summaryrefslogtreecommitdiff
path: root/t
Commit message (Collapse)AuthorAgeFilesLines
* Replace XXX by correct bug numberKarl Williamson2013-01-191-1/+1
| | | | | Commit 4d68ffa0f7f345bc1ae6751744518ba4bc3859bd failed to get the correct bug number in a comment
* Deprecate certain rare uses of backslashes within regexesKarl Williamson2013-01-192-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are three pairs of characters that Perl recognizes as metacharacters in regular expression patterns: {}, [], and (). These can be used as well to delimit patterns, as in: m{foo} s(foo)(bar) Since they are metacharacters, they have special meaning to regular expression patterns, and it turns out that you can't turn off that special meaning by the normal means of preceding them with a backslash, if you use them, paired, within a pattern delimitted by them. For example, in m{foo\{1,3\}} the backslashes do not change the behavior, and this matches "f", "o" followed by one to three more occurrences of "o". Usages like this, where they are interpreted as metacharacters, are exceedingly rare; we think there are none, for example, in all of CPAN. Hence, this deprecation should affect very little code. It does give notice, however, that any such code needs to change, which will in turn allow us to change the behavior in future Perl versions so that the backslashes do have an effect, and without fear that we are silently breaking any existing code. =head1 Performance Enhancements
* Revert "Deprecate literal unescaped "{" in regexes."Karl Williamson2013-01-192-8/+3
| | | | | | | | | This reverts commit 2a53d3314d380af5ab5283758219417c6dfa36e9. Not the entire commit was reverted, but the deprecation message is gone. This caused too many problems. See thread http://www.nntp.perl.org/group/perl.perl5.porters/2012/11/msg195425.html (which lists previous threads).
* Extend strictness for qr/(?[ \N{} ])/Karl Williamson2013-01-193-0/+7
| | | | | | | | | | This recently added regex syntax imposes stricter rules on parsing than normal. However, this did not include parsing \N{} constructs that occur within it. This commit does that, making fatal the warnings that come from \N{} I will add to perldiag the newly added messages along with the others for (?[ ]) before 5.18 ships
* Skip t/io/eintr.t on older Darwins.James E Keenan2013-01-181-2/+4
| | | | | | | Identify the OS version by capturing the first two parts of the M.m.p version number. For RT #116262
* further refinement to EISDIR tests from `perl dir`Ricardo Signes2013-01-181-3/+2
| | | | | | | | 1. actually use the EISDIR string, rather than getting it and not using it; this was a refactoring screw-up 2. don't hardcode the Win32 EACCES error, either, use the same "$!" mechanism
* in testing EISDIR string, check known value firstRicardo Signes2013-01-161-5/+6
| | | | | | | It was not enough to ensure the English value, as some platforms use a different string entirely. Rather than goof around with figuring them out, just get the known value by making an EISDIR and stringifying it, then compare to that.
* Remove the Rhapsody port.Nicholas Clark2013-01-161-2/+1
| | | | | | | Rhapsody was an Apple OS that later evolved into Darwin and Mac OS X. It was initially only released to developers, but later became Mac OS X Server, with releases in 1999 and 2000. It was obsoleted by Mac OS X 10.0, released in March 2001.
* avoid having to worry whether the test runs in a localeRicardo Signes2013-01-151-0/+4
| | | | | | This is a time-honored tradition from such places as t/op. Tony Cook alerted me to failures caused by this test on machines smoking in non-English locales.
* Add warnings for "\08", /\017/Karl Williamson2013-01-141-0/+4
| | | | | | | | | | | | | This was discussed in thread http://perl.markmail.org/thread/avtzvtpzemvg2ki2 but I never got around to this portion of the consensus, until now. I did a cpan grep http://grep.cpan.me/?q=%28^|[^\\]%29\\[0-7]{1%2C2}[8-9]&page=1 and eyeballing the results, saw three cases where this warning might show up; one of which was for EBCDIC. The others looked to be false positives, such as in .css files.
* reg_mesg.t: Add, revise test namesKarl Williamson2013-01-131-3/+3
|
* porting/args_assert.t: Needs to look in inline.h tooKarl Williamson2013-01-121-0/+2
| | | | | inline.h is a special header file that contains C functions, and hence perhaps PERL_ARGS_ASSERTS.
* expect a different error for `perl dir` on Win32Ricardo Signes2013-01-121-1/+5
| | | | | | | We've known that this is how Win32 behaves, as it was documented in the ticket for which this is a fix. I don't think it's worth the bother of ensuring we get EISDIR, as long as we don't just exit silently!
