On 2019-Jul-02, Daniel Gustafsson wrote:
> > On 2 Jul 2019, at 22:16, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> > even if we made a test case that presumed
> > --enable-nls and tried to exercise this, the lack of translations
> > for the new words would get in the way for a long while.
>
> For testing though, couldn’t we have an autogenerated .po which has a unique
> and predictable dummy value translation for every string (the string backwards
> or something), which can be used for testing? This is all hand-wavy since I
> haven’t tried actually doing it, but it seems a better option than waiting for
> .po files to be available. Or am I missing the point of the value of the
> discussed test?
Hmm, no, I think that's precisely it, and that sounds like a pretty good
starter idea ... but I wouldn't want to be the one to have to set this
up -- it seems pretty laborious.
Anyway I'm not objecting to the patch -- I agree that we're already not
testing translatability and that this patch shouldn't be forced to start
doing it.
--
Álvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services