Re: [PATCH v5] http: do not ignore proxy path

> At 2024-08-02 10:10-0700, Junio C Hamano <gitster@pobox.com> sent:
>
>> Ryan Hendrickson <ryan.hendrickson@alum.mit.edu> writes:
>>
>>> Hmm. I'd be inclined to take the preliminary clean-up approach, but
>>> some of the existing strings (there are also two "Unsupported
>>> ..."/"Supported ..." strings near the "Could not set..."s) are going
>>> through gettext, and I'm reluctant to interfere with the l10n process.
>>
>> I do not see what you mean by interfering with the localization.
>>
>> If we are updating text to be translated anyway, giving translators
>> the strings that need to be translated _earlier_ rather than later
>> would be more helpful to them, no?
>
> Probably true, but as a new contributor I don't know whether changing
> msgids means more people need to review the patch, more files need to
> be changed, a translation team needs to be notified, the change needs
> to be pushed a different branch... whatever your process is. Localized
> strings are generally more of a headache for drive-by contributors, in
> my experience across different projects.
>
>>> Is this blocking feedback? This strikes me as speculative
>>> over-engineering
>>
>> No, it is loosening a pattern that is overly tight and as a side
>> effect shortening the line to more readable length ;-).
>
> Blocking or not?