Inline patterns in XML

Question

Hi,

It's likely that this matter has been queried already, but I couldn't find a good match, if so my apologies.

I have the below XML structure, where I want to treat the variables {} as inline tags in Studio (and ideally also newlines \n).

  <Term Translate="true">
    <Id>Check36VStatus.FailText</Id>
    <String>{0} - {1} - Have you turned on the Rider with the key?</String>
    <Reference>Text shown when a test step fails for product X</Reference>
    <Added>0001-01-01T00:00:00</Added>
  </Term>
  <Term Translate="true">
    <Id>Check36VStatus.SuccessText</Id>
    <String>{0} - {1}</String>
    <Reference>Text shown when the test step is successfully executed for product X</Reference>
    <Added>0001-01-01T00:00:00</Added>
  </Term>
  <Term Translate="true">
    <Id>Check36VStatusTextDescription</Id>
    <String>Turn on the machine by turning the key.\n\nImportant, if this test fails then all other tests will fail as well.</String>
    <Reference>Description of a UI component for product X</Reference>
    <Added>0001-01-01T00:00:00</Added>
  </Term>

I've tried to use Embedded content with both of the two below settings, but neither causes the brackets to be rendered as tags.

I don't want to use the TM Variable list as I need a stricter enforcement of the rule. Also, I realise that a workaround would be to tag the source text with something like <DNT>{0}</DNT> but I'd prefer not to have to search and replace in the source/final files.

Is the ECP/Inline tags the correct way to address this, and if so what am I doing wrong?

Thanks!

Simon

Paul · Accepted Answer

Hi Simon, 
 
No that won't work. The identifier is all lowercase and you used "Xpath". Try "xpath" and it should be fine. 
 
Regards 
 
Paul

Jerzy Czopik · Answer

Have you defined the TXT embedded content processor in your XML file type? If so, did you enter a proper structure information in all relevant tags? If yes, try to use {[^}]*} or at least {.*}, but the second one will mask ANYTHING between [ and }, which will result in wrong parsing. Your rule has an error - you used {*.}... You can also test {.+?} - this means { followed by at least one character and stopping when } is found (lazy rule). 
 But why don't you simply use XML parser with legacy embedded content and use the same regex, but there?

Paul · Answer

Hi Simon,

Jerzy referred to the XML filetype. You, for some reason, have chosen to use the text filetype and not the XML filetype. So if you persevere with your method I think your problem is just that your regex is incorrect. You used this

{*.}

You should have used this:

{.*?}

But looking at your file the XML filetype would be better because you'll then have your elements taken care of automatically and you'll only have to use the embedded content processor for the {0} and \n\n etc.

Regards

Paul

Paul Filkin | RWS Group

________________________
Design your own training!
You've done the courses and still need to go a little further, or still not clear?
Tell us what you need in our Community Solutions Hub

etc. Regards Paul

Paul · Answer

Unknown said: 
 Between my XML segments I have something like this: 
 La corretta installazione della macchina.{CR}La conoscenza del funzionamento 
 Where {CR} would be the paragraph break character. 
 Is there any way I can tell trados to use it as a segment split character (and of course hide it)? 
 
 Hi Enrico, 
 To do this you need to create a segmentation rule on your TM. I guess the lack of a space after the period is preventing the segmentation normally. 
 Regards 
 Paul

Paul · Answer

Hi Enrico, 
 ok - after you managed to split the segments with a segmentation rule and you end up with this: 
 
 La corretta installazione della macchina.{CR} 
 La conoscenza del funzionamento 
 
 You just need to make a tag excluded like this: 
 
 Now it should be perfect! 
 Regards 
 Paul

Paul · Answer

Unknown said: Oh, this is what exclude stands for. And now I understand, why it does NOT segment here. Would it be possible to add that so far to the embedded content? Otherwise we will need to create special TMs for such cases... 
 Hi Jerzy, 
 I recall asking for this a long time ago and there are technical limitations apparently because of the way the files are converted to sdlxliff. I think you just need to knock up a set of rules for your TMs when needed, and use them for segmentation only with no read/write capability. Would be easier if it was possible on the filetype I agree... but it's not :-( 
 Paul

Trados Studio > 1. Trados Studio

Inline patterns in XML