/[pcre]/code/trunk/doc/pcreapi.3
ViewVC logotype

Diff of /code/trunk/doc/pcreapi.3

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 535 by ph10, Thu Jun 3 19:18:24 2010 UTC revision 567 by ph10, Sat Nov 6 17:10:00 2010 UTC
# Line 765  out of use. To avoid confusion, they hav Line 765  out of use. To avoid confusion, they hav
765    50  [this code is not in use]    50  [this code is not in use]
766    51  octal value is greater than \e377 (not in UTF-8 mode)    51  octal value is greater than \e377 (not in UTF-8 mode)
767    52  internal error: overran compiling workspace    52  internal error: overran compiling workspace
768    53  internal error: previously-checked referenced subpattern not found    53  internal error: previously-checked referenced subpattern
769            not found
770    54  DEFINE group contains more than one branch    54  DEFINE group contains more than one branch
771    55  repeating a DEFINE group is not allowed    55  repeating a DEFINE group is not allowed
772    56  inconsistent NEWLINE options    56  inconsistent NEWLINE options
# Line 778  out of use. To avoid confusion, they hav Line 779  out of use. To avoid confusion, they hav
779    62  subpattern name expected    62  subpattern name expected
780    63  digit expected after (?+    63  digit expected after (?+
781    64  ] is an invalid data character in JavaScript compatibility mode    64  ] is an invalid data character in JavaScript compatibility mode
782    65  different names for subpatterns of the same number are not allowed    65  different names for subpatterns of the same number are
783            not allowed
784    66  (*MARK) must have an argument    66  (*MARK) must have an argument
785    67  this version of PCRE is not compiled with PCRE_UCP support    67  this version of PCRE is not compiled with PCRE_UCP support
786  .sp  .sp
# Line 846  Studying a pattern is also useful for no Line 848  Studying a pattern is also useful for no
848  single fixed starting character. A bitmap of possible starting bytes is  single fixed starting character. A bitmap of possible starting bytes is
849  created. This speeds up finding a position in the subject at which to start  created. This speeds up finding a position in the subject at which to start
850  matching.  matching.
851    .P
852    The two optimizations just described can be disabled by setting the
853    PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
854    \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
855    callouts, or make use of (*MARK), and you make use of these in cases where
856    matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
857    .\" HTML <a href="#execoptions">
858    .\" </a>
859    below.
860    .\"
861  .  .
862  .  .
863  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
# Line 1438  the Line 1450  the
1450  .\" HREF  .\" HREF
1451  \fBpcredemo\fP  \fBpcredemo\fP
1452  .\"  .\"
1453  sample program.  sample program. In the most general case, you have to check to see if the
1454    newline convention recognizes CRLF as a newline, and if so, and the current
1455    character is CR followed by LF, advance the starting offset by two characters
1456    instead of one.
1457  .sp  .sp
1458    PCRE_NO_START_OPTIMIZE    PCRE_NO_START_OPTIMIZE
1459  .sp  .sp
1460  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1461  a match, in order to speed up the process. For example, if it is known that a  a match, in order to speed up the process. For example, if it is known that an
1462  match must start with a specific character, it searches the subject for that  unanchored match must start with a specific character, it searches the subject
1463  character, and fails immediately if it cannot find it, without actually running  for that character, and fails immediately if it cannot find it, without
1464  the main matching function. When callouts are in use, these optimizations can  actually running the main matching function. This means that a special item
1465  cause them to be skipped. This option disables the "start-up" optimizations,  such as (*COMMIT) at the start of a pattern is not considered until after a
1466  causing performance to suffer, but ensuring that the callouts do occur.  suitable starting point for the match has been found. When callouts or (*MARK)
1467    items are in use, these "start-up" optimizations can cause them to be skipped
1468    if the pattern is never actually used. The start-up optimizations are in effect
1469    a pre-scan of the subject that takes place before the pattern is run.
1470    .P
1471    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1472    causing performance to suffer, but ensuring that in cases where the result is
1473    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1474    are considered at every possible starting position in the subject string.
