This week on the Perl 6 mailing lists
Remember, the next Parrot bug day is 13 January, 2007. Join the rest of the team at #parrot (irc.perl.org) to work on closing as many bugs as possible before the next Parrot release.
Language
Earlier, Luke Palmer asked for a clear definition of when math should use floating points and when it should be integer-based. In response, Darren Duncan highlighted a recent #perl6 discussion on the topic. He proposed distinct operators to allow users to explicitly choose the math mode. Jonathan Lang mentioned that in the one other instance where there were type-specific versions of a common operator, a third 'generic' version was required. Dr. Ruud suggested giving numerics multiple faces.
More recently, Larry Wall responded to the thread, adding his thoughts
and musing on whether standard Perl should include convenience operators
like idiv or if there should simply be pragmatic support for them.
A great deal of discussion followed.
Parrot Porters
[perl #41157] [PATCH] Fix a typo in t/op/cmp-nonbranch.t
In ticket [perl #41157], Lee Duhem supplied a patch to correct a typographical error.
[perl #41158] [BUG] Here Docs in test C<cmp> cause t/op/cmp-nonbranch.t abnormal exit
Lee Duhem reported an error in a test; this was done in ticket [perl #41158]. Jerry Gay reported that he believed the issue was resolved in r16381 and asked Lee to check it. Lee reported that he thought it was a bug, resolved in ticket [perl #41165].
Allison Randal announced that she had modified Punie to work with PAST-pm. She also gave her thoughts on working with PAST-pm, which Patrick R. Michaud replied to.
[perl #41163] [PATCH] suppress uninitialized value warning in config/inter/yacc.pm
In ticket [perl #41163], Lee Duhem submitted a patch to suppress some uninitialized value warnings. Jonathan Worthington had a question about the patch, and Lee agreed with the suggestion.
[perl #41164] [BUG] 'make world' fails with msvc due to unresolved external
Jerry Gay created ticket [perl #41164] to report a bug, which was resolved in r16383.
[perl #41165] [PATCH] appropriate handle CRLE in Here Docs
Ticket [perl #41165] included a patch from Lee Duhem to correct a problem he reported earlier in '[perl #41158] [BUG] Here Docs in test C cause t/op/cmp-nonbranch.t abnormal exit '.
[svn:parrot-pdd] r16391 - trunk/docs/pdds/clip
With this commit, Allison Randal moved the I/O PDD out of the clip directory, making it more official. Discussion continued on 'I/O PDD - ready for implementation'.
I/O PDD - ready for implementation
After moving the I/O PDD out of the clip directory (see '[svn:parrot-pdd] r16391 - trunk/docs/pdds/clip'), Allison Randal announced that implementation can begin on the details in the document. She also welcomed comments.
Jonathan Worthington replied that he liked what he had read, but felt that more information was needed on how role composition should work. Allison replied that she was still at work on defining roles in the objects PDD. She offered a short explanation of how it would probably work.
Some discussion on how to implement error values also took place, with Jonathan, Allison and Larry Wall contributing.
[perl #41168] graceful "no compiler" error message?
In ticket [perl #41168], Will Coleda added a request for a 'no
compiler found' error message in configure.pl.
[perl #41171] [PATCH] Fix to t/src/compiler.t so Parrot passes tests on Solaris
Steve Peters submitted a patch as ticket [perl #41171].
Re: [perl #41020] [PATCH] pmc2c.pl functionality extracted into separate package
Earlier James Keenan supplied some patches. After some discussion, the patches were recently applied as r16345.
However, James felt there were still some problems with the patches. He provided a new patch in the hopes that it would solve the problem.
[NEW]: README for t/tools/pmc2cutils/
James Keenan submitted a README file for t/tools/pmc2utils, which
was applied as r16409.
[perl #41173] [PATCH] Intel C++ is not really gcc
Steve Peters created ticket [perl #41173] to report a problem
caused by the Intel C++ compiler defining __GNUC__ although it is
not compatible with gcc. He provided a patch which attempts to address
the issue. It was applied in r16412.
[perl #41174] [PATCH] Silence warnings in
In ticket [perl #41174], Steve Peters contributed a patch to silence warnings in the Intel C++ compiler when 0 is used as an enum without being cast. It was applied in r16414.
[perl #41175] [PATCH] Remove unused label
In ticket [perl #41175], Steve Peters supplied a patch to remove
an unused label in src/inter_call.c. It was applied in r16415.
[perl #41180] [PATCH] Determine compiler before processing hints
Steve Peters noted in ticket [perl #41180] that sometimes the compiler should be known before processing the hints. He supplied a patch to correct the issue, which was applied as r16416.
[perl #41182] [PATCH] Preliminary support for suncc on Linux
Steve Peters created a patch to add initial support for compiling with suncc on Linux. This was submitted as ticket [perl #41182]. It was committed as r16417.
[perl #41185] [BUG]: examples/shootout/regexdna.pir.input: Persistently failing test
In ticket [perl #41185], James Keenan reported a problem with a
failing test in shootout.t on Mac OSX.
[perl #41186] [BUG]: tools/build/ops2pm.pl: Variable used in both package scope and loop iteration
Ticket [perl #41186] was created by James Keenan to report a possible
bug in ops2pm.pl. He wondered if it was intentional that a variable was
used in two different scopes. His feeling was that using the same variable
name would confuse someone.
[perl #41187] [PATCH] fix failed test t/examples/past.t
Lee Duhem reported that r16418 broke a test. He supplied a patch, but felt that there might be a better solution. Ticket [perl #41187] follows this report.
Users
gather { if $xy.was_taken { ... } }
Earlier, Gilbert R. Röhrbein asked if there is a way to look inside a gather-block if something was already taken. He included some example code of how he thought it would work.
Gabriele Renzi replied with a code correction, but noted that it would not work on Pugs 6.2.13.
99 problems in Perl6: problem 28
Gabriele Renzi decided to look at problem #28 from Ovid's list of 99 problems in Perl 6. Gabriele included a solution, but wondered if there was a better way to solve the exercise. Ovid suggested checking in the code, if it passed tests. Gaal Yahas provided the commit bit.
99 problems in Perl6: 32 and a question on number coercion
Gabriele Renzi provided an answer to problem number #32 from the list of
99 problems in Perl 6 and added it to the repository. Gabriele wondered
what the expected behavior was for gcd. Steffen Schwigon had some
questions on the code, which Gabriele answered.
Steffen Schwigon
perl6-ish idiom for code and tests in one file
Gabriele Renzi inquired what the Perl 6 idiom would be for keeping code
and tests in the same file. Juerd liked the idea of a macro or a standard
subroutine, as did Gaal Yahas. Larry Wall offered an alternative using
MAIN, but noted that autocalls to MAIN are not yet implemented. He
suggested a temporary workaround.
Compiler
[PATCH] fix wrong $include_path (of parrot) in Makefile.PL
Lee Duhem submitted a patch to fix an incorrect Parrot include path. He noted that he is still seeing an error.
Acknowlegements
This summary was prepared using Mail::Summary::Tools, available on CPAN.
If you appreciate Perl, consider contributing to the Perl Foundation to help support the development of Perl.
Thank you to everyone who has pointed out mistakes and offered suggestions for improving this series. Comments on this summary can be sent to Ann Barcomb, kudra@domaintje.com.
I apologize for the lateness and brevity of this summary, which are due to ongoing computer problems.
Distribution
This summary can be found in the following places:
See Also
