gmane.comp.lang.perl.perl5.porters

RFC: Unicode::UCD::all_casefolds()

Karl Williamson — 2012-05-24T18:42:34

Unicode::UCD::casefold() is a function which takes a code point argument 
and returns a hash of information about its casefold.  The first time 
the function is called, it constructs and caches a hash which contains a 
sub-hash for every code point that has an "interesting" casefold (there 
are only somewhat over a thousand such code points).

I have found it useful in writing .t tests to have access to the entire 
data structure, so I've written a minimal function to return that, and 
have tentatively called it all_casefolds().

My question is: "Would this be a function that others would conceivably 
want to use, and hence should be documented"?  If so, is there a better 
name?

Re: [perl #113006] perllocale Spanish old reference

Tom Christiansen — 2012-05-24T16:54:21

This is all certainly true.

However, it is also something that Perl has no control over. 
Do *not* use vendor locales if you want guaranteed behaviour. 

    % perl -MUnicode::Collate::Locale -E 'say for Unicode::Collate::Locale->new(locale => "es")->sort(qw(color chocolate))'
    chocolate
    color

    % perl -MUnicode::Collate::Locale -E 'say for Unicode::Collate::Locale->new(locale => "es__traditional")->sort(qw(color chocolate))'
    color
    chocolate

*That* is guaranteed behaviour.  The rest, isn't.

--tom

RE: Perl 5.16.0 RC0

Steve Hay — 2012-05-24T16:47:57

Tony Cook wrote on 2012-05-24:
PAUSE.
somewhere.
else

Yes, I found originally that the full test suite hangs (originally in
CPANPLUS tests and then again in Module-Build tests, but now seemingly
only in the latter), but when you try to look harder at individual hangs
then they don't hang any more :-(



I can reproduce the same result with the nmake command, which is useful
to avoid having to run the *whole* test suite every time, but the perl
harness command you've given runs without hanging for me :-/

I currently get three hanging tests in Module-Build (I'm ignoring here
the fork.t script which still intermittently hangs too): runthrough.t,
test_types.t and xs.t, and they all have the same call stack when
breaking into them when hung:

[Frames below may be incorrect and/or missing, no symbols loaded
for ntdll.dll]
 ntdll.dll!779bf901() 
 mswsock.dll!74c06e67() 
 mswsock.dll!74bfe8d0() 
 mswsock.dll!74bf7d8c() 
 ws2_32.dll!754d5a7a() 
 ws2_32.dll!754d5fe8() 
 ws2_32.dll!754c394b() 
 perl517.dll!win32_close(int fd)  Line 3065 + 0x9 bytesC
 perl517.dll!PerlLIOClose(IPerlLIO * piPerl, int handle)  Line
948 + 0x9 bytesC++
 perl517.dll!PerlIOUnix_close(interpreter * my_perl, _PerlIO * *
f)  Line 2871 + 0x1c bytesC
 perl517.dll!PerlIOBase_close(interpreter * my_perl, _PerlIO * *
f)  Line 2216 + 0x10 bytesC
 perl517.dll!PerlIOBuf_close(interpreter * my_perl, _PerlIO * *
f)  Line 4235 + 0xd bytesC
 perl517.dll!PerlIO__close(interpreter * my_perl, _PerlIO * * f)
Line 1474 + 0x10 bytesC
 perl517.dll!Perl_PerlIO_close(interpreter * my_perl, _PerlIO * *
f)  Line 1487 + 0xd bytesC
 perl517.dll!Perl_do_openn(interpreter * my_perl, gv * gv, const
char * oname, long len, int as_raw, int rawmode, int rawperm, _PerlIO *
* supplied_fp, sv * * svp, long num_svs)  Line 664 + 0xd bytesC
 perl517.dll!Perl_pp_open(interpreter * my_perl)  Line 625 + 0x2e
bytesC
 perl517.dll!Perl_runops_debug(interpreter * my_perl)  Line 2118
+ 0xf bytesC
 perl517.dll!S_run_body(interpreter * my_perl, long oldscope)
Line 2397 + 0xf bytesC
 perl517.dll!perl_run(interpreter * my_perl)  Line 2320 + 0xd
bytesC
 perl517.dll!RunPerl(int argc, char * * argv, char * * env)  Line
270 + 0x9 bytesC++
 perl.exe!main(int argc, char * * argv, char * * env)  Line 23 +
0x12 bytesC
 perl.exe!__tmainCRTStartup()  Line 555 + 0x17 bytesC
 kernel32.dll!7604339a() 
 ntdll.dll!779d9ef2() 
 ntdll.dll!779d9ec5() 
 ntdll.dll!779ce38c()

Re: [perl #113006] perllocale Spanish old reference

Joaquín Ferrero — 2012-05-24T16:08:19

El 24/05/12 02:53, James E Keenan via RT escribió:

Hi, Jim.

