Nice work, I like byte puzzles ;-) . I cant test your code atm, but I have a few comments below.
Line 231: > + The field label is separated from the field contents by one random byte. That byte isn't random! It actually specifies how long the string is in characters (or bytes!) Iirc, the SongShowPlus file format is similar. Line 327: Should the tab be invalid as well? I think we should keep the users database clean from formatting. I think webkit will render this as a single space. Line 331: Should this be moved to core/file/__unit__.py I know this could be used else where. Iirc, some code I merged a couple months back does. Finally, I realise this might mean having to make extensive modifications to the import wizard (so not suitable at this time), but could it be made more user friendly by reading the xml file and giving the user a list of song titles, rather than a list of unidentifiable file names. As I said, maybe 2.1 or what ever is after the release? Phill -- https://code.launchpad.net/~sfindlay/openlp/songs-import-powersong/+merge/104102 Your team OpenLP Core is requested to review the proposed merge of lp:~sfindlay/openlp/songs-import-powersong into lp:openlp. _______________________________________________ Mailing list: https://launchpad.net/~openlp-core Post to : [email protected] Unsubscribe : https://launchpad.net/~openlp-core More help : https://help.launchpad.net/ListHelp
