[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: Meaningless kdelibs messages?
From:       Hamish Rodda <meddie () yoyo ! cc ! monash ! edu ! au>
Date:       2002-02-24 9:40:46
[Download RAW message or body]

On Saturday 23 February 2002 9:33, David Faure wrote:
>On Friday 22 February 2002 23:00, Gaute Hvoslef Kvalnes wrote:
>> I encountered these two messages in kdelibs.po, and I don't understand
>> what they mean. Perhaps they should be saying something like 'The request
>> expected to get a file in return'? (The request doesn't return anything,
>> it receives a return value. Or do I miss the point completely?)
>>
>> #: kio/kio/global.cpp:632
>> msgid ""
>> "The request expected to return a file, however the directory "
>> "<strong>%1</strong> was returned instead."
>>
>> #: kio/kio/global.cpp:640
>> msgid ""
>> "The request expected to return a directory, however the file "
>> "<strong>%1</strong> was returned instead."
>
>Well, I'm not sure what it should say, but here's what it's about:
>imagine you want to list a directory. You ask something else
>(here a "kioslave") to list directory /blah for you, issuing a request
>to it. If the ioslave detects that /blah is in fact a file, it will return
>the second error case above.

That's correct. This is a more verbose (and I hope clearer) version of the 
current error message ("%1 is a directory, but a file was expected.", and 
vice versa).  I don't think end users should ever see this error message, 
it's more for developers (which is a quite rare case).

>So maybe it should say.... "the request expected a directory
>but found a file instead" or something like that.

That's actually the same meaning to me (does the translation what you have 
just said have a different meaning to the translation of the strings in the 
po file?)

>Actually you're right,
>it doesn't return anything at all, just an error message.

>Anyway, I'm surprised to see those strings there, they're not used yet.
>Rodda: if all those strings are unused in KDE 3.0, why have they been
>left in ? In order not to lose the few existing translations ? This sounds
> like quite a lot of wasted time to me (for translators), to translate
> strings that can't ever be displayed yet, and that will probably change
> once you work on them again after 3.0.....

The vast majority won't change, I will make sure of that; the problem is only 
a few of the possible solution strings. Can i18n strings be given "msgid" 
values (it appears so in the examples above) so that if there is a change, 
translators simply review the old translation with the new english version?

From everything that was said back when they were committed I thought it was 
best to leave them in (translators working on beta2 would be using the po 
files Dirk created containging those strings); I even asked on i18n-doc and 
got only one response, saying they were ok to stay.

I would even like to cut out the few misleading strings and proceed with 
turning the detailed error messages on for 3.0, except that I don't want to 
divert attention away from debugging, seeing that these messages are 
considered a feature.

I would be grateful for your guidance...

Cheers,

Hamish
[prev in list] [next in list] [prev in thread] [next in thread]