nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Alternate languages and summidx_en/sp

ShovelFace

Is it possible to set up metadata field name 'keyword' to be populated by summidx_en if in an English record, and summidx_fr if in a Spanish word? I have the English keyword working as expected, but am unable to tie in the French record. Note that this is via TeamSite integration with MetaTagger, and the records are suffixed with a language identifier (e.g.: _en/_sp).

Thx.

Find more posts tagged with

Comments

ShovelFace

Any takers on this one? Class? Bueller? Bueller???

ShovelFace

Ok, let me try to explain better:

Suppose I have the following rack scripts:

summidx_en
Rack Script: parse ;
summarize -keywords 10
Field Type: Summarization based

summidx_sp
Rack Script: parse ;
summarize -keywords 10
Field Type: Summarization based

What I would like to do is, say, the user has the English record opened, I would like to use summidx_en to populate the keyword EA with the English keywords. If, on the other hand, they have the Spanish record opened, I want to populate the same keyword EA with the Spanish keywords. MetaTagger is not allowing me to create rackscripts with the same "keyword" identifier, say something like

keyword:
Rack Script: parse -useLexicon summidx_en -mode phrase ;
summarize -db summidx_en -keywords 5
Field Type: Summarization based

as I am not able to create the following if "keyword" alreay present:

keyword:
Rack Script: parse -useLexicon summidx_sp -mode phrase ;
summarize -db summidx_sp -keywords 5
Field Type: Summarization based

Any guidance here is greatly appreciated.

Thx.

Migrateduser

Best approach is to create two separate content processors and field sets - one per language - and merge the results in a final processor.

E.G.

content processor cp1_en { keywords_en, summary_en, ... }
content processor cp1_sp { keywords_sp, summary_sp, .... }
+ final processor group { "mergeattr.pl keywords keywords_en keywords_sp" }

That said, mergeattr.pl would have to be written, but it should be less than 100 lines.

Cheers,
C

ShovelFace

Thanks for the input.

I've created the content processors as described, and have attempted to create the postprocessor (code below) vs a Final processor. Idea is move the temporary value (i.e.: keyword_en) into the TeamSite EA keyword. (only working on the English record for now). Postprocessor has keyword_en as a parameter.

When the Tagging UI opens, the "Re-Tag" button is not present next to the keyword field.

Any idea what I am missing? Perhaps I am passing the parameter wrong?

Migrateduser

TeamSite using a peculiar method of mapping TagUI fields to MT based on compatibilities with older versions of MT. For it to "see" a field, a field of that name needs to be defined in the MT server, not present in the returned MD.

The workaround is:
define a field name "keywords", but don't add it to you content processor. Add it to a dummy CP that doesn't run (set the regex to "nevernevernever").

Cheers,C

ShovelFace

Hey C,

Note sure what you mean by "set the regex to 'nevernevernever'". Do you mean to create a location regex that is always false?

Thx.

SF

Migrateduser

Yup - create a regex that is always false. That will ensure that the field is active from a TS perspective but the model will not be used from an MT perspective.

TestReport- Row Line Error.rptdesign