The current gmarkup parse routines will throw away any whitespace that runs before/between/after toplevel xml elements - and whatever other non-element text chars exists in an xml file will make the parser routine to error out when before the first element.
While that not only violates xml standards - it makes a current project of mine a bit more complicated where I want to parse an xml style snippet into a larger tree. (and interesting too, one can not let some real text run in between toplevel xml elements). All these problems - without need. The current scheme makes my project a bit more complicated since I have to make a fake call to promote the parser from STATE_START to STATE_INSIDE_TEXT - issuing some "<_>" text to the parser, and let the parser-handlers filter it out again. See - it is not particularly complicated to make that state-promotion a valid option of a gmarkup parser - it can be achieved with a GMarkupParseFlag to the _new call, so it initializes the state with INSIDE_TEXT - and it makes people like me a bit happier who want to look at documents with an xml syntax - and keep any text around xml elements. It would be nice of you to add such a flag - it's simple, it wouldn't hurt other software you are currently chewing on, and can be reused in a lot of projects doing some simplistic xml parsing. thanks, guido (not subscribed to ML) // mdk 9.0 with glib 2.0.6 // who am I? google for "guidod" _______________________________________________ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list