Annotation of embedaddon/pcre/doc/pcrematching.3, revision 1.1

1.1     ! misho       1: .TH PCREMATCHING 3
        !             2: .SH NAME
        !             3: PCRE - Perl-compatible regular expressions
        !             4: .SH "PCRE MATCHING ALGORITHMS"
        !             5: .rs
        !             6: .sp
        !             7: This document describes the two different algorithms that are available in PCRE
        !             8: for matching a compiled regular expression against a given subject string. The
        !             9: "standard" algorithm is the one provided by the \fBpcre_exec()\fP function.
        !            10: This works in the same was as Perl's matching function, and provides a
        !            11: Perl-compatible matching operation.
        !            12: .P
        !            13: An alternative algorithm is provided by the \fBpcre_dfa_exec()\fP function;
        !            14: this operates in a different way, and is not Perl-compatible. It has advantages
        !            15: and disadvantages compared with the standard algorithm, and these are described
        !            16: below.
        !            17: .P
        !            18: When there is only one possible way in which a given subject string can match a
        !            19: pattern, the two algorithms give the same answer. A difference arises, however,
        !            20: when there are multiple possibilities. For example, if the pattern
        !            21: .sp
        !            22:   ^<.*>
        !            23: .sp
        !            24: is matched against the string
        !            25: .sp
        !            26:   <something> <something else> <something further>
        !            27: .sp
        !            28: there are three possible answers. The standard algorithm finds only one of
        !            29: them, whereas the alternative algorithm finds all three.
        !            30: .
        !            31: .SH "REGULAR EXPRESSIONS AS TREES"
        !            32: .rs
        !            33: .sp
        !            34: The set of strings that are matched by a regular expression can be represented
        !            35: as a tree structure. An unlimited repetition in the pattern makes the tree of
        !            36: infinite size, but it is still a tree. Matching the pattern to a given subject
        !            37: string (from a given starting point) can be thought of as a search of the tree.
        !            38: There are two ways to search a tree: depth-first and breadth-first, and these
        !            39: correspond to the two matching algorithms provided by PCRE.
        !            40: .
        !            41: .SH "THE STANDARD MATCHING ALGORITHM"
        !            42: .rs
        !            43: .sp
        !            44: In the terminology of Jeffrey Friedl's book "Mastering Regular
        !            45: Expressions", the standard algorithm is an "NFA algorithm". It conducts a
        !            46: depth-first search of the pattern tree. That is, it proceeds along a single
        !            47: path through the tree, checking that the subject matches what is required. When
        !            48: there is a mismatch, the algorithm tries any alternatives at the current point,
        !            49: and if they all fail, it backs up to the previous branch point in the tree, and
        !            50: tries the next alternative branch at that level. This often involves backing up
        !            51: (moving to the left) in the subject string as well. The order in which
        !            52: repetition branches are tried is controlled by the greedy or ungreedy nature of
        !            53: the quantifier.
        !            54: .P
        !            55: If a leaf node is reached, a matching string has been found, and at that point
        !            56: the algorithm stops. Thus, if there is more than one possible match, this
        !            57: algorithm returns the first one that it finds. Whether this is the shortest,
        !            58: the longest, or some intermediate length depends on the way the greedy and
        !            59: ungreedy repetition quantifiers are specified in the pattern.
        !            60: .P
        !            61: Because it ends up with a single path through the tree, it is relatively
        !            62: straightforward for this algorithm to keep track of the substrings that are
        !            63: matched by portions of the pattern in parentheses. This provides support for
        !            64: capturing parentheses and back references.
        !            65: .
        !            66: .SH "THE ALTERNATIVE MATCHING ALGORITHM"
        !            67: .rs
        !            68: .sp
        !            69: This algorithm conducts a breadth-first search of the tree. Starting from the
        !            70: first matching point in the subject, it scans the subject string from left to
        !            71: right, once, character by character, and as it does this, it remembers all the
        !            72: paths through the tree that represent valid matches. In Friedl's terminology,
        !            73: this is a kind of "DFA algorithm", though it is not implemented as a
        !            74: traditional finite state machine (it keeps multiple states active
        !            75: simultaneously).
        !            76: .P
        !            77: Although the general principle of this matching algorithm is that it scans the
        !            78: subject string only once, without backtracking, there is one exception: when a
        !            79: lookaround assertion is encountered, the characters following or preceding the
        !            80: current point have to be independently inspected.
        !            81: .P
        !            82: The scan continues until either the end of the subject is reached, or there are
        !            83: no more unterminated paths. At this point, terminated paths represent the
        !            84: different matching possibilities (if there are none, the match has failed).
        !            85: Thus, if there is more than one possible match, this algorithm finds all of
        !            86: them, and in particular, it finds the longest. The matches are returned in
        !            87: decreasing order of length. There is an option to stop the algorithm after the
        !            88: first match (which is necessarily the shortest) is found.
        !            89: .P
        !            90: Note that all the matches that are found start at the same point in the
        !            91: subject. If the pattern
        !            92: .sp
        !            93:   cat(er(pillar)?)?
