine, and it is important to keep them distinct in your thinking as they overlap conceptually in many places, but where they don't overlap the difference is critical. For the majority of regnode types the two concepts are (nearly) identical in practice. The two types are \&\f(CW\*(C`REGNODE_AFTER\*(C'\fR which is used heavily during compilation but only occasionally during execution and \f(CW\*(C`regnext\*(C'\fR which is used heavily during execution, and only occasionally during compilation. .IP """REGNODE_AFTER""" 4 .IX Item """REGNODE_AFTER""" This is the "positionally next regnode" in the compiled regex program. For the smaller regnode types it is \f(CW\*(C`regnode_ptr+1\*(C'\fR under the hood, but as regnode sizes vary and can change over time we offer macros which hide the gory details. .Sp It is heavily used in the compiler phase but is only used by a few select regnode types in the execution phase. It is also heavily used in the code for dumping the regexp program for debugging. .Sp There are a selection of macros which can be used to compute this as efficiently as possible depending on the circumstances. The canonical macro is \f(CWREGNODE_AFTER()\fR, which is the most powerful and should handle any case we have, but is also potentially the slowest. There are two additional macros for the special case that you KNOW the current regnode size is constant, and you know its type or opcode. In which case you can use \f(CWREGNODE_AFTER_opcode()\fR or \f(CWREGNODE_AFTER_type()\fR. .Sp In older versions of the regex engine \f(CWREGNODE_AFTER()\fR was called \&\f(CW\*(C`NEXTOPER\*(C'\fR but this was found to be confusing and it was renamed. There is also a \f(CWREGNODE_BEFORE()\fR, but it is unsafe and should not be used in new code. .IP """regnext""" 4 .IX Item """regnext""" This is the regnode which can be reached by jumping forward by the value of the \f(CWNEXT_OFF()\fR member of the regnode, or in a few cases for longer jumps by the \f(CW\*(C`arg1\*(C'\fR field of the \f(CW\*(C`regnode_1\*(C'\fR structure. The subroutine \&\f(CWregnext()\fR handles this transparently. In the majority of cases the \&\f(CW\*(C`regnext\*(C'\fR for a regnode is the regnode which should be executed after the current one has successfully matched, but in some cases this may not be true. In loop control and branch control regnode types the regnext may signify something special, for BRANCH nodes \f(CW\*(C`regnext\*(C'\fR is the next BRANCH that should be executed if the current one fails execution, and some loop control regnodes set the regnext to be the end of the loop so they can jump to their cleanup if the current iteration fails to match. .PP Most regnode types do not create a branch in the execution flow, and leaving aside optimizations the two concepts of "next" are the same. For instance the \f(CW\*(C`regnext\*(C'\fR and \f(CW\*(C`REGNODE_AFTER\*(C'\fR of a SBOL opcode are the same during compilation phase. The main place this is not true is \&\f(CW\*(C`BRANCH\*(C'\fR regnodes where the \f(CW\*(C`REGNODE_AFTER\*(C'\fR represents the start of the pattern in the branch and the \f(CW\*(C`regnext\*(C'\fR represents the linkage to the next BRANCH should this one fail to match, or 0 if it is the last branch. The looping logic for quantifiers also makes similar use of the distinction between the two types, with \f(CW\*(C`REGNODE_AFTER\*(C'\fR being the inside of the loop construct, and the \f(CW\*(C`regnext\*(C'\fR pointing at the end of the loop. .PP During compilation the engine may not know what the regnext is for a given node, so during compilation \f(CW\*(C`regnext\*(C'\fR is only used where it must be used and is known to be correct. At the very end of the compilation phase we walk the regex program and correct the regnext data as appropriate, and also perform various optimizations which may result in regnodes that were required during construction becoming redundant, or we may replace a large regnode with a much smaller one and filling in the gap with OPTIMIZED regnodes. Thus we might start with something like this: .PP .Vb 5 \& BRANCH \& EXACT "foo" \& BRANCH \& EXACT "bar" \& EXACT "!" .Ve .PP and replace it with something like: .PP .Vb 5 \& TRIE foo|bar \& OPTIMIZED \& OPTIMIZED \& OPTIMIZED \& EXACT "!" .Ve .PP the \f(CW\*(C`REGNODE_AFTER\*(C'\fR for the \f(CW\*(C`TRIE\*(C'\fR node would be an \f(CW\*(C`OPTIMIZED\*(C'\fR regnode, and in theory the \f(CW\*(C`regnext\*(C'\fR would be the same as the \&\f(CW\*(C`REGNODE_AFTER\*(C'\fR. But it would be inefficient to execute the OPTIMIZED regnode as a noop three times, so the optimizer fixes the \f(CW\*(C`regnext\*(C'\fR so such nodes are skipped during execution phase. .PP During execution phases we use the \f(CWregnext()\fR almost exclusively, and only use \f(CW\*(C`REGNODE_AFTER\*(C'\fR in special cases where it has a well defined meaning for a given regnode type. For instance /x+/ results in .PP .Vb 3 \& PLUS \& EXACT "x" \& END .Ve .PP the \f(CW\*(C`regnext\*(C'\fR of the \f(CW\*(C`PLUS\*(C'\fR regnode is the \f(CW\*(C`END\*(C'\fR regnode, and the \&\f(CW\*(C`REGNODE_AFTER\*(C'\fR of the \f(CW\*(C`PLUS\*(C'\fR regnode is the \f(CW\*(C`EXACT\*(C'\fR regnode. The \&\f(CW\*(C`regnext\*(C'\fR and \f(CW\*(C`REGNODE_AFTER\*(C'\fR of the \f(CW\*(C`EXACT\*(C'\fR regnode is the \&\f(CW\*(C`END\*(C'\fR regnode. .SH "Process Overview" .IX Header "Process Overview" Broadly speaking, performing a match of a string against a pattern involves the following steps: .IP "A. Compilation" 5 .IX Item "A. Compilation" .RS 5 .PD 0 .IP "1. Parsing" 5 .IX Item "1. Parsing" .IP "2. Peep-hole optimisation and analysis" 5 .IX Item "2. Peep-hole optimisation and analysis" .RE .RS 5 .RE .IP "B. Execution" 5 .IX Item "B. Execution" .RS 5 .IP "3. Start position and no-match optimisations" 5 .IX Item "3. Start position and no-match optimisations" .IP "4. Program execution" 5 .IX Item "4. Program execution" .RE .RS 5 .RE .PD .PP Where these steps occur in the actual execution of a perl program is determined by whether the pattern involves interpolating any string variables. If interpolation occurs, then compilation happens at run time. If it does not, then compilation is performed at compile time. (The \f(CW\*(C`/o\*(C'\fR modifier changes this, as does \f(CW\*(C`qr//\*(C'\fR to a certain extent.) The engine doesn't really care that much. .SS Compilation .IX Subsection "Compilation" This code resides primarily in \fIregcomp.c\fR, along with the header files \&\fIregcomp.h\fR, \fIregexp.h\fR and \fIregnodes.h\fR. .PP Compilation starts with \f(CWpregcomp()\fR, which is mostly an initialisation wrapper which farms work out to two other routines for the heavy lifting: the first is \f(CWreg()\fR, which is the start point for parsing; the second, \&\f(CWstudy_chunk()\fR, is responsible for optimisation. .PP Initialisation in \f(CWpregcomp()\fR mostly involves the creation and data-filling of a special structure, \f(CW\*(C`RExC_state_t\*(C'\fR (defined in \fIregcomp.c\fR). Almost all internally-used routines in \fIregcomp.h\fR take a pointer to one of these structures as their first argument, with the name \f(CW\*(C`pRExC_state\*(C'\fR. This structure is used to store the compilation state and contains many fields. Likewise there are many macros which operate on this variable: anything that looks like \f(CW\*(C`RExC_xxxx\*(C'\fR is a macro that operates on this pointer/structure. .PP \&\f(CWreg()\fR is the start of the parse process. It is responsible for parsing an arbitrary chunk of pattern up to either the end of the string, or the first closing parenthesis it encounters in the pattern. This means it can be used to parse the top-level regex, or any section inside of a grouping parenthesis. It also handles the "special parens" that perl's regexes have. For instance when parsing \f(CW\*(C`/x(?:foo)y/\*(C'\fR, \&\f(CWreg()\fR will at one point be called to parse from the "?" symbol up to and including the ")". .PP Additionally, \f(CWreg()\fR is responsible for parsing the one or more branches from the pattern, and for "finishing them off" by correctly setting their next pointers. In order to do the parsing, it repeatedly calls out to \f(CWregbranch()\fR, which is responsible for handling up to the first \f(CW\*(C`|\*(C'\fR symbol it sees. .PP \&\f(CWregbranch()\fR in turn calls \f(CWregpiece()\fR which handles "things" followed by a quantifier. In order to parse the "things", \f(CWregatom()\fR is called. This is the lowest level routine, which parses out constant strings, character classes, and the various special symbols like \f(CW\*(C`$\*(C'\fR. If \f(CWregatom()\fR encounters a "(" character it in turn calls \f(CWreg()\fR. .PP There used to be two main passes