[solved] How do I access lexicon strings in multiple languages?

I have a snippet that sends emails to users, run by a cronjob at set intervals so no current context applies.

The thing is, I need to be able to access lexicons in different languages so English and French users get text in their respective languages.

How do I do it in php? This is roughly what I am doing:

But the second language to load is the one that is used thereafter regardless of the language required.

Is there any way to do this using the api or do I need to manually include both lexicon files, load them into arrays and then access the strings by language + key? [ed. note: okyanet last edited this post 12 years, 5 months ago.]

Really, I would just like to know if this is possible at all with the current state of Revo - before I go ahead and do something inefficient

What if I dynamically set the culture key within the loop... will try that out. Edit: didn't work.

Otherwise, feature request perhaps? [ed. note: okyanet last edited this post 12 years, 5 months ago.]

Do you know whether it's the load that's failing or getting the user's language?

I'd use a different variable name for $user, though I don't think it will solve your problem.

Hi Bob, thanks for the reply.

I'm not doing exactly what I've shown above, it was just to illustrate the problem. There are no issues with the rest of my code and the load works fine - but it doesn't seem to be possible to load two languages at once.

In the same way, if you try to load two lexicon languages using the tag syntax within a resource it will still only make one language available:

This will results in the string repeated in one language only, rather than the first in English and the second in French.

As an example of where this might be an issue (not related to my problem), how would you make the following language choice menu using lexicon tags:

English - Deutsch - Francais - Italiano - Espanol

You'd need to grab the language name from each lexicon language right? [ed. note: okyanet last edited this post 12 years, 5 months ago.]

Only one lexicon language can be in memory at a time for use in a snippet, though recalling load() each time through the loop should work, in theory. OTOH, loading them manually and using str_replace would probably be faster.

I'm surprised that it doesn't work with tags though. I'm almost certain that multiple language tags worked for me when I tested it for the book, though that was a while ago.

I was using my phone before, so let me ask my question more clearly:

In your foreach loop, what you describe could be caused by $user->get('language') always returning the same thing, or by load() ignoring the language argument. I was asking if you'd checked to see which was the case.

Quote from: BobRay at Dec 01, 2011, 07:31 PM

Only one lexicon language can be in memory at a time for use in a snippet, though recalling load() each time through the loop should work, in theory.

That's what I'd figured, and as my lexicon file is quite small (just for the emails) it wouldn't really have been a huge problem.

I'm surprised that it doesn't work with tags though. I'm almost certain that multiple language tags worked for me when I tested it for the book, though that was a while ago.

Just tested again and the last language to be specified is the one that is used for all lexicon tags.

In your foreach loop, what you describe could be caused by $user->get('language') always returning the same thing, or by load() ignoring the language argument. I was asking if you'd checked to see which was the case.

$user->get('language') is working correctly, and in fact load() is not ignoring the argument either. Here's another test I did:

Which outputs:

So load() is working fine but the last 'new' language to load in is the one that stays in memory.

Actually I can easily find a workaround but it would be very useful if there was a way to do it natively - I'll file a feature request unless anyone jumps in with an obvious solution

Did you try uncached language tags?

_

Yeah, tried uncached tags but I think the problem is that only one language file can be loaded into memory so there's no apparent way to access two languages.

Have filed a feature request here:
http://bugs.modx.com/issues/6204

This issue is fixed in Git/nightlies, and will be in 2.2-rc2.

That is excellent news, many thanks Shaun!