RE: [PHP-CVS] Re: [PHP-DEV] Re: [PHP-CVS] com php-src: Fix dba configuration for Windows: ext/dba/config.w32
| From: | Anatol Belski | Date: | Sat, 20 Aug 2016 00:20:41 +0000 |
| Subject: | RE: [PHP-CVS] Re: [PHP-DEV] Re: [PHP-CVS] com php-src: Fix dba configuration for Windows: ext/dba/config.w32 | ||
| References: | 1 2 3 4 5 | Groups: | php.cvs php.internals |
| Request: | Send a blank email to internals+get-95329@lists.php.net to get a copy of this message | ||
> -----Original Message-----
> From: Christoph M. Becker [mailto:cmbecker69@gmx.de]
> Sent: Friday, August 19, 2016 4:29 PM
> To: Anatol Belski <anatol.php@belski.net>; PHP Internals
> <internals@lists.php.net>; php-cvs@lists.php.net
> Subject: [PHP-CVS] Re: [PHP-DEV] Re: [PHP-CVS] com php-src: Fix dba
> configuration for Windows: ext/dba/config.w32
>
> On 19.08.2016 at 15:22, Anatol Belski wrote:
>
> >> -----Original Message-----
> >> From: Christoph M. Becker [mailto:cmbecker69@gmx.de]
> >> Sent: Friday, August 19, 2016 1:38 PM
> >> To: Anatol Belski <anatol.php@belski.net>; php-cvs@lists.php.net;
> >> internals@lists.php.net
> >> Subject: [PHP-DEV] Re: [PHP-CVS] com php-src: Fix dba configuration
> >> for
> >> Windows: ext/dba/config.w32
> >>
> >> On 19.08.2016 at 13:17, Anatol Belski wrote:
> >>
> >>>> -----Original Message-----
> >>>> From: Christoph Michael Becker [mailto:cmb@php.net]
> >>>> Sent: Friday, August 19, 2016 11:42 AM
> >>>> To: php-cvs@lists.php.net
> >>>> Subject: [PHP-CVS] com php-src: Fix dba configuration for Windows:
> >>>> ext/dba/config.w32
> >>>>
> >>>> Commit: ad76e8a529eabf150f17d313bb035b329bc68dec
> >>>> Author: Christoph M. Becker <cmbecker69@gmx.de> Fri, 19 Aug
> 2016
> >>>> 11:42:16 +0200
> >>>> Parents: bc1214f25e7c9525336b34e09aec1f1db82b9894
> >>>> Branches: PHP-5.6 PHP-7.0 PHP-7.1 master
> >>>>
> >>>> Link: http://git.php.net/?p=php-
> >>>> src.git;a=commitdiff;h=ad76e8a529eabf150f17d313bb035b329bc68dec
> >>>>
> >>>> Log:
> >>>> Fix dba configuration for Windows
> >>>>
> >>>> To be able to build the dba extension on Windows, libdb was required.
> >>>> This is contrary to *nix where each handler can be configured
> >>>> individually. To avoid BC breaks, we only do minimal modifications,
> >>>> instead of adjusting the Windows configuration to match the *nix
> >>>> configuration, for now.
> >>>
> >>> I'm not sure it's ok to suddenly change this for stable branches,
> >>> and even not
> >> sure with 7.1. Effectively no builds for this were provided since 5.3, if I don't
> err.
> >>
> >> Yes, you're right.
> >>
> >>> So from the QA perspective, it's quite a risky move.
> >>> It should go by master, so we have time to catch up with bugs, deps
> >>> and users to start testing and using it again.
> >>
> >> This appears to be a chicken-and-egg problem. Currently, there are
> >> several known issues wrt. to the flatfile and inifile drivers on
> >> Windows. These drivers will be enabled as soon as libdb3.1s (or
> >> libdb6.1 as of PHP 7.0.0) are available. Users may than experience these
> issues.
> >
> >> However, it is not possible to test (and debug) with dba, if the libs
> >> are not available (unless one modifies config.w32), and it may be
> >> hard to get the respective libdb binaries. There may, however,
> >> dba_php.dlls available from third-party sites.
> >>
> >> Anyhow, if you still think the commit should only go to master, I can
> >> revert the changes. Or feel free to do so yourself. :-)
> >
> > The ini parsing functionality is available in the core anyway. The bundled libcdb
> originates to year 2000, it needs to be reviewed and possibly updated with the
> latest patches. In between, very good alternatives like Redis, MongoDB or even
> upscaledb and others are available for key/value storage. So ext/dba doesn't
> look like something that needs to be ungently available today.
> >
> > The bins might be available as patching the config is easy, or even just with
> using the Berkeley DB libs. However, the bug tracker is silent, which is not good.
> There was no testing since over 5 years on Windows side, there was likely no
> testing in PHP 7 as it's always disabled. IMHO selling that state just as is for
> stable is dangerous. In addition, at least qdbm, not sure with tcadb, could be
> supported. And the bundled libcdb is questionable anyway.
> >
> > It's an if/else question, as for me it's a decision that should be in favor of
> > the
> quality. That would require some time to be guaranteed. I'd ask you to please
> keep it only for master, so there's enough time for all the complex work. Clear,
> some bugs are always there, there's no reason to omit the usual QA work.
> Please keep the incentive, really appreciated!
>
> Okay, I'm going to revert the commits to PHP 5.6-7.1, leaving only master. :-)
>
Thanks! When the 7.1 stress is over, we'll surely have time to check the topic deeper.
Regards
Anatol