preprocessor for iwgenindex ?

nipper

I have a number of DCRs that I want to get summaries and keywords for. So I am going through the MT 4.1 Users manual (on Win2K) and running iwgenindex on the type xml.

I am suspecting I missed a step as this will probably use all the DCR, including image paths, radio buttons, etc in the summary and keyword list. I certainly do not want that.

What did I miss ? Tips/pointers/RTFMs appreciated.

Andy

Migrateduser

Easiest way is to invoke the preprocessor on the corpus (collection of docs) outside of MT to generate input text, then run iwgenindex in -noconvert mode.

Second way is to run the docs through MT and save the modified cracked text into an attribute that you can then use for running iwgenindex in -noconvert mode.