Dear All,

I am a heavy user of beagleto index all my scientific documentation archive.

I recently found some articles fail to be retrieved even when they contain extractable text. Namely, they are PDF files with particular words.

On word for all: "transesterification", where the "fi" is one linked character (as typographers say it should be). Unfortunately, though, the "fi" is not converted in separate letters when beagle-extract-content is run on the file, and consequently when I search for "transesterification" in beagle-search this file is not in the "hits" list.

Please not that if I search for "transesterification" inside evince, the word is found.

I think something should be done around this!

Marco

PS I put the file which exhibites this problem in http://ingchim.ing.uniroma1.it/users/mbravi/liu2007bpd.pdf