ViewVC logotype

Diff of /code/trunk/NEWS

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

revision 49 by nigel, Sat Feb 24 21:39:33 2007 UTC revision 1470 by ph10, Fri Apr 4 13:41:58 2014 UTC
# Line 1  Line 1 
1  News about PCRE releases  News about PCRE releases
2  ------------------------  ------------------------
4    Release 8.35 04-April-2014
5    --------------------------
7    There have been performance improvements for classes containing non-ASCII
8    characters and the "auto-possessification" feature has been extended. Other
9    minor improvements have been implemented and bugs fixed. There is a new callout
10    feature to enable applications to do detailed stack checks at compile time, to
11    avoid running out of stack for deeply nested parentheses. The JIT compiler has
12    been extended with experimental support for ARM-64, MIPS-64, and PPC-LE.
15    Release 8.34 15-December-2013
16    -----------------------------
18    As well as fixing the inevitable bugs, performance has been improved by
19    refactoring and extending the amount of "auto-possessification" that PCRE does.
20    Other notable changes:
22    .  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
23       an empty string. If it can, pcretest shows this in its information output.
25    .  A back reference to a named subpattern when there is more than one of the
26       same name now checks them in the order in which they appear in the pattern.
27       The first one that is set is used for the reference. Previously only the
28       first one was inspected. This change makes PCRE more compatible with Perl.
30    .  Unicode character properties were updated from Unicode 6.3.0.
32    .  The character VT has been added to the set of characters that match \s and
33       are generally treated as white space, following this same change in Perl
34       5.18. There is now no difference between "Perl space" and "POSIX space".
36    .  Perl has changed its handling of \8 and \9. If there is no previously
37       encountered capturing group of those numbers, they are treated as the
38       literal characters 8 and 9 instead of a binary zero followed by the
39       literals. PCRE now does the same.
41    .  Following Perl, added \o{} to specify codepoints in octal, making it
42       possible to specify values greater than 0777 and also making them
43       unambiguous.
45    .  In UCP mode, \s was not matching two of the characters that Perl matches,
46       namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
47       were matched by \h.
49    .  Add JIT support for the 64 bit TileGX architecture.
51    .  Upgraded the handling of the POSIX classes [:graph:], [:print:], and
52       [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
53       does in Unicode mode.
55    .  Perl no longer allows group names to start with digits, so I have made this
56       change also in PCRE.
58    .  Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
59       mean "start of word" and "end of word", respectively, as a transition aid.
62    Release 8.33 28-May-2013
63    --------------------------
65    A number of bugs are fixed, and some performance improvements have been made.
66    There are also some new features, of which these are the most important:
68    .  The behaviour of the backtracking verbs has been rationalized and
69       documented in more detail.
71    .  JIT now supports callouts and all of the backtracking verbs.
73    .  Unicode validation has been updated in the light of Unicode Corrigendum #9,
74       which points out that "non characters" are not "characters that may not
75       appear in Unicode strings" but rather "characters that are reserved for
76       internal use and have only local meaning".
78    .  (*LIMIT_MATCH=d) and (*LIMIT_RECURSION=d) have been added so that the
79       creator of a pattern can specify lower (but not higher) limits for the
80       matching process.
82    .  The PCRE_NEVER_UTF option is available to prevent pattern-writers from using
83       the (*UTF) feature, as this could be a security issue.
86    Release 8.32 30-November-2012
87    -----------------------------
89    This release fixes a number of bugs, but also has some new features. These are
90    the highlights:
92    .  There is now support for 32-bit character strings and UTF-32. Like the
93       16-bit support, this is done by compiling a separate 32-bit library.
95    .  \X now matches a Unicode extended grapheme cluster.
97    .  Case-independent matching of Unicode characters that have more than one
98       "other case" now makes all three (or more) characters equivalent. This
99       applies, for example, to Greek Sigma, which has two lowercase versions.
101    .  Unicode character properties are updated to Unicode 6.2.0.
103    .  The EBCDIC support, which had decayed, has had a spring clean.
