develooper Front page | perl.perl5.porters | Postings from September 2008

Re: [perl #41572][PATCH] Subclassing CGI::Pretty dies in ->new

Thread Previous | Thread Next
From:
Lincoln Stein
Date:
September 8, 2008 16:43
Subject:
Re: [perl #41572][PATCH] Subclassing CGI::Pretty dies in ->new
Message ID:
6dce9a0b0809081234n2de94d7dr519569da3a44ad58@mail.gmail.com
I appreciate your response and apologize if I sounded a bit exasperated in
the previous note. A lot of things are going on right now and my stress
level is rising.

Lincoln

On Mon, Sep 8, 2008 at 3:32 PM, Nicholas Clark <nick@ccl4.org> wrote:

> On Mon, Sep 08, 2008 at 10:04:03AM -0400, Lincoln Stein wrote:
> > Unfortunately this patch breaks the pretty.t regression tests. Exported
> > functions, such as h1(), no longer export. I am working on fixing this
> now.
> >
> > Yesterday's patch from Renee to fix parameter processing, broke about 50%
> of
> > the regression tests! Fortunately this was easy to fix.
> >
> > Does anyone run the regression tests before accepting patches into core?
> I'd
> > appreciate it if you perl5 porters could take CGI.pm upstream fixes only
> > from CPAN so that we are in sync.
>
> We've been trying very hard to keep core modules exactly in sync with CPAN,
> by
> sending things upstream, and then waiting for them to come back. Which can
> be
> a bit of a long wait with some dual life module authors, who are nowhere
> near
> as responsive as you are.
>
> Sometimes we forget (which happened once a few weeks ago with some change
> for CGI.pm) but apart from that, changes sent to p5p for CGI.pm have been
> sent to you (unless you were already on the Cc list).
>
> And no, no-one commits changes unless they pass tests locally. Well, unless
> they mess up (which I have once or twice by managing to run tests on a
> different build tree from the one I was committing on)
>
>
> Right now, blead is byte for byte identical with the modules in the CGI.pm
> 3.42 release. (And for our mutual sanity, long may it stay in sync)
>
> I had already done a full diff before you sent this message that I am
> replying
> to - it turns out that there are 3 differences in test scripts, two to do
> with
> where a test file is. We're likely to move the location of CGI.pm in the
> core tarball (into ext/) as part of a plan to make it easier to track
> dual-life
> code, which would mean that these would change, so I was holding off
> sending
> you a patch with the changes until that had happened.
>
> I've appended the full test diff. The third change is an EBCDIC fix, made
> as
> part of http://public.activestate.com/cgi-bin/perlbrowse/p/28568
>
> Change 28568 by stevep@stevep-kirk on 2006/07/13 20:18:59
>
>        Subject: [PATCH] z/OS: CPAN-ized ext/ and lib/
>        From: Jarkko Hietaniemi <jhi@iki.fi>
>        Date: Thu, 13 Jul 2006 23:10:27 +0300
>        Message-ID: <44B6A8B3.5070605@iki.fi>
>
>
> Nicholas Clark
>
> diff -pru ../CGI.pm-3.42/t/upload.t lib/CGI/t/upload.t
> --- ../CGI.pm-3.42/t/upload.t   2008-09-08 14:52:59.000000000 +0100
> +++ lib/CGI/t/upload.t  2008-09-08 19:40:04.000000000 +0100
> @@ -7,7 +7,17 @@
>
>  # Due to a bug in older versions of MakeMaker & Test::Harness, we must
>  # ensure the blib's are in @INC, else we might use the core CGI.pm
> -use lib qw(. ./blib/lib ./blib/arch);
> +
> +my $test_file;
> +if($ENV{PERL_CORE}) {
> +   chdir 't';
> +   @INC = '../lib';
> +   use File::Spec ();
> +   $test_file = File::Spec->catfile(qw(.. lib CGI t),
> "upload_post_text.txt");
> +} else {
> +   use lib qw(. ./blib/lib ./blib/arch);
> +   $test_file = "t/upload_post_text.txt";
> +}
>
>  use strict;
>
> @@ -67,7 +77,7 @@ my $q;
>
>  {
>     local *STDIN;
> -    open STDIN, '<t/upload_post_text.txt'
> +    open STDIN, "< $test_file"
>         or die 'missing test file t/upload_post_text.txt';
>     binmode STDIN;
>     $q = CGI->new;
> diff -pru ../CGI.pm-3.42/t/uploadInfo.t lib/CGI/t/uploadInfo.t
> --- ../CGI.pm-3.42/t/uploadInfo.t       2008-09-08 14:52:59.000000000 +0100
> +++ lib/CGI/t/uploadInfo.t      2008-09-08 19:40:49.000000000 +0100
> @@ -7,7 +7,17 @@
>
>  # Due to a bug in older versions of MakeMaker & Test::Harness, we must
>  # ensure the blib's are in @INC, else we might use the core CGI.pm
> -use lib qw(. ./blib/lib ./blib/arch);
> +
> +my $test_file;
> +if($ENV{PERL_CORE}) {
> +   chdir 't';
> +   @INC = '../lib';
> +   use File::Spec ();
> +   $test_file = File::Spec->catfile(qw(.. lib CGI t),
> "upload_post_text.txt");
> +} else {
> +   use lib qw(. ./blib/lib ./blib/arch);
> +   $test_file = "t/upload_post_text.txt";
> +}
>
>  use strict;
>
> @@ -68,7 +78,7 @@ my $q;
>
>  {
>     local *STDIN;
> -    open STDIN, '<t/upload_post_text.txt'
> +    open STDIN, "< $test_file"
>         or die 'missing test file t/upload_post_text.txt';
>     binmode STDIN;
>     $q = CGI->new;
> diff -pru ../CGI.pm-3.42/t/util-58.t lib/CGI/t/util-58.t
> --- ../CGI.pm-3.42/t/util-58.t  2003-04-14 19:32:22.000000000 +0100
> +++ lib/CGI/t/util-58.t 2006-07-13 21:18:14.000000000 +0100
> @@ -11,6 +11,11 @@ BEGIN {
>  use Test::More tests => 2;
>  use_ok("CGI::Util");
>  my $uri = "\x{5c0f}\x{98fc} \x{5f3e}.txt"; # KOGAI, Dan, in Kanji
> -is(CGI::Util::escape($uri), "%E5%B0%8F%E9%A3%BC%20%E5%BC%BE.txt",
> -   "# Escape string with UTF-8 flag");
> +if (ord('A') == 193) { # EBCDIC.
> +    is(CGI::Util::escape($uri),
> "%FC%C3%A0%EE%F9%E5%E7%F8%20%FC%C3%C7%CA.txt",
> +       "# Escape string with UTF-8 (UTF-EBCDIC) flag");
> +} else {
> +    is(CGI::Util::escape($uri), "%E5%B0%8F%E9%A3%BC%20%E5%BC%BE.txt",
> +       "# Escape string with UTF-8 flag");
> +}
>  __END__
>



-- 
Lincoln D. Stein

Ontario Institute for Cancer Research
101 College St., Suite 800
Toronto, ON, Canada M5G0A3
416 673-8514
Assistant: Stacey Quinn <Stacey.Quinn@oicr.on.ca>

Cold Spring Harbor Laboratory
1 Bungtown Road
Cold Spring Harbor, NY 11724 USA
(516) 367-8380
Assistant: Sandra Michelsen <michelse@cshl.edu>

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About