Sorry, I just found the error. Simple an empty attribute. Python showed me strange row numbers. I used a multi line initialization of a dict. The error always appeared on the first line although I was on an other line.
Interesting behavior. Jonatan JoReiners schrieb: > Hello, I have a really strange problem. I'm unable to figure it out on > my own. > > I parse very simple xml documents, without any check for their form. > These files look very similar and are encoded in UTF-8. > > Now minidom is always able to parse these files with > minidom.parse("file") . > Now when fetching I use this expression: > xmldoc.getElementsByTagName('DocNumb')[0].firstChild.data.encode('latin1') > > (I know it's not beautiful but convenient. ) > > It always work, but sometimes it simply fails to fetch any element. > Redoing it from hand in the python console with the same file always > works. > > I have no idea what the error is. I compared working and not working > files with the hex editor. I couldn't find any significant change. It's > too strange. If anybody has any clue, hint or anything, please let me > know. > > Regards, Jonatan -- http://mail.python.org/mailman/listinfo/python-list