105    .  A number of JIT optimizations have been added, which give faster JIT
106       execution speed. In addition, a new direct interface to JIT execution is
107       available. This bypasses some of the sanity checks of pcre_exec() to give a
108       noticeable speed-up.
110    .  A number of issues in pcregrep have been fixed, making it more compatible
111       with GNU grep. In particular, --exclude and --include (and variants) apply
112       to all files now, not just those obtained from scanning a directory
113       recursively. In Windows environments, the default action for directories is
114       now "skip" instead of "read" (which provokes an error).
116    .  If the --only-matching (-o) option in pcregrep is specified multiple
117       times, each one causes appropriate output. For example, -o1 -o2 outputs the
118       substrings matched by the 1st and 2nd capturing parentheses. A separating
119       string can be specified by --om-separator (default empty).
121    .  When PCRE is built via Autotools using a version of gcc that has the
122       "visibility" feature, it is used to hide internal library functions that are
123       not part of the public API.
126    Release 8.31 06-July-2012
127    -------------------------
129    This is mainly a bug-fixing release, with a small number of developments:
131    . The JIT compiler now supports partial matching and the (*MARK) and
132      (*COMMIT) verbs.
134    . PCRE_INFO_MAXLOOKBEHIND can be used to find the longest lookbehind in a
135      pattern.
137    . There should be a performance improvement when using the heap instead of the
138      stack for recursion.
140    . pcregrep can now be linked with libedit as an alternative to libreadline.
142    . pcregrep now has a --file-list option where the list of files to scan is
143      given as a file.
145    . pcregrep now recognizes binary files and there are related options.
147    . The Unicode tables have been updated to 6.1.0.
149    As always, the full list of changes is in the ChangeLog file.
152    Release 8.30 04-February-2012
153    -----------------------------
155    Release 8.30 introduces a major new feature: support for 16-bit character
156    strings, compiled as a separate library. There are a few changes to the
157    8-bit library, in addition to some bug fixes.
159    . The pcre_info() function, which has been obsolete for over 10 years, has
160      been removed.
162    . When a compiled pattern was saved to a file and later reloaded on a host
163      with different endianness, PCRE used automatically to swap the bytes in some
164      of the data fields. With the advent of the 16-bit library, where more of this
165      swapping is needed, it is no longer done automatically. Instead, the bad
166      endianness is detected and a specific error is given. The user can then call
167      a new function called pcre_pattern_to_host_byte_order() (or an equivalent
168      16-bit function) to do the swap.
170    . In UTF-8 mode, the values 0xd800 to 0xdfff are not legal Unicode
171      code points and are now faulted. (They are the so-called "surrogates"
172      that are reserved for coding high values in UTF-16.)
175    Release 8.21 12-Dec-2011
176    ------------------------
178    This is almost entirely a bug-fix release. The only new feature is the ability
179    to obtain the size of the memory used by the JIT compiler.
182    Release 8.20 21-Oct-2011
183    ------------------------
185    The main change in this release is the inclusion of Zoltan Herczeg's
186    just-in-time compiler support, which can be accessed by building PCRE with
187    --enable-jit. Large performance benefits can be had in many situations. 8.20
188    also fixes an unfortunate bug that was introduced in 8.13 as well as tidying up
189    a number of infelicities and differences from Perl.
192    Release 8.13 16-Aug-2011
193    ------------------------
195    This is mainly a bug-fix release. There has been a lot of internal refactoring.
196    The Unicode tables have been updated. The only new feature in the library is
197    the passing of *MARK information to callouts. Some additions have been made to
198    pcretest to make testing easier and more comprehensive. There is a new option
199    for pcregrep to adjust its internal buffer size.
202    Release 8.12 15-Jan-2011
203    ------------------------
205    This release fixes some bugs in pcregrep, one of which caused the tests to fail
206    on 64-bit big-endian systems. There are no changes to the code of the library.
209    Release 8.11 10-Dec-2010
210    ------------------------
212    A number of bugs in the library and in pcregrep have been fixed. As always, see
213    ChangeLog for details. The following are the non-bug-fix changes:
215    . Added --match-limit and --recursion-limit to pcregrep.