involved in parsing, the first to calculate the size of the compiled program, and the second to actually compile it. But now there is only one main pass, with an initial crude guess based on the length of the input pattern, which is increased if necessary as parsing proceeds, and afterwards, trimmed to the actual amount used. .PP However, it may happen that parsing must be restarted at the beginning when various circumstances occur along the way. An example is if the program turns out to be so large that there are jumps in it that won't fit in the normal 16 bits available. There are two special regops that can hold bigger jump destinations, BRANCHJ and LONGBRANCH. The parse is restarted, and these are used instead of the normal shorter ones. Whenever restarting the parse is required, the function returns failure and sets a flag as to what needs to be done. This is passed up to the top level routine which takes the appropriate action and restarts from scratch. In the case of needing longer jumps, the \f(CW\*(C`RExC_use_BRANCHJ\*(C'\fR flag is set in the \f(CW\*(C`RExC_state_t\*(C'\fR structure, which the functions know to inspect before deciding how to do branches. .PP In most instances, the function that discovers the issue sets the causal flag and returns failure immediately. "Parsing complications" contains an explicit example of how this works. In other cases, such as a forward reference to a numbered parenthetical grouping, we need to finish the parse to know if that numbered grouping actually appears in the pattern. In those cases, the parse is just redone at the end, with the knowledge of how many groupings occur in it. .PP The routine \f(CWregtail()\fR is called by both \f(CWreg()\fR and \f(CWregbranch()\fR in order to "set the tail pointer" correctly. When executing and we get to the end of a branch, we need to go to the node following the grouping parens. When parsing, however, we don't know where the end will be until we get there, so when we do we must go back and update the offsets as appropriate. \f(CW\*(C`regtail\*(C'\fR is used to make this easier. .PP A subtlety of the parsing process means that a regex like \f(CW\*(C`/foo/\*(C'\fR is originally parsed into an alternation with a single branch. It is only afterwards that the optimiser converts single branch alternations into the simpler form. .PP \fIParse Call Graph and a Grammar\fR .IX Subsection "Parse Call Graph and a Grammar" .PP The call graph looks like this: .PP .Vb 10 \& reg() # parse a top level regex, or inside of \& # parens \& regbranch() # parse a single branch of an alternation \& regpiece() # parse a pattern followed by a quantifier \& regatom() # parse a simple pattern \& regclass() # used to handle a class \& reg() # used to handle a parenthesised \& # subpattern \& .... \& ... \& regtail() # finish off the branch \& ... \& regtail() # finish off the branch sequence. Tie each \& # branch\*(Aqs tail to the tail of the \& # sequence \& # (NEW) In Debug mode this is \& # regtail_study(). .Ve .PP A grammar form might be something like this: .PP .Vb 11 \& atom : constant | class \& quant : \*(Aq*\*(Aq | \*(Aq+\*(Aq | \*(Aq?\*(Aq | \*(Aq{min,max}\*(Aq \& _branch: piece \& | piece _branch \& | nothing \& branch: _branch \& | _branch \*(Aq|\*(Aq branch \& group : \*(Aq(\*(Aq branch \*(Aq)\*(Aq \& _piece: atom | group \& piece : _piece \& | _piece quant .Ve .PP \fIParsing complications\fR .IX Subsection "Parsing complications" .PP The implication of the above description is that a pattern containing nested parentheses will result in a call graph which cycles through \f(CWreg()\fR, \&\f(CWregbranch()\fR, \f(CWregpiece()\fR, \f(CWregatom()\fR, \f(CWreg()\fR, \f(CWregbranch()\fR \fIetc\fR multiple times, until the deepest level of nesting is reached. All the above routines return a pointer to a \f(CW\*(C`regnode\*(C'\fR, which is usually the last regnode added to the program. However, one complication is that \fBreg()\fR returns NULL for parsing \f(CW\*(C`(?:)\*(C'\fR syntax for embedded modifiers, setting the flag \&\f(CW\*(C`TRYAGAIN\*(C'\fR. The \f(CW\*(C`TRYAGAIN\*(C'\fR propagates upwards until it is captured, in some cases by \f(CWregatom()\fR, but otherwise unconditionally by \&\f(CWregbranch()\fR. Hence it will never be returned by \f(CWregbranch()\fR to \&\f(CWreg()\fR. This flag permits patterns such as \f(CW\*(C`(?i)+\*(C'\fR to be detected as errors (\fIQuantifier follows nothing in regex; marked by <\-\- HERE in m/(?i)+ <\-\- HERE /\fR). .PP Another complication is that the representation used for the program differs if it needs to store Unicode, but it's not always possible to know for sure whether it does until midway through parsing. The Unicode representation for the program is larger, and cannot be matched as efficiently. (See "Unicode and Localisation Support" below for more details as to why.) If the pattern contains literal Unicode, it's obvious that the program needs to store Unicode. Otherwise, the parser optimistically assumes that the more efficient representation can be used, and starts sizing on this basis. However, if it then encounters something in the pattern which must be stored as Unicode, such as an \f(CW\*(C`\ex{...}\*(C'\fR escape sequence representing a character literal, then this means that all previously calculated sizes need to be redone, using values appropriate for the Unicode representation. This is another instance where the parsing needs to be restarted, and it can and is done immediately. The function returns failure, and sets the flag \f(CW\*(C`RESTART_UTF8\*(C'\fR (encapsulated by using the macro \f(CW\*(C`REQUIRE_UTF8\*(C'\fR). This restart request is propagated up the call chain in a similar fashion, until it is "caught" in \f(CWPerl_re_op_compile()\fR, which marks the pattern as containing Unicode, and restarts the sizing pass. It is also possible for constructions within run-time code blocks to turn out to need Unicode representation., which is signalled by \&\f(CWS_compile_runtime_code()\fR returning false to \f(CWPerl_re_op_compile()\fR. .PP The restart was previously implemented using a \f(CW\*(C`longjmp\*(C'\fR in \f(CWregatom()\fR back to a \f(CW\*(C`setjmp\*(C'\fR in \f(CWPerl_re_op_compile()\fR, but this proved to be problematic as the latter is a large function containing many automatic variables, which interact badly with the emergent control flow of \f(CW\*(C`setjmp\*(C'\fR. .PP \fIDebug Output\fR .IX Subsection "Debug Output" .PP Starting in the 5.9.x development version of perl you can \f(CW\*(C`use re Debug => \*(AqPARSE\*(Aq\*(C'\fR to see some trace information about the parse process. We will start with some simple patterns and build up to more complex patterns. .PP So when we parse \f(CW\*(C`/foo/\*(C'\fR we see something like the following table. The left shows what is being parsed, and the number indicates where the next regop would go. The stuff on the right is the trace output of the graph. The names are chosen to be short to make it less dense on the screen. 'tsdy' is a special form of \f(CWregtail()\fR which does some extra analysis. .PP .Vb 6 \& >foo< 1 reg \& brnc \& piec \& atom \& >< 4 tsdy~ EXACT (EXACT) (1) \& ~ attach to END (3) offset to 2 .Ve .PP The resulting program then looks like: .PP .Vb 2 \& 1: EXACT (3) \& 3: END(0) .Ve .PP As you can see, even though we parsed out a branch and a piece, it was ultimately only an atom. The final program shows us how things work. We have an \f(CW\*(C`EXACT\*(C'\fR regop, followed by an \f(CW\*(C`END\*(C'\fR regop. The number in parens indicates where the \f(CW\*(C`regnext\*(C'\fR of the node goes. The \f(CW\*(C`regnext\*(C'\fR of an \f(CW\*(C`END\*(C'\fR regop is unused, as \f(CW\*(C`END\*(C'\fR regops mean we have successfully matched. The number on the left indicates the position of the regop in the regnode array. .PP Now let's try a harder pattern. We will add a quantifier, so now we have the pattern \&\f(CW\*(C`/foo+/\*(C'\fR. We will see that \f(CWregbranch()\fR calls \f(CWregpiece()\fR twice. .PP .Vb 10 \& >foo+< 1 reg \& brnc \& piec \& atom \& >o+< 3 piec \& atom \& >< 6 tail~ EXACT (1) \& 7 tsdy~ EXACT (EXACT) (1) \& ~ PLUS (END) (3) \& ~ attach to END (6) offset to 3 .Ve .PP And we end up with the program: .PP .Vb 4 \& 1: EXACT (3) \& 3: PLUS(6) \& 4: EXACT (0) \& 6: END(0) .Ve .PP Now we have a special case. The \f(CW\*(C`EXACT\*(C'\fR regop has a \f(CW\*(C`regnext\*(C'\fR of 0. This is because if it matches it should try to match itself again. The \f(CW\*(C`PLUS\*(C'\fR regop handles the actual failure of the \f(CW\*(C`EXACT\*(C'\fR regop and acts appropriately (going to regnode 6 if the \f(CW\*(C`EXACT\*(C'\fR matched at least once, or failing if it didn't). .PP Now for something much more complex: \f(CW\*(C`/x(?:foo*|b[a][rR])(foo|bar)$/\*(C'\fR .PP .Vb 10 \& >x(?:foo*|b... 1 reg \& brnc \& piec \& atom \& >(?:foo*|b[... 