* Add a test to detect error when attempting to syntax-check a directory.James E Keenan2013-01-121-1/+12
| | | | For: RT #61362
* Add back-compat (?[ ]) testsKarl Williamson2013-01-113-21/+237
| | | | | | | | | | This adds testing of (?[ ]), using the same tests, t/re/re_tests< as are used by many of the regular expression .t files. Basically, it converts the [bracketed] character classes in these tests to the new syntax and verifies that they work there. Some tests won't work in one or the other, and the capability to skip depending on the .t is added
* reg_mesg.t: Test messages from (?[ ])Karl Williamson2013-01-111-0/+37
|
* New regex experimental feature: (?[ ])Karl Williamson2013-01-112-3/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a fancier [bracketed] character class which allows set operations, such as intersection and subtraction. The entry in perlre for this commit details its operation. Besides extending regular expressions to handle this functionality, recommended by Unicode, the intent here is to do three things: 1) Intersection has been simulated by regexes using zero-width look-around assertions, which are non-obvious. This allows replacing those with a more powerful and clearer syntax; the compiled regexes are smaller and faster. Everything is known at compile time. 2) Set operations have also been simulated by using user-defined Unicode properties. These are globals, have security implications, restricted names, and d don't allow as complex expressions as this new feature. 3) I hope that this feature will come to be viewed as a "better" bracketed character class. I took advantage of the fact that there is no embedded base to have to be compatibile with to forbid certain iffy practices with the existing ones, while remaining mostly backwards compatible. The main difference is that /x is always enabled, so white space can be pretty much freely used with these, but to specify a match on white space, it must be escaped. Things that should have been illegal are, such as \x{}, and \x{abcdefghi}. Things that look like a posix specifier but don't quite meet the rules now give an error instead of silently compiling. e.g., [:digit] is an error instead of the union of the characters that compose it. I may have omitted things; perhaps it should be an error to have the same letter occur twice, adjacent. Since this is experimental, we can make such changes based on field feed back. The intent is to keep this feature, since it is strongly recommended by Unicode. The exact syntax is subject to change, so is experimental.
* regcomp.c: Add capability for regclass() to return inversion listKarl Williamson2013-01-111-0/+1
| | | | | This is currently unused, but will have regclass() return an inversion list instead of a node.
* regcomp.c: Add capability for stricter parsing of []Karl Williamson2013-01-111-0/+3
| | | | | | | | | | This adds the capability, currently unused, of forbidding certain things in [bracketed] character classes. Included are things that warn bug still compile, such as false ranges, [\d-\w], and unrecognized escapes. Also forbidden are potentially ambiguous cases where \x (without braces) isn't followed by exactly two hex digits, or \000 where the number of octal digits isn't precisely three.
* regcomp.c: Add capability for strict [:posix:]Karl Williamson2013-01-111-0/+2
| | | | | This adds a parameter to regpposixcc() to enforce stricter rules on the posix class syntax. It is currently unused
* Add optional strict mode to grok_bslash_[xo]Karl Williamson2013-01-111-0/+3
| | | | | | This mode croaks on any iffy constructs that currently compile. It is not currently used; documentation of the error messages will be delivered later.
* Better error pos for grok_bslash_[xo]Karl Williamson2013-01-111-2/+6
| | | | | | | | | These functions advance the parse pointer for the caller. The regex code has the infrastructure to output a marker as to where the error was. This commit simply moves the parse pointer past all the legal digits in the input, which are likely supposed to be part of the number, which makes it likely that the missing right brace point is just past those.
* regexp.t: Add a period in test name skip reasonKarl Williamson2013-01-111-1/+1
| | | | This is easier to read.
* lib/warnings/regcomp: Add note about better place for theseKarl Williamson2013-01-111-0/+2
| | | | | reg_mesg.t has some more infrastructure, so it is probably easier to add new tests there.
* t/re/re_tests: Remove duplicate testKarl Williamson2013-01-081-1/+0
| | | | This is identical to the test two lines above.
* t/re/re_tests: Remove unused variable initializationsKarl Williamson2013-01-081-32/+32
| | | | | | | These lines all fail to compile, so matching doesn't happen, so it doesn't matter at all what the target string to be matched against is set to. It is misleading to put apparently meaningful stuff in that string.
* regexp.t: Skip tests that are supposed toKarl Williamson2013-01-081-5/+5
| | | | | | This reorders some if elsif ... blocks so that skip is tested for and done before actually trying the test. This only affected tests which were supposed to generate compiler errors.
* skip appropriately when XS::APItest isn't availableTony Cook2013-01-081-1/+3
| | | | XS::APItest isn't available under -Uusedl
* t/op/not.t: Add descriptions to all tests.James E Keenan2013-01-051-16/+31
| | | | | Also, add note() before tests 4 and 5 explaining rationale for addition of parentheses to second arguments.
* Add parens around second argument to tests 4 and 5.James E Keenan2013-01-051-2/+2
| | | | | | | | If a description were to be added to these tests, in the absence of parentheses the scalar prototype of CORE::not would enforce a scalar context onto the balance of the statement, leading to apparently anomalous behavior, viz., the descriptions would not be printed and test 5 would be reported to FAIL.