217    . Added an optional parentheses number to the -o and --only-matching options
218      of pcregrep.
220    . Changed the way PCRE_PARTIAL_HARD affects the matching of $, \z, \Z, \b, and
221      \B.
223    . Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
224      bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
226    . Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
227      START_OPTIMIZE option, which is now allowed at compile time
230    Release 8.10 25-Jun-2010
231    ------------------------
233    There are two major additions: support for (*MARK) and friends, and the option
234    PCRE_UCP, which changes the behaviour of \b, \d, \s, and \w (and their
235    opposites) so that they make use of Unicode properties. There are also a number
236    of lesser new features, and several bugs have been fixed. A new option,
237    --line-buffered, has been added to pcregrep, for use when it is connected to
238    pipes.
241    Release 8.02 19-Mar-2010
242    ------------------------
244    Another bug-fix release.
247    Release 8.01 19-Jan-2010
248    ------------------------
250    This is a bug-fix release. Several bugs in the code itself and some bugs and
251    infelicities in the build system have been fixed.
254    Release 8.00 19-Oct-09
255    ----------------------
257    Bugs have been fixed in the library and in pcregrep. There are also some
258    enhancements. Restrictions on patterns used for partial matching have been
259    removed, extra information is given for partial matches, the partial matching
260    process has been improved, and an option to make a partial match override a
261    full match is available. The "study" process has been enhanced by finding a
262    lower bound matching length. Groups with duplicate numbers may now have
263    duplicated names without the use of PCRE_DUPNAMES. However, they may not have
264    different names. The documentation has been revised to reflect these changes.
265    The version number has been expanded to 3 digits as it is clear that the rate
266    of change is not slowing down.
269    Release 7.9 11-Apr-09
270    ---------------------
272    Mostly bugfixes and tidies with just a couple of minor functional additions.
275    Release 7.8 05-Sep-08
276    ---------------------
278    More bug fixes, plus a performance improvement in Unicode character property
279    lookup.
282    Release 7.7 07-May-08
283    ---------------------
285    This is once again mainly a bug-fix release, but there are a couple of new
286    features.
289    Release 7.6 28-Jan-08
290    ---------------------
292    The main reason for having this release so soon after 7.5 is because it fixes a
293    potential buffer overflow problem in pcre_compile() when run in UTF-8 mode. In
294    addition, the CMake configuration files have been brought up to date.
297    Release 7.5 10-Jan-08
298    ---------------------
300    This is mainly a bug-fix release. However the ability to link pcregrep with
301    libz or libbz2 and the ability to link pcretest with libreadline have been
302    added. Also the --line-offsets and --file-offsets options were added to
303    pcregrep.
306    Release 7.4 21-Sep-07
307    ---------------------
309    The only change of specification is the addition of options to control whether
310    \R matches any Unicode line ending (the default) or just CR, LF, and CRLF.
311    Otherwise, the changes are bug fixes and a refactoring to reduce the number of
312    relocations needed in a shared library. There have also been some documentation
313    updates, in particular, some more information about using CMake to build PCRE
314    has been added to the NON-UNIX-USE file.
317    Release 7.3 28-Aug-07
318    ---------------------
320    Most changes are bug fixes. Some that are not:
322    1. There is some support for Perl 5.10's experimental "backtracking control
323       verbs" such as (*PRUNE).
325    2. UTF-8 checking is now as per RFC 3629 instead of RFC 2279; this is more
326       restrictive in the strings it accepts.
328    3. Checking for potential integer overflow has been made more dynamic, and as a
329       consequence there is no longer a hard limit on the size of a subpattern that
330       has a limited repeat count.
332    4. When CRLF is a valid line-ending sequence, pcre_exec() and pcre_dfa_exec()
333       no longer advance by two characters instead of one when an unanchored match
334       fails at CRLF if there are explicit CR or LF matches within the pattern.
335       This gets rid of some anomalous effects that previously occurred.
337    5. Some PCRE-specific settings for varying the newline options at the start of
338       a pattern have been added.
