Re: [PATCH] Add L10N percentages to translation strings.

"Kevin J. McCarthy" <[email protected]> Fri, 1 May 2026 21:11:05 +0800

Newsgroups	gmane.mail.mutt.devel
Message-ID	<afSmaTwgJGZKuzxN@qinghai>

On Fri, May 01, 2026 at 02:53:57PM +0200, Oswald Buddenhagen via Mutt-dev wrote:
>On Fri, May 01, 2026 at 08:41:37PM +0800, Kevin J. McCarthy wrote:
>>I could put the initial "%s %s" in each piece, but I still don't 
>>think it would make it much clearer.
>>
>dunno, i think it would - adding _() in four places would be easier
>to do and more legible than having a dedicated formatter function, 
>esp.  with the verbosity that coding this in c implies at some call 
>sites.
>
>>I was already a bit apprehensive about i10n the percentages, so I 
>>would be reluctant to take on localization of the whole string.
>>
>i know that feeling, but in case of doubt it's usually better to err 
>on the side of flexibility. in most cases, translators will simply 
>copy the source string - no harm done. to be extra-safe, it makes 
>sense to add these L10N comments that clarify the context and explain 
>which parts should (not) be localized. in rare cases, it might be 
>necessary to add a disambiguating argument _inside_ the gettext call 
>(e.g., to tell apart instances of "%s %s" that are used in different 
>contexts).

Okay, let me take a second look at this tomorrow.  I think I understand 
what you're getting at, that the progress meter parts themselves could 
be rearranged, with different separators and parenthesis.

-- 
Kevin J. McCarthy
GPG Fingerprint: 8975 A9B3 3AA3 7910 385C  5308 ADEF 7684 8031 6BDA

signature.asc (application/pgp-signature, 833 B)