3 piec \& atom \& >?:foo*|b[a... reg \& >foo*|b[a][... brnc \& piec \& atom \& >o*|b[a][rR... 5 piec \& atom \& >|b[a][rR])... 8 tail~ EXACT (3) \& >b[a][rR])(... 9 brnc \& 10 piec \& atom \& >[a][rR])(f... 12 piec \& atom \& >a][rR])(fo... clas \& >[rR])(foo|... 14 tail~ EXACT (10) \& piec \& atom \& >rR])(foo|b... clas \& >)(foo|bar)... 25 tail~ EXACT (12) \& tail~ BRANCH (3) \& 26 tsdy~ BRANCH (END) (9) \& ~ attach to TAIL (25) offset to 16 \& tsdy~ EXACT (EXACT) (4) \& ~ STAR (END) (6) \& ~ attach to TAIL (25) offset to 19 \& tsdy~ EXACT (EXACT) (10) \& ~ EXACT (EXACT) (12) \& ~ ANYOF[Rr] (END) (14) \& ~ attach to TAIL (25) offset to 11 \& >(foo|bar)$< tail~ EXACT (1) \& piec \& atom \& >foo|bar)$< reg \& 28 brnc \& piec \& atom \& >|bar)$< 31 tail~ OPEN1 (26) \& >bar)$< brnc \& 32 piec \& atom \& >)$< 34 tail~ BRANCH (28) \& 36 tsdy~ BRANCH (END) (31) \& ~ attach to CLOSE1 (34) offset to 3 \& tsdy~ EXACT (EXACT) (29) \& ~ attach to CLOSE1 (34) offset to 5 \& tsdy~ EXACT (EXACT) (32) \& ~ attach to CLOSE1 (34) offset to 2 \& >$< tail~ BRANCH (3) \& ~ BRANCH (9) \& ~ TAIL (25) \& piec \& atom \& >< 37 tail~ OPEN1 (26) \& ~ BRANCH (28) \& ~ BRANCH (31) \& ~ CLOSE1 (34) \& 38 tsdy~ EXACT (EXACT) (1) \& ~ BRANCH (END) (3) \& ~ BRANCH (END) (9) \& ~ TAIL (END) (25) \& ~ OPEN1 (END) (26) \& ~ BRANCH (END) (28) \& ~ BRANCH (END) (31) \& ~ CLOSE1 (END) (34) \& ~ EOL (END) (36) \& ~ attach to END (37) offset to 1 .Ve .PP Resulting in the program .PP .Vb 10 \& 1: EXACT (3) \& 3: BRANCH(9) \& 4: EXACT (6) \& 6: STAR(26) \& 7: EXACT (0) \& 9: BRANCH(25) \& 10: EXACT (14) \& 12: OPTIMIZED (2 nodes) \& 14: ANYOF[Rr](26) \& 25: TAIL(26) \& 26: OPEN1(28) \& 28: TRIE\-EXACT(34) \& [StS:1 Wds:2 Cs:6 Uq:5 #Sts:7 Mn:3 Mx:3 Stcls:bf] \& \& \& 30: OPTIMIZED (4 nodes) \& 34: CLOSE1(36) \& 36: EOL(37) \& 37: END(0) .Ve .PP Here we can see a much more complex program, with various optimisations in play. At regnode 10 we see an example where a character class with only one character in it was turned into an \f(CW\*(C`EXACT\*(C'\fR node. We can also see where an entire alternation was turned into a \f(CW\*(C`TRIE\-EXACT\*(C'\fR node. As a consequence, some of the regnodes have been marked as optimised away. We can see that the \f(CW\*(C`$\*(C'\fR symbol has been converted into an \f(CW\*(C`EOL\*(C'\fR regop, a special piece of code that looks for \f(CW\*(C`\en\*(C'\fR or the end of the string. .PP The next pointer for \f(CW\*(C`BRANCH\*(C'\fRes is interesting in that it points at where execution should go if the branch fails. When executing, if the engine tries to traverse from a branch to a \f(CW\*(C`regnext\*(C'\fR that isn't a branch then the engine will know that the entire set of branches has failed. .PP \fIPeep-hole Optimisation and Analysis\fR .IX Subsection "Peep-hole Optimisation and Analysis" .PP The regular expression engine can be a weighty tool to wield. On long strings and complex patterns it can end up having to do a lot of work to find a match, and even more to decide that no match is possible. Consider a situation like the following pattern. .PP .Vb 1 \& \*(Aqababababababababababab\*(Aq =~ /(a|b)*z/ .Ve .PP The \f(CW\*(C`(a|b)*\*(C'\fR part can match at every char in the string, and then fail every time because there is no \f(CW\*(C`z\*(C'\fR in the string. So obviously we can avoid using the regex engine unless there is a \f(CW\*(C`z\*(C'\fR in the string. Likewise in a pattern like: .PP .Vb 1 \& /foo(\ew+)bar/ .Ve .PP In this case we know that the string must contain a \f(CW\*(C`foo\*(C'\fR which must be followed by \f(CW\*(C`bar\*(C'\fR. We can use Fast Boyer-Moore matching as implemented in \f(CWfbm_instr()\fR to find the location of these strings. If they don't exist then we don't need to resort to the much more expensive regex engine. Even better, if they do exist then we can use their positions to reduce the search space that the regex engine needs to cover to determine if the entire pattern matches. .PP There are various aspects of the pattern that can be used to facilitate optimisations along these lines: .IP \(bu 5 anchored fixed strings .IP \(bu 5 floating fixed strings .IP \(bu 5 minimum and maximum length requirements .IP \(bu 5 start class .IP \(bu 5 Beginning/End of line positions .PP Another form of optimisation that can occur is the post-parse "peep-hole" optimisation, where inefficient constructs are replaced by more efficient constructs. The \f(CW\*(C`TAIL\*(C'\fR regops which are used during parsing to mark the end of branches and the end of groups are examples of this. These regops are used as place-holders during construction and "always match" so they can be "optimised away" by making the things that point to the \f(CW\*(C`TAIL\*(C'\fR point to the thing that \f(CW\*(C`TAIL\*(C'\fR points to, thus "skipping" the node. .PP Another optimisation that can occur is that of "\f(CW\*(C`EXACT\*(C'\fR merging" which is where two consecutive \f(CW\*(C`EXACT\*(C'\fR nodes are merged into a single regop. An even more aggressive form of this is that a branch sequence of the form \f(CW\*(C`EXACT BRANCH ... EXACT\*(C'\fR can be converted into a \&\f(CW\*(C`TRIE\-EXACT\*(C'\fR regop. .PP All of this occurs in the routine \f(CWstudy_chunk()\fR which uses a special structure \f(CW\*(C`scan_data_t\*(C'\fR to store the analysis that it has performed, and does the "peep-hole" optimisations as it goes. .PP The code involved in \f(CWstudy_chunk()\fR is extremely cryptic. Be careful. :\-) .SS Execution .IX Subsection "Execution" Execution of a regex generally involves two phases, the first being finding the start point in the string where we should match from, and the second being running the regop interpreter. .PP If we can tell that there is no valid start point then we don't bother running the interpreter at all. Likewise, if we know from the analysis phase that we cannot detect a short-cut to the start position, we go straight to the interpreter. .PP The two entry points are \f(CWre_intuit_start()\fR and \f(CWpregexec()\fR. These routines have a somewhat incestuous relationship with overlap between their functions, and \f(CWpregexec()\fR may even call \f(CWre_intuit_start()\fR on its own. Nevertheless other parts of the perl source code may call into either, or both. .PP Execution of the interpreter itself used to be recursive, but thanks to the efforts of Dave Mitchell in the 5.9.x development track, that has changed: now an internal stack is maintained on the heap and the routine is fully iterative. This can make it tricky as the code is quite conservative about what state it stores, with the result that two consecutive lines in the code can actually be running in totally different contexts due to the simulated recursion. .PP \fIStart position and no-match optimisations\fR .IX Subsection "Start position and no-match optimisations" .PP \&\f(CWre_intuit_start()\fR is responsible for handling start points and no-match optimisations as determined by the results of the analysis done by \&\f(CWstudy_chunk()\fR (and described in "Peep-hole Optimisation and Analysis"). .PP The basic structure of this routine is to try to find the start\- and/or end-points of where the pattern could match, and to ensure that the string is long enough to match the pattern. It tries to use more efficient methods over less efficient methods and may involve considerable cross-checking of constraints to find the place in the string that matches. For instance it may try to determine that a given fixed string must be not only present but a certain number of chars before the end of the string, or whatever. .PP It calls several other routines, such as \f(CWfbm_instr()\fR which does Fast Boyer Moore matching and \f(CWfind_byclass()\fR which is responsible for finding the start using the first mandatory regop in the program. .PP When the optimisation criteria have been satisfied, \f(CWreg_try()\fR is called to perform the match. .PP \fIProgram execution\fR .IX Subsection "Program execution" .PP \&\f(CWpregexec()\fR is the main entry point for running a regex. It contains support for initialising the regex interpreter's state, running \&\f(CWre_intuit_start()\fR if needed, and running the interpreter on the string from various start positions as needed. When it is necessary to use the regex interpreter \f(CWpregexec()\fR calls \f(CWregtry()\fR. .PP \&\f(CWregtry()\fR is the entry point into the regex interpreter. It expects as arguments a pointer to a \f(CW\*(C`regmatch_info\*(C'\fR structure and a pointer to a string. It returns an integer 