Hi Paul

Find it attached. And this is the original webpage (saved as complete webpage).

HTML Sample.zip 

Hi Piero,

If you have one file it might be easiest to follow the advice from Evzen especially because you could also tweak the segmentation rules to separate out the keywords into separate segments.  You could do this in the full file too but it may not be appropriate to segment on a comma elsewhere.

But... if you'd like to persevere here's a way to do it.  First reset your meta rule so that the content attribute is not translatable.  Then add a new rule also called meta and move it above your existing rule like this:

Then edit the rule so the condition is this:

meta[@name="keywords"]

And add one attribute for content and make it translatable:

In effect you are creating a rule that only parses the meta element when it contains the attribute name="keywords".  Then it picks out the content of content attribute and makes it translatable.  This produces this in the editor which is what you wanted:

You may have to do this for each type of content you need.  So you might want to create one for this too:

meta[@name="description"]

Then you get this:

Which renders this:

The point being that priority in the list is important and you just need to create specific rules for each part you wish to extract.  If the html filetype accepted more flexible xpath you could create a single rule for this but I was unable to make this work... so separate rules seems to do the job.

Hope this helps.

Paul

Hi Piero,

If you have one file it might be easiest to follow the advice from Evzen especially because you could also tweak the segmentation rules to separate out the keywords into separate segments.  You could do this in the full file too but it may not be appropriate to segment on a comma elsewhere.

But... if you'd like to persevere here's a way to do it.  First reset your meta rule so that the content attribute is not translatable.  Then add a new rule also called meta and move it above your existing rule like this:

Then edit the rule so the condition is this:

meta[@name="keywords"]

And add one attribute for content and make it translatable:

In effect you are creating a rule that only parses the meta element when it contains the attribute name="keywords".  Then it picks out the content of content attribute and makes it translatable.  This produces this in the editor which is what you wanted:

You may have to do this for each type of content you need.  So you might want to create one for this too:

meta[@name="description"]

Then you get this:

Which renders this:

The point being that priority in the list is important and you just need to create specific rules for each part you wish to extract.  If the html filetype accepted more flexible xpath you could create a single rule for this but I was unable to make this work... so separate rules seems to do the job.

Hope this helps.

Paul

Unknown said:

If the html filetype accepted more flexible xpath you could create a single rule for this but I was unable to make this work... so separate rules seems to do the job.

The HTML parser does not accept full set of XPath rules as the XML parser... this was confirmed by Patrik the other day in one of our discussions.
I would vote for enhancing the functionality one day, it would come handy from time to time, e.g. with various customer-crafted files "enriched" by non-HTML tags and other content (AKA "messy files"), with very specific requirements for what should/shouldn't be translated.