On 07/21/09 10:28, "Sven-S. Porst" <ssp-li...@earthlingsoft.net> wrote:
> I just looked over the Coins stuff again. And while it looks like the > general code is much better now thanks to using XPath. > > However, it looks like the original improvement to the DOI regexp was > reverted in the process and that we should redo that. That improvement was totally inadequate for URLs, let alone DOIs. > We seem to have a method for that -stringByExtractingDOIFromString in > BibItem_PubMedLookup already, so it may be a good idea to use that. Why do you need a regex? The one for stringByExtractingDOIFromString is for an arbitrary string that may contain a DOI somewhere. In Coins, you already know it's a DOI (because of the prefix) and have the length based on the split string, right? ------------------------------------------------------------------------------ _______________________________________________ Bibdesk-develop mailing list Bibdesk-develop@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-develop
