I've had this problem myself because of a lot of german and french references in my work. My solution is as follows:
1. Use a RegExp tool like RegexBuddy (http://www.regexbuddy.com) that has a full-featured regular-expression engine and search and replace features (grep + sed + find can do this under *NIX). 2. Search in all .bib files for characters in the range \x80-\xFF (all extended ASCII characters) 3. Now you have a list of matches and you can visually identify the offending characters. 4. For each offending character, simply craft a single search-and-replace that replaces an extended caracter to its LaTeX accented equivalent. For instance, I would search for all 'ü' and replace with '\"u', 'é' is replaced with '\'e', 'À' is replaced with '\`a', etc. I found that to replace these in about 500 references took about 20 minutes. Then, for every new reference you add manually or download from a database, just make sure to replace extended characters with LaTeX versions with escaped diacritics and you should be fine. Hope this helps. Best regards, Tennessee Carmel-Veilleux, ing. jr (OIQ) Electrical engineering masters student, ETS (http://www.etsmtl.ca) Project AREXIMAS (http://areximas.etsmtl.ca) On 21 August 2010 10:17, Wolfgang Engelmann <[email protected]>wrote: > Sorry, this is slightly off topic, but if sombody is using jabref as a data > base program for the bib-files of Lyx, > > could he/she tell me how to avoid the export of abstract, school, > institutions? Those contain apparently sometimes wrong encodings, which > bibtex/biblatex/biber doesn't like. > > I tired finding out from the handbook, but must have overlooked the part > describing it. > > Wolfgang >