* t/op/oct.t: Add descriptions to tests lacking them.James E Keenan2013-01-051-15/+19
|
* t/op/or.t: Provide descriptions for tests lacking them.James E Keenan2013-01-051-5/+5
|
* t/op/pos.t: Provide descriptions for remaining tests lacking them.James E Keenan2013-01-051-7/+7
|
* t/op/pwent.t: Add descriptions for all tests.James E Keenan2013-01-051-2/+3
|
* t/op/pow.t: Add descriptions for tests lacking them.James E Keenan2013-01-051-5/+5
| | | | | Those which were still lacking descriptions were all testing that exponentiation has precedence over negation.
* t/op/fork.t: Add description for one special test.James E Keenan2013-01-051-1/+1
|
* charnames: Deprecate character names with spacing issuesKarl Williamson2013-01-042-0/+46
| | | | | | | A user-defined character name with trailing or multiple spaces in a row is likely a typo, and hence likely won't match what the other uses of it. These names also won't work if we extend :loose to these. This now generates a warning.
* charnames: :alias alone implies :fullKarl Williamson2013-01-041-8/+0
| | | | | | | | | The documentation says this is how it should behave, but only 1 of the three paths in the code did it, and in fact there was a test to the contrary. I'm only adding a test for one of the two fixed paths, as the other one appears to require a weird file name.
* charnames: comment, error message clarificationsKarl Williamson2013-01-041-1/+1
|
* make m?$pat? match only once under ithreadsDavid Mitchell2013-01-041-1/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [perl #115080] m?...? is only supposed to match once, until reset. Normally this is done by setting the PMf_USED flag on the PMOP. Under ithreads we can't modify ops, so instead we indicate by setting the regex's SV to readonly. (This is a bit of a hack: the flag should be associated with the PMOP, not the regex). This breaks with run-time regexes when the pattern gets recompiled; for example: for my $c (qw(a b c)) { print "matched $c\n" if $c =~ m?^$c$?; } outputs matched a on unthreaded, but matched a matched b matched c on threaded. The re_eval jumbo fix made this more noticeable by sometimes recompiling even when the pattern text hasn't changed (to make closures work ok). The quick fix is to propagate the readonlyness of the old re to the new re. (The proper fix would be to store the flag state in a pad slot associated with the PMOP). Needless to say, I've gone for the quick fix.
* Regenerated customized.dat file due to pod2*.PL changesChris 'BinGOs' Williams2013-01-041-68/+64
|
* Fix period at the end of perldiag.pod and remove entry from TODO in diag.tMatthew Horsfall (alh)2013-01-031-1/+0
|
* regcomp.c: Refactor out a switch statementKarl Williamson2012-12-311-1/+0
| | | | | | | Previous work had collapsed most of the cases of this switch. This removes the entire switch, allowing more of the logic to be collapsed into single code paths. Most of this commit is just moving things around; the heavy lifting has been done in previous commits.
* locale.t, pat_advanced.t: Remove TODOsKarl Williamson2012-12-311-1/+0
| | | | The previous commit fixed these TODOs.
* re/pat_advanced.t: Add TODO test for recent regressionKarl Williamson2012-12-311-0/+8
| | | | | | | | | Commit 3018b823898645e44b8c37c70ac5c6302b031381 added a regression for the Posix classes [:upper:] and [:lower:] when matching case-insensitively. If an above-Latin1 code point has been matched by one of these classes at the time another regex is compiled which also has the same class as the first one, and the second regex is /i, the case-insensitivity is ignored.
* Correct typo in test description.James E Keenan2012-12-301-1/+1
|
* regcomp.c: Clean up ANYOF_CLASS handling.Karl Williamson2012-12-281-0/+2
| | | | | | | | | | | The ANYOF_CLASS flag is used in ANYOF nodes (for [bracketed] and the synthetic start class) only when matching something like \w, [:punct:] etc., under /l (locale). It should not be set unless /l is specified. However, it was always getting set for the synthetic start class. This commit fixes that. The previous code was masking errors in which it was being tested for unnecessarily, and for much of the 5.17 series, the synthetic start class was always set to test for locale, which was a waste of cpu when no locale was specified.
* Add regcomp.c warning checks to t/porting/diag.t.Matthew Horsfall (alh)2012-12-271-7/+35
| | | | | | * Support regcomp.c ckWARN and vWARN macros * Update pod/perldiag.pod for fixes discovered with new checks * Allow t/porting/diag.t to match printfs with flags more liberally
* pat_advanced.t: Add testsKarl Williamson2012-12-271-0/+4
| | | | I almost broke this, so adding a precautionary test.
* Revert "porting/podcheck corrections."David Mitchell2012-12-261-3/+1
| | | | | | | | | This reverts commit f6a6501216dee24e251d4482bd3a1f6daf4ac0da. The fix seems wrong and is causing podcheck.t test failures, and (for my system at least), reverting it removes those errors and doesn't create new errors. Whatever was originally causing podcheck errors needs to be fixed, rather than trying to mask it.
View Lorry Controller Status