Re: Re: Status of doc migration to Git
| From: | Hannes Magnusson | Date: | Tue, 25 Mar 2014 03:13:46 +0000 |
| Subject: | Re: Re: Status of doc migration to Git | ||
| References: | 1 2 3 4 5 | Groups: | php.doc |
| Request: | Send a blank email to phpdoc+get-969384908@lists.php.net to get a copy of this message | ||
On Mon, Mar 24, 2014 at 8:06 PM, André L F S Bacci
<andrebacci.listas@gmail.com> wrote:
> On Mon, Mar 24, 2014 at 11:11 PM, Hannes Magnusson
> <hannes.magnusson@gmail.com> wrote:
>>
>> On Mon, Mar 24, 2014 at 6:59 PM, Hannes Magnusson
>> <hannes.magnusson@gmail.com> wrote:
>> > On Mon, Mar 24, 2014 at 1:55 PM, Maciej Sobaczewski <sobak@php.net>
>> > wrote:
>> >>> - Translation statuses; we could probably insert the Git hash in
>> >>> place
>> >>> of the SVN revision number, that doesn't show how out-of-date the file
>> >>> is though. Is there some other convenient way to replace the usage of
>> >>> EN-Revision values, with Git? How do translators use this, if at all?
>> >>> Do you?
>> >>
>> >> This would require rewritting whole system. The fact that revision ID's
>> >> are
>> >
>> > Thats the question I guess.
>> > When we move to git.. What are our options?
>> >
>> > Maybe this requirement isn't really useful for anyone. Just some
>> > overzealous control freak by the "system.
>> >
>> > What incremental numbers give us is see "how many revisions" a file is
>> > out of date.. But really? Does that really matter?
>>
>> Whops. I forgot we no longer use cvs.. Seeing how many "revisions" a
>> translation is out-of-date based on the revision number alone isn't
>> possible - it requires manual checking, or scripting, just like you
>> would need if you are comparing git hashes.
>>
>> The only reason why the "system" cares that its a number is because
>> the revision match is done as x < y, rather then x != y.
>> Moving to git hashes would be fine.
>
>
> What about a timestamp as revision number, and hash aside. It would help get
> diffs and show how old-of-date are the file.
>
> The expansion only need to occurs in en (via filter clean/smudge), and
> translations would be static.
The automatic-expansion is the problem. git doesn't have reserved
keywords like cvs/svn.
-Hannes