  From the Real Academia Española (Royal Spanish Academy) :

1. Exclusión de los dígrafos ch y ll del abecedario (removing of 'ch' and 'll' digraphs from alphabet)

http://www.rae.es/rae/gestores/gespub000018.nsf/%28voAnexos%29/arch8100821B76809110C12571B80038BA4A/$File/CuestionesparaelFAQdeconsultas.htm#novOrto1

And, yes: these changes are worldwide.

Other link: http://en.wikipedia.org/wiki/Spanish_orthography#Alphabet

Re: RFC: Proposal to change the editor hints in core files to expand tabs

demerphq — 2012-05-24T15:18:57

Glad to hear that.

But I recommend that you do this as one big patch and NOT on a per file basis.

Otherwise people will have to use -w on pretty well every diff. If we
do it as a one off then the -w is needed only for that diff.

I also really do not like the idea of mixing whitespace changes with
real changes. It makes merging and rebasing extremely difficult, and
it makes code history diving much more difficult.

cheers,
Yves

Re: RFC: Proposal to change the editor hints in core files to expand tabs

Ricardo Signes — 2012-05-24T15:10:51

* Karl Williamson <public< at >khwilliamson.com> [2012-05-20T23:34:36]

I am all for it.

Re: [PATCH] perlapi.pod clarification patchlet: newSVpvn operates on buffers - not strings

Karl Williamson — 2012-05-24T14:56:14

It looks like I was confusing this with other newSVpvn_foo functions. 
Shouldn't they all be changed correspondingly?

Re: [perl #112796] The reentrant API should always have prototypes

Craig A. Berry — 2012-05-24T14:42:40

On May 24, 2012, at 2:49 AM, Father Chrysostomos via RT wrote:



Done, thanks for the reminder.  Please close the ticket.

________________________________________
Craig A. Berry
mailto:craigberry< at >mac.com

"... getting out of a sonnet is much more
 difficult than getting in."
                 Brad Leithauser

Re: [perl #112784] mathoms broken under C++

Craig A. Berry — 2012-05-24T14:42:21

On May 24, 2012, at 2:48 AM, Father Chrysostomos via RT wrote:



Done, thanks for the reminder.  Please close the ticket.

________________________________________
Craig A. Berry
mailto:craigberry< at >mac.com

"... getting out of a sonnet is much more
 difficult than getting in."
                 Brad Leithauser

Smoke [blead] v5.16.0-209-gef5eb41 FAIL(F) MSWin32 Win2000 SP4 (x86/1 cpu)

George Greer — 2012-05-24T12:48:00

Smoke logs available at http://m-l.org/~perl/smoke/perl/

Note: VirtualBox clock issues have been apparently fixed by host reboot/upgrade.

Automated smoke report for 5.17.0 patch ef5eb418bf9cb919bb2d82393f4570a790061e2e v5.16.0-209-gef5eb41
perl-win2k: Intel(R) Core(TM) i7 CPU 920 < at > 2.67GHz(~2662 MHz) (x86/1 cpu)
    on        MSWin32 - Win2000 SP4
    using     cl version 14.00.50727.762
    smoketime 4 hours 20 minutes (average 1 hour 5 minutes)

Summary: FAIL(F)

O = OK  F = Failure(s), extended report at the bottom
X = Failure(s) under TEST but not under harness
? = still running or test results not (yet) available
Build failures during:       - = unknown or N/A
c = Configure, m = make, M = make (after miniperl), t = make test-prep

v5.16.0-209-gef5eb41  Configuration (common) none
----------- ---------------------------------------------------------
O O         
O F         -Duseithreads
| +--------- -DDEBUGGING
+----------- no debugging


Locally applied patches:
    uncommitted-changes
    SMOKEef5eb418bf9cb919bb2d82393f4570a790061e2e

Failures: (common-args) none
[default] -DDEBUGGING -Duseithreads
../t/op/threads.t...........................................FAILED
    Non-zero exit status: 9
    Bad plan.  You planned 26 tests but ran 9.

