Re: gettext, plural form and translation - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: gettext, plural form and translation
Date
Msg-id 49C2489C.4010208@gmx.net
Whole thread Raw
In response to Re: gettext, plural form and translation  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Tom Lane wrote:
> Peter Eisentraut <peter_e@gmx.net> writes:
>> Greg Stark wrote:
>>> If the "(n rows)" is the *only* message that needs it then I think it
>>> would be simpler to just make it "(Rows: n)" instead. But I wouldn't
>>> be surprised if there were other messages with similar issues.
> 
>> There are a few more, e.g.,
> 
>> %d index pages have been deleted
>> %d connections
>> Identifier must be less than %d characters.
> 
> What's supposed to happen when a message contains more than one
> number (for example, most of the vacuum activity messages)?

Heh.  Good point.  That is not supported.  It would obviously explode 
the API.  But I agree it's a problem.

Btw., you can find out how much of a problem by using

for x in $(find . -name "*.pot"); do msggrep -K -E -e '%[diu] 
[[:alpha:]].*%[diu] [[:alpha:]]' $x; done

and manually hand-filtering the rest.  I count about 16 problem 
messages.  They are mostly vacuum messages as well as messages of the 
kind "expected %d things but received only %d items".  There are a 
number of additional messages that circumvent the problem by writing 
"expected %d things but received only %d", but that is not possible in 
all cases.


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: gettext, plural form and translation
Next
From: Aidan Van Dyk
Date:
Subject: Re: gettext, plural form and translation