341    Release 7.2 19-Jun-07
342    ---------------------
344    WARNING: saved patterns that were compiled by earlier versions of PCRE must be
345    recompiled for use with 7.2 (necessitated by the addition of \K, \h, \H, \v,
346    and \V).
348    Correction to the notes for 7.1: the note about shared libraries for Windows is
349    wrong. Previously, three libraries were built, but each could function
350    independently. For example, the pcreposix library also included all the
351    functions from the basic pcre library. The change is that the three libraries
352    are no longer independent. They are like the Unix libraries. To use the
353    pcreposix functions, for example, you need to link with both the pcreposix and
354    the basic pcre library.
356    Some more features from Perl 5.10 have been added:
358      (?-n) and (?+n) relative references for recursion and subroutines.
360      (?(-n) and (?(+n) relative references as conditions.
362      \k{name} and \g{name} are synonyms for \k<name>.
364      \K to reset the start of the matched string; for example, (foo)\Kbar
365      matches bar preceded by foo, but only sets bar as the matched string.
367      (?| introduces a group where the capturing parentheses in each alternative
368      start from the same number; for example, (?|(abc)|(xyz)) sets capturing
369      parentheses number 1 in both cases.
371      \h, \H, \v, \V match horizontal and vertical whitespace, respectively.
374    Release 7.1 24-Apr-07
375    ---------------------
377    There is only one new feature in this release: a linebreak setting of
378    PCRE_NEWLINE_ANYCRLF. It is a cut-down version of PCRE_NEWLINE_ANY, which
379    recognizes only CRLF, CR, and LF as linebreaks.
381    A few bugs are fixed (see ChangeLog for details), but the major change is a
382    complete re-implementation of the build system. This now has full Autotools
383    support and so is now "standard" in some sense. It should help with compiling
384    PCRE in a wide variety of environments.
386    NOTE: when building shared libraries for Windows, three dlls are now built,
387    called libpcre, libpcreposix, and libpcrecpp. Previously, everything was
388    included in a single dll.
390    Another important change is that the dftables auxiliary program is no longer
391    compiled and run at "make" time by default. Instead, a default set of character
392    tables (assuming ASCII coding) is used. If you want to use dftables to generate
393    the character tables as previously, add --enable-rebuild-chartables to the
394    "configure" command. You must do this if you are compiling PCRE to run on a
395    system that uses EBCDIC code.
397    There is a discussion about character tables in the README file. The default is
398    not to use dftables so that that there is no problem when cross-compiling.
401    Release 7.0 19-Dec-06
402    ---------------------
404    This release has a new major number because there have been some internal
405    upheavals to facilitate the addition of new optimizations and other facilities,
406    and to make subsequent maintenance and extension easier. Compilation is likely
407    to be a bit slower, but there should be no major effect on runtime performance.
408    Previously compiled patterns are NOT upwards compatible with this release. If
409    you have saved compiled patterns from a previous release, you will have to
410    re-compile them. Important changes that are visible to users are:
412    1. The Unicode property tables have been updated to Unicode 5.0.0, which adds
413       some more scripts.
415    2. The option PCRE_NEWLINE_ANY causes PCRE to recognize any Unicode newline
416       sequence as a newline.
418    3. The \R escape matches a single Unicode newline sequence as a single unit.
420    4. New features that will appear in Perl 5.10 are now in PCRE. These include
421       alternative Perl syntax for named parentheses, and Perl syntax for
422       recursion.
424    5. The C++ wrapper interface has been extended by the addition of a
425       QuoteMeta function and the ability to allow copy construction and
426       assignment.
428    For a complete list of changes, see the ChangeLog file.
431    Release 6.7 04-Jul-06
432    ---------------------
434    The main additions to this release are the ability to use the same name for
435    multiple sets of parentheses, and support for CRLF line endings in both the
436    library and pcregrep (and in pcretest for testing).
438    Thanks to Ian Taylor, the stack usage for many kinds of pattern has been
439    significantly reduced for certain subject strings.
442    Release 6.5 01-Feb-06
443    ---------------------
445    Important changes in this release:
447    1. A number of new features have been added to pcregrep.