1475    Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1476    Consider the pattern
1477    .sp
1478      (*COMMIT)ABC
1479    .sp
1480    When this is compiled, PCRE records the fact that a match must start with the
1481    character "A". Suppose the subject string is "DEFABC". The start-up
1482    optimization scans along the subject, finds "A" and runs the first match
1483    attempt from there. The (*COMMIT) item means that the pattern must match the
1484    current starting position, which in this case, it does. However, if the same
1485    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1486    subject string does not happen. The first match attempt is run starting from
1487    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1488    the overall result is "no match". If the pattern is studied, more start-up
1489    optimizations may be used. For example, a minimum length for the subject may be
1490    recorded. Consider the pattern
1491    .sp
1492      (*MARK:A)(X|Y)
1493    .sp
1494    The minimum length for a match is one character. If the subject is "ABC", there
1495    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1496    If the pattern is studied, the final attempt does not take place, because PCRE
1497    knows that the subject is too short, and so the (*MARK) is never encountered.
1498    In this case, studying the pattern does not affect the overall match result,
1499    which is still "no match", but it does affect the auxiliary information that is
1500    returned.
1501  .sp  .sp
1502    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1503  .sp  .sp
# Line 1466  in the main Line 1515  in the main
1515  \fBpcre\fP  \fBpcre\fP
1516  .\"  .\"
1517  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1518  the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,  the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains a value that does
1519    not point to the start of a UTF-8 character (or to the end of the subject),
1520  PCRE_ERROR_BADUTF8_OFFSET is returned.  PCRE_ERROR_BADUTF8_OFFSET is returned.
1521  .P  .P
1522  If you already know that your subject is valid, and you want to skip these  If you already know that your subject is valid, and you want to skip these
# Line 1474  checks for performance reasons, you can Line 1524  checks for performance reasons, you can
1524  calling \fBpcre_exec()\fP. You might want to do this for the second and  calling \fBpcre_exec()\fP. You might want to do this for the second and
1525  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1526  all the matches in a single subject string. However, you should be sure that  all the matches in a single subject string. However, you should be sure that
1527  the value of \fIstartoffset\fP points to the start of a UTF-8 character. When  the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1528  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1529  subject, or a value of \fIstartoffset\fP that does not point to the start of a  invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1530  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1531  .sp  .sp
1532    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
1533    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
# Line 1486  These options turn on the partial matchi Line 1536  These options turn on the partial matchi
1536  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1537  occurs if the end of the subject string is reached successfully, but there are  occurs if the end of the subject string is reached successfully, but there are
1538  not enough subject characters to complete the match. If this happens when  not enough subject characters to complete the match. If this happens when
1539  PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1540  PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues  testing any remaining alternatives. Only if no complete match can be found is
1541  by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL  PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1542  returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that  PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1543  was inspected when the partial match was found is set as the first matching  but only if no complete match can be found.
1544  string. There is a more detailed discussion in the  .P
1545    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1546    partial match is found, \fBpcre_exec()\fP immediately returns
1547    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1548    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1549    important that an alternative complete match.
1550    .P
1551    In both cases, the portion of the string that was inspected when the partial
1552    match was found is set as the first matching string. There is a more detailed
1553    discussion of partial and multi-segment matching, with examples, in the
1554  .\" HREF  .\" HREF
1555  \fBpcrepartial\fP  \fBpcrepartial\fP
1556  .\"  .\"
1557  documentation.  documentation.
1558  .  .
1559    .
1560  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1561  .rs  .rs
1562  .sp  .sp
1563  The subject string is passed to \fBpcre_exec()\fP as a pointer in  The subject string is passed to \fBpcre_exec()\fP as a pointer in
1564  \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset  \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1565  in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of  in \fIstartoffset\fP. If this is negative or greater than the length of the
1566  a UTF-8 character. Unlike the pattern string, the subject may contain binary  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET.