1 for success and a 0 for failure. It is basically a set-up wrapper around \f(CWregmatch()\fR. .PP \&\f(CW\*(C`regmatch\*(C'\fR is the main "recursive loop" of the interpreter. It is basically a giant switch statement that implements a state machine, where the possible states are the regops themselves, plus a number of additional intermediate and failure states. A few of the states are implemented as subroutines but the bulk are inline code. .SH MISCELLANEOUS .IX Header "MISCELLANEOUS" .SS "Unicode and Localisation Support" .IX Subsection "Unicode and Localisation Support" When dealing with strings containing characters that cannot be represented using an eight-bit character set, perl uses an internal representation that is a permissive version of Unicode's UTF\-8 encoding[2]. This uses single bytes to represent characters from the ASCII character set, and sequences of two or more bytes for all other characters. (See perlunitut for more information about the relationship between UTF\-8 and perl's encoding, utf8. The difference isn't important for this discussion.) .PP No matter how you look at it, Unicode support is going to be a pain in a regex engine. Tricks that might be fine when you have 256 possible characters often won't scale to handle the size of the UTF\-8 character set. Things you can take for granted with ASCII may not be true with Unicode. For instance, in ASCII, it is safe to assume that \&\f(CW\*(C`sizeof(char1) == sizeof(char2)\*(C'\fR, but in UTF\-8 it isn't. Unicode case folding is vastly more complex than the simple rules of ASCII, and even when not using Unicode but only localised single byte encodings, things can get tricky (for example, \fBLATIN SMALL LETTER SHARP S\fR (U+00DF, ß) should match 'SS' in localised case-insensitive matching). .PP Making things worse is that UTF\-8 support was a later addition to the regex engine (as it was to perl) and this necessarily made things a lot more complicated. Obviously it is easier to design a regex engine with Unicode support in mind from the beginning than it is to retrofit it to one that wasn't. .PP Nearly all regops that involve looking at the input string have two cases, one for UTF\-8, and one not. In fact, it's often more complex than that, as the pattern may be UTF\-8 as well. .PP Care must be taken when making changes to make sure that you handle UTF\-8 properly, both at compile time and at execution time, including when the string and pattern are mismatched. .SS "Base Structures" .IX Subsection "Base Structures" The \f(CW\*(C`regexp\*(C'\fR structure described in perlreapi is common to all regex engines. Two of its fields are intended for the private use of the regex engine that compiled the pattern. These are the \&\f(CW\*(C`intflags\*(C'\fR and pprivate members. The \f(CW\*(C`pprivate\*(C'\fR is a void pointer to an arbitrary structure whose use and management is the responsibility of the compiling engine. perl will never modify either of these values. In the case of the stock engine the structure pointed to by \&\f(CW\*(C`pprivate\*(C'\fR is called \f(CW\*(C`regexp_internal\*(C'\fR. .PP Its \f(CW\*(C`pprivate\*(C'\fR and \f(CW\*(C`intflags\*(C'\fR fields contain data specific to each engine. .PP There are two structures used to store a compiled regular expression. One, the \f(CW\*(C`regexp\*(C'\fR structure described in perlreapi is populated by the engine currently being used and some of its fields read by perl to implement things such as the stringification of \f(CW\*(C`qr//\*(C'\fR. .PP The other structure is pointed to by the \f(CW\*(C`regexp\*(C'\fR struct's \&\f(CW\*(C`pprivate\*(C'\fR and is in addition to \f(CW\*(C`intflags\*(C'\fR in the same struct considered to be the property of the regex engine which compiled the regular expression; .PP The regexp structure contains all the data that perl needs to be aware of to properly work with the regular expression. It includes data about optimisations that perl can use to determine if the regex engine should really be used, and various other control info that is needed to properly execute patterns in various contexts such as is the pattern anchored in some way, or what flags were used during the compile, or whether the program contains special constructs that perl needs to be aware of. .PP In addition it contains two fields that are intended for the private