Passed Todo tests: (common-args) none
[default] 
[default] -DDEBUGGING
[default] -Duseithreads
[default] -DDEBUGGING -Duseithreads
../cpan/CGI/t/tmpdir.t......................................PASSED
    3-9

Compiler messages(MSWin32):
op.c(6531) : warning C4244: 'initializing' : conversion from 'U32' to 'const char', possible loss of data
op.c(10633) : warning C4028: formal parameter 1 different from declaration
..\pp.c(5980) : warning C4244: 'function' : conversion from 'U32' to 'const char', possible loss of data
..\pp_hot.c(2685) : warning C4244: 'initializing' : conversion from 'I32' to 'U8', possible loss of data
..\pp_sort.c(1668) : warning C4244: 'initializing' : conversion from 'I32' to 'U8', possible loss of data
..\toke.c(765) : warning C4244: '=' : conversion from 'U32' to 'U8', possible loss of data
IO.xs(488) : warning C4101: 'ret' : unreferenced local variable
ListUtil.xs(297) : warning C4244: 'initializing' : conversion from 'I32' to 'U8', possible loss of data
ListUtil.xs(335) : warning C4244: 'initializing' : conversion from 'I32' to 'U8', possible loss of data
Base64.xs(394) : warning C4146: unary minus operator applied to unsigned type, result still unsigned
Base64.xs(396) : warning C4146: unary minus operator applied to unsigned type, result still unsigned
POSIX.xs(705) : warning C4101: 'charstring' : unreferenced local variable
scalar.xs(175) : warning C4018: '>' : signed/unsigned mismatch
Piece.xs(950) : warning C4101: 'ret' : unreferenced local variable
APItest.xs(3046) : warning C4244: 'initializing' : conversion from 'I32' to 'U8', possible loss of data
APItest.xs(3358) : warning C4101: 'tmp' : unreferenced local variable
op.c(10633) : warning C4028: formal parameter 2 different from declaration

Re: Removing from the opcode tree

demerphq — 2012-05-24T10:59:14

As soon as they materialize could I have a couple too?

Yves

Re: Perl 5.16.0 RC0

Tony Cook — 2012-05-24T10:09:09

Another strangeness in the behaviour here:

  nmake test

hangs in various Module-Build tests.

But:

  nmake test TEST_FILES=..\cpan\Module-Build\t\*.t

runs the Module-Build tests without a problem.

  nmake test TEST_FILES="base/*.t ../cpan/Module-Build/t/*.t"

hangs in perl_mb_opt.t, as does:

  cd ../t
  .\perl harness base/cond.t ../cpan/Module-Build/t/perl_mb_opt.t

Tony

Re: Perl secret operators manual page

Aristotle Pagaltzis — 2012-05-24T09:17:22

* Philippe Bruhat (BooK) <philippe.bruhat< at >free.fr> [2012-05-24 02:30]:

Yes please.

However mistaken the anthropological interpretation of the interaction
after which the operator was named – the Rorschach interpretation of the
inverted curly bracket as a picture of two faces opposite one another
touching noses is perfect, once it has been pointed it is very difficult
not to see it every time you look at the operator.

And the name has successfully me(me)tastasised for that reason. So on
descriptivist grounds the “wrong” name has to be mentioned. “Hourglass”
is an apt interpretation of the visual also, but it can only be listed
as an alternative.

I see no reason not to add a note to point out the anthropological error
however. It might even contribute just a little to a wider awareness.

Regards,

[perl #112786] build broken under clang++

Father Chrysostomos via RT — 2012-05-24T08:36:10

Thank you.  Applied as ef5eb41.

Re: Removing from the opcode tree

Paul Johnson — 2012-05-24T08:22:13


I think this is the only real obstacle to getting Devel::Cover working
under threads.

So I'm going to close my eyes and concentrate really hard on willing
magical large gobs of time into existence.

Re: Perl secret operators manual page

Philippe Bruhat (BooK — 2012-05-24T08:01:38

I'm not sure about providing a link from perlop. I think of it more like
an "easter egg", a little nugget of sugar or chocolate that people have
to look for or to be pointed to. Hidden, sweet and nutricious.

[perl #112796] The reentrant API should always have prototypes

Father Chrysostomos via RT — 2012-05-24T07:49:57

Are you going to apply it?

[perl #112784] mathoms broken under C++

Father Chrysostomos via RT — 2012-05-24T07:48:39

Are you going to apply this?

[perl #111446] pod2html generates illegal UTF-8

Father Chrysostomos via RT — 2012-05-24T07:47:34

The release has happened. :-)

Re: Perl secret operators manual page

Elizabeth Mattijsen — 2012-05-24T07:44:15

I do think it belongs in the core distribution, just as "perldoc cheat" does.  And with a link from perlop, as that would be the place where people would look first to look up documentation on this strange operator they found in someone else's code.



Liz

Re: Perl secret operators manual page

Richard Foley — 2012-05-24T07:01:32

Indeed -} Political Correctness {- rears it's ugly head once more.