Cornee van der Linden 

I'm not sure if there is a better way to do this, but I would do this based on this sample (it always helps to provide a sample!):

<topic xmlns:astX="http://www.astoriasoftware.com/translation">
    <title>4FG456S0</title>
    <shortdesc>Short description:
        <ph>Bloodbags LR Set </ph>
        <term astX:md="true">CPD/SAG-M Blood Bag Set With RBC Leukoreduction Filter</term>
        <term astX:md="false">Standard Blood Bag Set</term>
        <term astX:md="true">Leukoreduced Blood Bag Set</term>
        <term astX:md="false">Platelet Concentrate Bag Set</term>
        <term astX:md="true">Red Cell Preservation Bag Set</term>
        <keyword>Instructions for Use</keyword>
    </shortdesc>
</topic>

First you need to add the namespace as this is a custom value you are using:

Then I used this Xpath for the term element:

//term[@astX:md='true']

I also made the rule structure just in case as it somehow makes no sense to have this as inline with your usecase:

Finally I added this rule and moved it to the very bottom of the list of parser rules as I want this to run first:

//*

I make this non-translatable:

Then when I preview I see this:

As opposed to this without the updated rule for term:

I hope that helps?

Hello Paul, thanks for getting back to me. Really appreciate it. I tried as you suggested, but it does not seem to work. Note that those strings we don't want translated usually have an attribute called astX:sn and then a value that can be anything like for example 00000023WHG804D9E2060GYZ whereas the strings we want translated always have the astX:md attribute with the value set to true

 Cornee van der Linden 

Cornee van der Linden said:

Note that those strings we don't want translated usually have an attribute called astX:sn and then a value that can be anything like for example 00000023WHG804D9E2060GYZ

Usually or always?

Cornee van der Linden said:

the strings we want translated always have the astX:md attribute with the value set to true

ok - so this is ok and in line with my test.

Cornee van der Linden said:

I tried as you suggested, but it does not seem to work.

I think the next step would be for you to create a sample file for me so I don't have to make one up that may not be representative of what you are seeing.  If I provide a solution for you based on my best guess at what you have and it doesn't work for you with your files I have absolutely no idea why not!!

You do not have to give me your customer files.  Just take a copy of one, cut it down in size so it contains at least one example of the different scenarios you have, and then change the content so the text doesn't break an NDA.

That is a trivial task and will help me to give you a better answer.

Hello Paul,

Sorry, I should have said 'always'; the strings we don't want translated always have the attribute called astX:sn and then a value that can be anything like for example 00000023WHG804D9E2060GYZ

I'm not an LSP but work for a medical device company, so I can send you a sample file. I tried to attach it here, but it doesn't let me. Can I make it somehow available to you in a different way?

Thanks,

Cornee

Cornee van der Linden 

Either use Insert -> Code and copy paste as XML; or Insert -> Image/Video/File and attaché either as XML or as a zip.  If you still can't do that email it to pfilkin at sdl dotcom

Let me email you some files. That might be easier.

Thanks again,

Cornee

Cornee van der Linden 

One solution based on what I understand from you so far would be this:

I added two new rules at the top, and then disabled every other rule.

The first rule is this and is set as always translatable:

//*[@astX:md='true']

This will extract the text from any element that has the attribute astX:md='true'.

Then I added a rule to not extract anything at all to make sure I would only apply the first rule to a clean sheet:

//*

Maybe this simple idea will suit your needs?

Cornee van der Linden 

One solution based on what I understand from you so far would be this:

I added two new rules at the top, and then disabled every other rule.

The first rule is this and is set as always translatable:

//*[@astX:md='true']

This will extract the text from any element that has the attribute astX:md='true'.

Then I added a rule to not extract anything at all to make sure I would only apply the first rule to a clean sheet:

//*

Maybe this simple idea will suit your needs?

HI Paul,

Yes, this seems to be working! Thanks again for helping so quickly. Really appreciate it.