Thread: Re: fixing tsearch locale support

Re: fixing tsearch locale support

From
Jeff Davis
Date:
On Mon, 2024-12-09 at 11:11 +0100, Peter Eisentraut wrote:
> I have expanded this patch set.  The first three patches are the same
> as
> before.  I have added a new patch that gets rid of lowerstr() from
> ts_locale.c and replaces it with the standard str_tolower() that
> everyone else is using.

+1 to the patch series.

Note: I posted a patch series to support case folding:

https://commitfest.postgresql.org/51/5436/

and we may want to use case folding for some of these purposes
internally. But this is not a blocker.

There is some kind of compatibility risk with these changes, so we will
need a release note. And we should try to get all the changes in one
release to avoid repeated small breakages.


Regards,
    Jeff Davis




Re: fixing tsearch locale support

From
Peter Eisentraut
Date:
On 12.12.24 19:20, Jeff Davis wrote:
> On Mon, 2024-12-09 at 11:11 +0100, Peter Eisentraut wrote:
>> I have expanded this patch set.  The first three patches are the same
>> as
>> before.  I have added a new patch that gets rid of lowerstr() from
>> ts_locale.c and replaces it with the standard str_tolower() that
>> everyone else is using.
> 
> +1 to the patch series.
> 
> Note: I posted a patch series to support case folding:
> 
> https://commitfest.postgresql.org/51/5436/
> 
> and we may want to use case folding for some of these purposes
> internally. But this is not a blocker.
> 
> There is some kind of compatibility risk with these changes, so we will
> need a release note. And we should try to get all the changes in one
> release to avoid repeated small breakages.

I have committed this and made a note on the open items wiki page about 
making a release note or similar.

I'll close this commitfest entry now and will come back with a new patch 
series for t_isalpha/t_isalnum when the locale-provider-aware character 
classification functions are available.