Re: DBD::mysql path forward

Pali,

Yes, I agree, we'll have to create a fork pre revert and stop accepting PRs

How might we allow people time to test the fixes to give them time? Just 
have them use the fork, I would assume?

Regards,

Patrick

On 9/12/17 6:27 AM, [email protected] wrote:
> On Tuesday 12 September 2017 05:40:31 Patrick M. Galbraith wrote:
>> Hi all,
>>
>> After talking to Pali and looking at how other drivers have handled the
>> issue, the best way forward will be to deal with solving the UTF-8 issue
>> correctly as was attempted in May. This will be a problem for some, but only
>> due to having to been accustomed to a work-around that relies on a
>> intermittent and buggy implementation. The plan is to give ample time to let
>> people prepare and comment as well as give them a chance to try the changes
>> and adjust prior to stable release. I'm going to look through the mailing
>> list and find particularly those who had problems back in May and June when
>> we tried this before and work with them in advance of the change.
> Hi!
>
> To prove fact that other DBI drivers (e.g. Pg or SQLite) had fixed
> similar/same UTF-8 issue as MySQL has and behave Perl-correctly, I would
> provide test cases so you would see difference between Pg, SQLite and
> mysql DBI drivers.
>
> There are also many modules on cpan which uses mysql DBI driver. But I'm
> not sure if all those modules expects buggy behavior or not...
>
>> DBD::mysql should be on par with all the other DBD drivers and allow
>> transparent usage regardless of backing RDBMS, and this really is the only
>> way. It will also allow for other fixes to proceed and the driver to
>> continue improving and supporting all versions and enhancements of MySQL and
>> MariaDB.
>>
>> Please do give us your thoughts on this as we want to be as helpful and
>> transparent as possible.
> If we are going to have UTF-8 fix in DBD::mysql, I would suggest one
> thing right now:
>
> STOP MERGING AND ACCEPTING PULL REQUEST TO GITHUB DBD::MYSQL PROJECT!
>
> Reason is simple: Current git master repository at version 4.043
> (against which are creating new pull requests) is reset to version 4.041
> and couple of changes, including lot of bug fixes from 4.042 are
> missing
>
> People in 4.043 (and git master) re-discovered 4.041 bugs, which were
> fixed in 4.042 and started to fixing them again, either wrongly or in
> different way. Or just started complaining about them, that they are
> again present.
>
> Those (reverted) UTF-8 commits changes couple of lines, plus other fixes
> depends on it. Once git master repository starts to divergent, it would
> be very hard to incorporate UTF-8 fixes again.
>
> And speaking by myself, for me it would be easier to create a fork of
> DBD::mysql from point before big revert as trying to re-apply and rebase
> those patches on top of master with couple of changes...