Hans,
I am sorry for replaying that late. Actually what I did, was just
a unification. I did not attempt to improve the messages as such,
although I should have.
All the texts I touched do get displayed to the user, they are
not (hidden) log or debug messages. Some of them were terminated
with dot (like a complete sentence) and some were not. Some
were translated and some were not. After my changes, all should be
translated and not terminated, to be consistent with all the other texts
returned by mnote_*_entry_get_value().
Yes, I think it gets better user experience. The users may also help us
in registering new not-yet-known values of new cameras, if they clearly
understand the situation. Not every user of SW using libexif speaks
English. For example, many users of the Czech software I participate
on don't speak English, but we do have good bug-reporting system
from our users towards us. Actually this product and feedback from it
is the reason why I joined the libexif team and why I do so many changes...
-- Jan
>> --- mnote-pentax-entry.c 13 May 2007 19:21:47 -0000 1.11
>> +++ mnote-pentax-entry.c 24 May 2007 06:47:14 -0000 1.12
>> @@ -341,7 +341,7 @@ mnote_pentax_entry_get_value (MnotePenta
>> /* search the tag */
>> for (i = 0; (items[i].tag && items[i].tag != entry->tag); i++); if
>> (!items[i].tag) {
>> - strncpy (val, "Internal error", maxlen);
>> + strncpy (val, _("Internal error"), maxlen);
>> break;
>> }
>
> OK... so we have one of the worst error messages ever (because it does
> not give ANY hint as to what may have been the cause), and one which
> definitely needs developer attention to fix if a user sees it.
>
> Does it make the user experience better if that kind of message is
> translated? The language for discussions with the developers is mainly
> English, and we don't have translators for bug fixing discussions :)
>> @@ -350,7 +350,7 @@ mnote_pentax_entry_get_value (MnotePenta
>> (items[i].elem[j].index < vs); j++);
>> if (items[i].elem[j].index != vs) {
>> snprintf (val, maxlen,
>> - "Internal error (unknown value %i)", vs);
>> + _("Internal error (unknown value %i)"), vs);
>> break;
>> }
>> strncpy (val, items[i].elem[j].string, maxlen);
>
> OK, that is a little better. At least some hint.
|