449    2. The Unicode property tables have been updated to Unicode 4.1.0, and the
450       supported properties have been extended with script names such as "Arabic",
451       and the derived properties "Any" and "L&". This has necessitated a change to
452       the interal format of compiled patterns. Any saved compiled patterns that
453       use \p or \P must be recompiled.
455    3. The specification of recursion in patterns has been changed so that all
456       recursive subpatterns are automatically treated as atomic groups. Thus, for
457       example, (?R) is treated as if it were (?>(?R)). This is necessary because
458       otherwise there are situations where recursion does not work.
460    See the ChangeLog for a complete list of changes, which include a number of bug
461    fixes and tidies.
464    Release 6.0 07-Jun-05
465    ---------------------
467    The release number has been increased to 6.0 because of the addition of several
468    major new pieces of functionality.
470    A new function, pcre_dfa_exec(), which implements pattern matching using a DFA
471    algorithm, has been added. This has a number of advantages for certain cases,
472    though it does run more slowly, and lacks the ability to capture substrings. On
473    the other hand, it does find all matches, not just the first, and it works
474    better for partial matching. The pcrematching man page discusses the
475    differences.
477    The pcretest program has been enhanced so that it can make use of the new
478    pcre_dfa_exec() matching function and the extra features it provides.
480    The distribution now includes a C++ wrapper library. This is built
481    automatically if a C++ compiler is found. The pcrecpp man page discusses this
482    interface.
484    The code itself has been re-organized into many more files, one for each
485    function, so it no longer requires everything to be linked in when static
486    linkage is used. As a consequence, some internal functions have had to have
487    their names exposed. These functions all have names starting with _pcre_. They
488    are undocumented, and are not intended for use by outside callers.
490    The pcregrep program has been enhanced with new functionality such as
491    multiline-matching and options for output more matching context. See the
492    ChangeLog for a complete list of changes to the library and the utility
493    programs.
496    Release 5.0 13-Sep-04
497    ---------------------
499    The licence under which PCRE is released has been changed to the more
500    conventional "BSD" licence.
502    In the code, some bugs have been fixed, and there are also some major changes
503    in this release (which is why I've increased the number to 5.0). Some changes
504    are internal rearrangements, and some provide a number of new facilities. The
505    new features are:
507    1. There's an "automatic callout" feature that inserts callouts before every
508       item in the regex, and there's a new callout field that gives the position
509       in the pattern - useful for debugging and tracing.
511    2. The extra_data structure can now be used to pass in a set of character
512       tables at exec time. This is useful if compiled regex are saved and re-used
513       at a later time when the tables may not be at the same address. If the
514       default internal tables are used, the pointer saved with the compiled
515       pattern is now set to NULL, which means that you don't need to do anything
516       special unless you are using custom tables.
518    3. It is possible, with some restrictions on the content of the regex, to
519       request "partial" matching. A special return code is given if all of the
520       subject string matched part of the regex. This could be useful for testing
521       an input field as it is being typed.
523    4. There is now some optional support for Unicode character properties, which
524       means that the patterns items such as \p{Lu} and \X can now be used. Only
525       the general category properties are supported. If PCRE is compiled with this
526       support, an additional 90K data structure is include, which increases the
527       size of the library dramatically.
529    5. There is support for saving compiled patterns and re-using them later.
531    6. There is support for running regular expressions that were compiled on a
532       different host with the opposite endianness.
534    7. The pcretest program has been extended to accommodate the new features.
536    The main internal rearrangement is that sequences of literal characters are no
537    longer handled as strings. Instead, each character is handled on its own. This
538    makes some UTF-8 handling easier, and makes the support of partial matching
539    possible. Compiled patterns containing long literal strings will be larger as a
540    result of this change; I hope that performance will not be much affected.
543    Release 4.5 01-Dec-03
544    ---------------------
546    Again mainly a bug-fix and tidying release, with only a couple of new features:
548    1. It's possible now to compile PCRE so that it does not use recursive
549    function calls when matching. Instead it gets memory from the heap. This slows
550    things down, but may be necessary on systems with limited stacks.
