leech_yahoo_terms does nothing else than taking a node's body and title and submitting it to the Yahoo Term Extraction Service. The Term Extraction Service returns an array of keywords, leech_yahoo_terms stores them into a designated taxonomy vocabulary.

This is a pretty general task and shouldn't be tied into leech as it is now.

leech_yahoo_terms could be easily a standalone module that allows for autotagging of any kind of content (yahooterms.module).

I open this issue to have a place where to collect specs, feature requests and patches for such a module.

Comments

Ian Ward’s picture

In addition to running on nodes, the ability to run on user comments and have a comment tag cloud could be useful.

Christoph C. Cemper’s picture

Yep, I agree and got this func in a "Yahooterm" module hack here already...
I don'T wanna mess around with simple patches, diff files and all that crap...

what about creating a new module for that?

alex_b’s picture

Hi Christoph,

I would appreciate that very much.

I guess the only major change leech_yahoo_terms module would need is a section on the settings page on which a user defines what node types to submit to the Yahoo Terms extraction service and on what event that should happen (e. g. on create and/or update).

If you put up such a module, I ll adapt leech to it and throw out leech_yahoo_terms. Everything that makes leech lighter makes me happier.

Alex

esppat’s picture

Assigned: Unassigned » esppat

A "yahooterm" module would be great, yes ...

I am not a php master, but I could help a bit (almost 20 years of dev.).

One of my urgent needs is to have multiple involved vocabularies :
- the first vocabulary which always receive new tags from yahoo (as now if the read-only switch is off)
- the second vocabulary which contains the FORBIDDEN words (those NOT to automatically add to the first vocab.)
- the third vocabulary which is read-only (like the current one if read-only is set)

The second vocab. should be maintained "by hand" by the webmaster, and like this, the newly added words in the first vocab. would stay very smart.
Of course, all this need a function to "reindex" all concerned and existing nodes ... The function should also remove from the first vocab all words which are added in the second one.

This would permit to be very smooth when using the module.
Let's go ?

You can reach me at importchinexport {at} gmail com
See you