use of the regex engine that compiled the pattern. These are the \f(CW\*(C`intflags\*(C'\fR and pprivate members. The \f(CW\*(C`pprivate\*(C'\fR is a void pointer to an arbitrary structure whose use and management is the responsibility of the compiling engine. perl will never modify either of these values. .PP As mentioned earlier, in the case of the default engines, the \f(CW\*(C`pprivate\*(C'\fR will be a pointer to a regexp_internal structure which holds the compiled program and any additional data that is private to the regex engine implementation. .PP \fIPerl's \fR\f(CI\*(C`pprivate\*(C'\fR\fI structure\fR .IX Subsection "Perl's pprivate structure" .PP The following structure is used as the \f(CW\*(C`pprivate\*(C'\fR struct by perl's regex engine. Since it is specific to perl it is only of curiosity value to other engine implementations. .PP .Vb 8 \& typedef struct regexp_internal { \& regnode *regstclass; \& struct reg_data *data; \& struct reg_code_blocks *code_blocks; \& U32 proglen; \& U32 name_list_idx; \& regnode program[1]; \& } regexp_internal; .Ve .PP Description of the attributes is as follows: .ie n .IP """regstclass""" 5 .el .IP \f(CWregstclass\fR 5 .IX Item "regstclass" Special regop that is used by \f(CWre_intuit_start()\fR to check if a pattern can match at a certain position. For instance if the regex engine knows that the pattern must start with a 'Z' then it can scan the string until it finds one and then launch the regex engine from there. The routine that handles this is called \f(CWfind_by_class()\fR. Sometimes this field points at a regop embedded in the program, and sometimes it points at an independent synthetic regop that has been constructed by the optimiser. .ie n .IP """data""" 5 .el .IP \f(CWdata\fR 5 .IX Item "data" This field points at a \f(CW\*(C`reg_data\*(C'\fR structure, which is defined as follows .Sp .Vb 5 \& struct reg_data { \& U32 count; \& U8 *what; \& void* data[1]; \& }; .Ve .Sp This structure is used for handling data structures that the regex engine needs to handle specially during a clone or free operation on the compiled product. Each element in the data array has a corresponding element in the what array. During compilation regops that need special structures stored will add an element to each array using the \fBadd_data()\fR routine and then store the index in the regop. .Sp In modern perls the 0th element of this structure is reserved and is NEVER used to store anything of use. This is to allow things that need to index into this array to represent "no value". .ie n .IP """code_blocks""" 5 .el .IP \f(CWcode_blocks\fR 5 .IX Item "code_blocks" This optional structure is used to manage \f(CW\*(C`(?{})\*(C'\fR constructs in the pattern. It is made up of the following structures. .Sp .Vb 7 \& /* record the position of a (?{...}) within a pattern */ \& struct reg_code_block { \& STRLEN start; \& STRLEN end; \& OP *block; \& REGEXP *src_regex; \& }; \& \& /* array of reg_code_block\*(Aqs plus header info */ \& struct reg_code_blocks { \& int refcnt; /* we may be pointed to from a regex \& and from the savestack */ \& int count; /* how many code blocks */ \& struct reg_code_block *cb; /* array of reg_code_block\*(Aqs */ \& }; .Ve .ie n .IP """proglen""" 5 .el .IP \f(CWproglen\fR 5 .IX Item "proglen" Stores the length of the compiled program in units of regops. .ie n .IP """name_list_idx""" 5 .el .IP \f(CWname_list_idx\fR 5 .IX Item "name_list_idx" This is the index into the data array where an AV is stored that contains the names of any named capture buffers in the pattern, should there be any. This is only used in the debugging version of the regex engine and when RXp_PAREN_NAMES(prog) is true. It will be 0 if there is no such data. .ie n .IP """program""" 5 .el .IP \f(CWprogram\fR 5 .IX Item "program" Compiled program. Inlined into the structure so the entire struct can be treated as a single blob. .SH "SEE ALSO" .IX Header "SEE ALSO" perlreapi .PP perlre .PP perlunitut .SH AUTHOR .IX Header "AUTHOR" by Yves Orton, 2006. .PP With excerpts from Perl, and contributions and suggestions from Ronald J. Kimball, Dave Mitchell, Dominic Dunlop, Mark Jason Dominus, Stephen McCamant, and David Landgren. .PP Now maintained by Perl 5 Porters. .SH LICENCE .IX Header "LICENCE" Same terms as Perl. .SH REFERENCES .IX Header "REFERENCES" [1] .PP [2]