552    2. UTF-8 string checking has been tightened to reject overlong sequences and to
553    check that a starting offset points to the start of a character. Failure of the
554    latter returns a new error code: PCRE_ERROR_BADUTF8_OFFSET.
556    3. PCRE can now be compiled for systems that use EBCDIC code.
559    Release 4.4 21-Aug-03
560    ---------------------
562    This is mainly a bug-fix and tidying release. The only new feature is that PCRE
563    checks UTF-8 strings for validity by default. There is an option to suppress
564    this, just in case anybody wants that teeny extra bit of performance.
567    Releases 4.1 - 4.3
568    ------------------
570    Sorry, I forgot about updating the NEWS file for these releases. Please take a
571    look at ChangeLog.
574    Release 4.0 17-Feb-03
575    ---------------------
577    There have been a lot of changes for the 4.0 release, adding additional
578    functionality and mending bugs. Below is a list of the highlights of the new
579    functionality. For full details of these features, please consult the
580    documentation. For a complete list of changes, see the ChangeLog file.
582    1. Support for Perl's \Q...\E escapes.
584    2. "Possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's Java
585    package. They provide some syntactic sugar for simple cases of "atomic
586    grouping".
588    3. Support for the \G assertion. It is true when the current matching position
589    is at the start point of the match.
591    4. A new feature that provides some of the functionality that Perl provides
592    with (?{...}). The facility is termed a "callout". The way it is done in PCRE
593    is for the caller to provide an optional function, by setting pcre_callout to
594    its entry point. To get the function called, the regex must include (?C) at
595    appropriate points.
597    5. Support for recursive calls to individual subpatterns. This makes it really
598    easy to get totally confused.
600    6. Support for named subpatterns. The Python syntax (?P<name>...) is used to
601    name a group.
603    7. Several extensions to UTF-8 support; it is now fairly complete. There is an
604    option for pcregrep to make it operate in UTF-8 mode.
606    8. The single man page has been split into a number of separate man pages.
607    These also give rise to individual HTML pages which are put in a separate
608    directory. There is an index.html page that lists them all. Some hyperlinking
609    between the pages has been installed.
612    Release 3.5 15-Aug-01
613    ---------------------
615    1. The configuring system has been upgraded to use later versions of autoconf
616    and libtool. By default it builds both a shared and a static library if the OS
617    supports it. You can use --disable-shared or --disable-static on the configure
618    command if you want only one of them.
620    2. The pcretest utility is now installed along with pcregrep because it is
621    useful for users (to test regexs) and by doing this, it automatically gets
622    relinked by libtool. The documentation has been turned into a man page, so
623    there are now .1, .txt, and .html versions in /doc.
625    3. Upgrades to pcregrep:
626       (i)   Added long-form option names like gnu grep.
627       (ii)  Added --help to list all options with an explanatory phrase.
628       (iii) Added -r, --recursive to recurse into sub-directories.
629       (iv)  Added -f, --file to read patterns from a file.
631    4. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
632    script, to force use of CR or LF instead of \n in the source. On non-Unix
633    systems, the value can be set in config.h.
635    5. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
636    absolute limit. Changed the text of the error message to make this clear, and
637    likewise updated the man page.
639    6. The limit of 99 on the number of capturing subpatterns has been removed.
640    The new limit is 65535, which I hope will not be a "real" limit.
643  Release 3.3 01-Aug-00  Release 3.3 01-Aug-00
644  ---------------------  ---------------------
# Line 34  some of the new functionality in Perl 5. Line 673  some of the new functionality in Perl 5.
673  Another (I hope this is the last!) change has been made to the API for the  Another (I hope this is the last!) change has been made to the API for the
674  pcre_compile() function. An additional argument has been added to make it  pcre_compile() function. An additional argument has been added to make it
675  possible to pass over a pointer to character tables built in the current  possible to pass over a pointer to character tables built in the current
676  locale by pcre_maketables(). To use the default tables, this new arguement  locale by pcre_maketables(). To use the default tables, this new argument
677  should be passed as NULL.  should be passed as NULL.

Removed from v.49  
changed lines
  Added in v.1470

  ViewVC Help
Powered by ViewVC 1.1.5