        !            94: .sp
        !            95: is matched against the string "the caterpillar catchment", the result will be
        !            96: the three strings "caterpillar", "cater", and "cat" that start at the fifth
        !            97: character of the subject. The algorithm does not automatically move on to find
        !            98: matches that start at later positions.
        !            99: .P
        !           100: There are a number of features of PCRE regular expressions that are not
        !           101: supported by the alternative matching algorithm. They are as follows:
        !           102: .P
        !           103: 1. Because the algorithm finds all possible matches, the greedy or ungreedy
        !           104: nature of repetition quantifiers is not relevant. Greedy and ungreedy
        !           105: quantifiers are treated in exactly the same way. However, possessive
        !           106: quantifiers can make a difference when what follows could also match what is
        !           107: quantified, for example in a pattern like this:
        !           108: .sp
        !           109:   ^a++\ew!
        !           110: .sp
        !           111: This pattern matches "aaab!" but not "aaa!", which would be matched by a
        !           112: non-possessive quantifier. Similarly, if an atomic group is present, it is
        !           113: matched as if it were a standalone pattern at the current point, and the
        !           114: longest match is then "locked in" for the rest of the overall pattern.
        !           115: .P
        !           116: 2. When dealing with multiple paths through the tree simultaneously, it is not
        !           117: straightforward to keep track of captured substrings for the different matching
        !           118: possibilities, and PCRE's implementation of this algorithm does not attempt to
        !           119: do this. This means that no captured substrings are available.
        !           120: .P
        !           121: 3. Because no substrings are captured, back references within the pattern are
        !           122: not supported, and cause errors if encountered.
        !           123: .P
        !           124: 4. For the same reason, conditional expressions that use a backreference as the
        !           125: condition or test for a specific group recursion are not supported.
        !           126: .P
        !           127: 5. Because many paths through the tree may be active, the \eK escape sequence,
        !           128: which resets the start of the match when encountered (but may be on some paths
        !           129: and not on others), is not supported. It causes an error if encountered.
        !           130: .P
        !           131: 6. Callouts are supported, but the value of the \fIcapture_top\fP field is
        !           132: always 1, and the value of the \fIcapture_last\fP field is always -1.
        !           133: .P
        !           134: 7. The \eC escape sequence, which (in the standard algorithm) matches a single
        !           135: byte, even in UTF-8 mode, is not supported in UTF-8 mode, because the
        !           136: alternative algorithm moves through the subject string one character at a time,
        !           137: for all active paths through the tree.
        !           138: .P
        !           139: 8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE) are not
        !           140: supported. (*FAIL) is supported, and behaves like a failing negative assertion.
        !           141: .
        !           142: .SH "ADVANTAGES OF THE ALTERNATIVE ALGORITHM"
        !           143: .rs
        !           144: .sp
        !           145: Using the alternative matching algorithm provides the following advantages:
        !           146: .P
        !           147: 1. All possible matches (at a single point in the subject) are automatically
        !           148: found, and in particular, the longest match is found. To find more than one
        !           149: match using the standard algorithm, you have to do kludgy things with
        !           150: callouts.
        !           151: .P
        !           152: 2. Because the alternative algorithm scans the subject string just once, and
        !           153: never needs to backtrack, it is possible to pass very long subject strings to
        !           154: the matching function in several pieces, checking for partial matching each
        !           155: time. Although it is possible to do multi-segment matching using the standard
        !           156: algorithm (\fBpcre_exec()\fP), by retaining partially matched substrings, it is
        !           157: more complicated. The
        !           158: .\" HREF
        !           159: \fBpcrepartial\fP
        !           160: .\"
        !           161: documentation gives details of partial matching and discusses multi-segment
        !           162: matching.
        !           163: .
        !           164: .
        !           165: .SH "DISADVANTAGES OF THE ALTERNATIVE ALGORITHM"
        !           166: .rs
        !           167: .sp
        !           168: The alternative algorithm suffers from a number of disadvantages:
        !           169: .P
        !           170: 1. It is substantially slower than the standard algorithm. This is partly
        !           171: because it has to search for all possible matches, but is also because it is
        !           172: less susceptible to optimization.
        !           173: .P
        !           174: 2. Capturing parentheses and back references are not supported.
        !           175: .P
        !           176: 3. Although atomic groups are supported, their use does not provide the
        !           177: performance advantage that it does for the standard algorithm.
        !           178: .
        !           179: .
        !           180: .SH AUTHOR
        !           181: .rs
        !           182: .sp
        !           183: .nf
        !           184: Philip Hazel
        !           185: University Computing Service
        !           186: Cambridge CB2 3QH, England.
        !           187: .fi
        !           188: .
        !           189: .
        !           190: .SH REVISION
        !           191: .rs
        !           192: .sp
        !           193: .nf
        !           194: Last updated: 19 November 2011
        !           195: Copyright (c) 1997-2010 University of Cambridge.
        !           196: .fi

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>