RIS_Parser breaks with file containing UTF-8 bom

When importing a file in RIS format that's encoded in UTF-8 and contains a UTF-8 bom the first entry gets lost because the $start variable is initialised to 5 and this causes eg. TY = JOUR to become(tag = '', value = 'UR);

This might be caused by using strpos to determine the position and drupal_substr for the actual splitting of the line.

Maybe the following ticket can help: this seems to fix a similar problem: http://drupal.org/node/173858

In the end I found a simple solution, I just added an empty first line to the file, this makes the BOM stick to an empty line and makes the actual first line correct.