$txt['error_temp_attachments_flushed']

[chadon]

Hi,

I just wanted to let you know about a little mistake in a string with 2 sentences where it should only be one in Post.english.php, SMF 2.1.

Please note that any files which had been previously attached but not posted. Have now been removed.

Another one in the same file:

$txt['error_temp_attachments_lost']

... or that post has been submitedd

[emanuele]

Thanks chadon.
I fixed the first one, while the second is a bit more tricky...the two are different, but I would try to revise a bit the sentence to make it easier to translate. I didn't notice before.

[chadon]

Yes I was only talking about the spelling of "submitedd" in the second one.

[Matthew K.]

Thanks for pointing it out! In the future if you find any issues it'd be awesome if you could add an issue to our GitHub project.

We'll fix that typo shortly!

[chadon]

In the future if you find any issues it'd be awesome if you could add an issue to our GitHub project.

Will do.

[emanuele]

...I'm not sure what I read... lol

Sorry for the misunderstanding.

For sake of clarity, what instead I wanted to say is that the sentence is built like this now:
$txt['here'] = 'here';
$txt['main_sentence'] = 'This should go %1$s';
And then sprintf the first (in the form of link) into the second:
sprintf($txt['main_sentence'], '<a href="somelink">' . $txt['here'] . '</a>');

I'd prefer to have:
$txt['main_sentence'] = 'This should go <a href="%1$s">here</a>';
and then:
sprintf($txt['main_sentence'], 'somelink');

[chadon]

...I'm not sure what I read... lol

I'm not sure either. lol

Is this related to the errors I reported?

Anyways, if I may suggest something, I think it would be easier for the translators' work to keep your first example:
$txt['here'] = 'here';
$txt['main_sentence'] = 'This should go %1$s';
And then sprintf the first (in the form of link) into the second:
sprintf($txt['main_sentence'], '<a href="somelink">' . $txt['here'] . '</a>');

rather than:

$txt['main_sentence'] = 'This should go <a href="%1$s">here</a>';
and then:
sprintf($txt['main_sentence'], 'somelink');

When I translate strings, I see many groups of words or full sentences that are identical and in your second example, we would have to translate the word "here" in many strings while with the first example, we would have to translate it only once if the code ' . $txt['here'] . ' is placed in all the files.

I'm not a coder and I don't know if this would be easy to do but it would be save a lot of work for the translators.

[Arantor]

It would not be appropriate to do that. Other languages do not use the same 'here' in each place, and thus you actually can't do that.

As it is, there is at least one bug related to doing that sort of thing for the word 'members' where it is used in different cases depending on the instance.

[chadon]

Yes but it would be the translators job to place it in the right place like this:

$txt['here'] = 'here';
$txt['main_sentence'] = 'This should go %1$s';

or

$txt['main_sentence'] = '%1$s it should go';

An other exemple where we do a lot of repetitive translation:

We are sorry but an error occured while...  this or that

The first part of the sentence could be translated only once. Not only it would save work for the translators but also save space in the language files.

[Arantor]

The first part of the sentence could be translated only once. Not only it would save work for the translators but also save space in the language files.

Except that isn't the point. While that works in English, it does not work in all languages. You simply cannot do this.

At the cost of spending more time repeating myself, there are already bugs reported in SMF where this *already happens* and it needs to be changed because different languages need to conjugate the term differently.


Even in your own example it won't work properly, because in the first example it will be 'This should go here' while in the latter it would become 'here it should go' - note the lack of capitalisation. Now imagine that the word will be different in other languages, not to mention that other languages have different rules around capitalisation. You cannot impose English-like rules on other languages.

[chadon]

OK, you certainly know more about it than me, it was just a suggestion.

Thank you anyway for your time.

[emanuele]

Yep, there is the risk of multiply strings, but in that case there is the possibility of using copy&paste.
Unfortunately it could be difficult to find a trade-off.

[Arantor]

I don't really see that you can have a trade off. It will work in some places and not in others, for some languages.

[emanuele]

Yep, in that specific case it's "easy" and it make sense to have it done properly.
There are other situations where "do it properly" is quite difficult (I'm thinking about plurals and in general handling of sentences containing numbers, my 2nd pet after banEdit), at least with the current way of handling strings.

[emanuele]

Fixed.