1567  zero bytes. When the starting offset is zero, the search for a match starts at  .P
1568  the beginning of the subject, and this is by far the most common case.  In UTF-8 mode, the byte offset must point to the start of a UTF-8 character (or
1569    the end of the subject). Unlike the pattern string, the subject may contain
1570    binary zero bytes. When the starting offset is zero, the search for a match
1571    starts at the beginning of the subject, and this is by far the most common
1572    case.
1573  .P  .P
1574  A non-zero starting offset is useful when searching for another match in the  A non-zero starting offset is useful when searching for another match in the
1575  same subject by calling \fBpcre_exec()\fP again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
# Line 1525  start of the subject, which is deemed to Line 1589  start of the subject, which is deemed to
1589  set to 4, it finds the second occurrence of "iss" because it is able to look  set to 4, it finds the second occurrence of "iss" because it is able to look
1590  behind the starting point to discover that it is preceded by a letter.  behind the starting point to discover that it is preceded by a letter.
1591  .P  .P
1592    Finding all the matches in a subject is tricky when the pattern can match an
1593    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1594    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1595    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1596    and trying an ordinary match again. There is some code that demonstrates how to
1597    do this in the
1598    .\" HREF
1599    \fBpcredemo\fP
1600    .\"
1601    sample program. In the most general case, you have to check to see if the
1602    newline convention recognizes CRLF as a newline, and if so, and the current
1603    character is CR followed by LF, advance the starting offset by two characters
1604    instead of one.
1605    .P
1606  If a non-zero starting offset is passed when the pattern is anchored, one  If a non-zero starting offset is passed when the pattern is anchored, one
1607  attempt to match at the given offset is made. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1608  pattern does not require the match to be at the start of the subject.  pattern does not require the match to be at the start of the subject.
1609  .  .
1610    .
1611  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1612  .rs  .rs
1613  .sp  .sp
# Line 1711  description above. Line 1790  description above.
1790    PCRE_ERROR_BADNEWLINE     (-23)    PCRE_ERROR_BADNEWLINE     (-23)
1791  .sp  .sp
1792  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1793    .sp
1794      PCRE_ERROR_BADOFFSET      (-24)
1795    .sp
1796    The value of \fIstartoffset\fP was negative or greater than the length of the
1797    subject, that is, the value in \fIlength\fP.
1798  .P  .P
1799  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1800  .  .
# Line 1995  Here is an example of a simple call to \ Line 2079  Here is an example of a simple call to \
2079  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2080  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2081  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2082  PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST,  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2083  and PCRE_DFA_RESTART. All but the last four of these are exactly the same as  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2084  for \fBpcre_exec()\fP, so their description is not repeated here.  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2085    so their description is not repeated here.
2086  .sp  .sp
2087    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
2088    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
# Line 2012  is converted into PCRE_ERROR_PARTIAL if Line 2097  is converted into PCRE_ERROR_PARTIAL if
2097  there have been no complete matches, but there is still at least one matching  there have been no complete matches, but there is still at least one matching
2098  possibility. The portion of the string that was inspected when the longest  possibility. The portion of the string that was inspected when the longest
2099  partial match was found is set as the first matching string in both cases.  partial match was found is set as the first matching string in both cases.
2100    There is a more detailed discussion of partial and multi-segment matching, with
2101    examples, in the
2102    .\" HREF
2103    \fBpcrepartial\fP
2104    .\"
2105    documentation.
2106  .sp  .sp
2107    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2108  .sp  .sp
# Line 2131  Cambridge CB2 3QH, England. Line 2222  Cambridge CB2 3QH, England.
2222  .rs  .rs
2223  .sp  .sp
2224  .nf  .nf
2225  Last updated: 01 June 2010  Last updated: 06 November 2010
2226  Copyright (c) 1997-2010 University of Cambridge.  Copyright (c) 1997-2010 University of Cambridge.
2227  .fi  .fi

Legend:
Removed from v.535  
changed lines
  Added in v.567

  ViewVC Help
Powered by ViewVC 1.1.5