I have been working with a translation service, using the XLIFF 2 provider (we'll be doing some jobs that way, others via the Translation Dashboard). We're working with sample content at the moment to ensure the file that their system returns is going to import correctly back into our Umbraco instance.
We are going to need some RTEs translated, which will contain formatting, links, images etc, as per this example:
I see that Translation Manager (vLatest) separates out the alt text of the image. Can it do this for the title text of links as well? The translators did manage to translate the title text by updating the 'originalData' in the XLIFF file but I assume they shouldn't be updating that really?
Also... should there be the 'originalData' tag in the XLIFF file? Our translators were trying this on an umbraco instance themselves and their RTE content was being persisted as CDATA without any 'originalData' nodes. We're a bit confused to why it's different!
But xliff 2.0 doesn't have a native way for us taking the title element out of the original data bit (Well i don't think it does- i am re-reading the spec again, to see if I can understand it)
That explains the difference then, sorry, I don't remember seeing that option :(
I think the translators are going to be happy with the CDATA block - am sending them another file to test now having unticked the 'Split Html' option. If so then no need for the 'separating out' of the title tag. Fingers crossed!
Yep, so far I've seen a lot of different translators/companies like the Xliff files in loads of different ways (hence the options).
I do think I have found the way the title link is meant to be split (on an obscure xliff mailing list post). but it's probably a bit of work - more than 1 or 2 hours ;) so it's unlikely to land in a release soon.
Hey, thanks for checking. The translators have confirmed that they can handle the CDATA format, and we've imported their resulting file successfuly, which is a relief!
We discussed previously that the Link TinyMCE toolbar option doesn't work in the Translation Dashboard. Without that the editors only option to check/edit the link title would be via View Source I believe. If this was separated out in the way that the image alt text is then this would be easy for them. So perhaps that's another reason to do it (but of course the prioritisation is up to you!)
We've extended the img splitting we already had and made it configurable, so there is an HTML tag settings tab now where you can say what tags you want to extract attributes from.
So the default is to extract all title and alt attributes from img and a tags (obv a tags don't really have alt, but if they did)
Is it possible to extract the 'title' tag of links in RTE when translating?
Hi Kevin,
I have been working with a translation service, using the XLIFF 2 provider (we'll be doing some jobs that way, others via the Translation Dashboard). We're working with sample content at the moment to ensure the file that their system returns is going to import correctly back into our Umbraco instance.
We are going to need some RTEs translated, which will contain formatting, links, images etc, as per this example:
I see that Translation Manager (vLatest) separates out the alt text of the image. Can it do this for the title text of links as well? The translators did manage to translate the title text by updating the 'originalData' in the XLIFF file but I assume they shouldn't be updating that really?
Many thanks
Lotte
Hi,
I am fairly sure I have written the code for that somewhere :) - it might be in the next version, I will go have a look.
Kevin
Also... should there be the 'originalData' tag in the XLIFF file? Our translators were trying this on an umbraco instance themselves and their RTE content was being persisted as CDATA without any 'originalData' nodes. We're a bit confused to why it's different!
Hi
yeah if you have the split option turned on (in the provider settings) you should see something like :
But xliff 2.0 doesn't have a native way for us taking the title element out of the original data bit (Well i don't think it does- i am re-reading the spec again, to see if I can understand it)
That explains the difference then, sorry, I don't remember seeing that option :(
I think the translators are going to be happy with the CDATA block - am sending them another file to test now having unticked the 'Split Html' option. If so then no need for the 'separating out' of the title tag. Fingers crossed!
Thank you very much
Lotte
Yep, so far I've seen a lot of different translators/companies like the Xliff files in loads of different ways (hence the options).
I do think I have found the way the title link is meant to be split (on an obscure xliff mailing list post). but it's probably a bit of work - more than 1 or 2 hours ;) so it's unlikely to land in a release soon.
Hey, thanks for checking. The translators have confirmed that they can handle the CDATA format, and we've imported their resulting file successfuly, which is a relief!
We discussed previously that the Link TinyMCE toolbar option doesn't work in the Translation Dashboard. Without that the editors only option to check/edit the link title would be via View Source I believe. If this was separated out in the way that the image alt text is then this would be easy for them. So perhaps that's another reason to do it (but of course the prioritisation is up to you!)
Thanks
Lotte
Hi
you might not need it now, but we have put the splitting tag thing into v2.2.1 release of translation manager
We've extended the
imgsplitting we already had and made it configurable, so there is an HTML tag settings tab now where you can say what tags you want to extract attributes from.So the default is to extract all title and alt attributes from
imgandatags (obvatags don't really have alt, but if they did)That's great to know. Especially as the goalposts keep moving on this project at the moment ;-)
is working on a reply...