Hey- Bart van den Eijnden (OSGIS) wrote: > Hey Tim, > > sorry for not providing a test case, it just slipped my mind in this > case. I was also trying to see if there was interest before going to the > trouble, so you've beat me to it in a way. Thanks. >
No problem. Thanks for the patch. I just need a vacation after too much test writing. > I've just run tests before and after applying the patch in IE7, and can > confirm things work okay, except the usual suspect overview map failing > 1 test, but there is an open ticket for that already. > > Thanks again. > > Best regards, > Bart > > Tim Schaub wrote: >> Hey- >> >> Christopher Schmidt wrote: >> >>> On Wed, Feb 04, 2009 at 09:53:27AM -0700, Tim Schaub wrote: >>> >>>> Hey- >>>> >>>> [email protected] wrote: >>>> >>>>> Typo in my post, the last one should be _geometry, so: >>>>> >>>>> if(node.childNodes.length == 1 && >>>>> node.firstChild.nodeType == 3) { >>>>> if(this.extractAttributes) { >>>>> name = "_attribute"; >>>>> } >>>>> } else if (node.childNodes.length == 0) { >>>>> name = "_attribute"; >>>>> } else { >>>>> name = "_geometry"; >>>>> } >>>>> >>>>> >>>> Yeah, I like the look of that better :). >>>> >>>> I think this sounds good. Please open a ticket. >>>> >>> http://trac.openlayers.org/ticket/1927 >>> >> >> Bart, I added some tests. If someone runs them in IE, I'll commit. >> >> Also, if you can't provide a test with a patch, at least some sample >> data helps. I get tired of making up GML all the time. >> >> Tim >> >> >>>> Tim >>>> >>>> >>>>>> Hi list, >>>>>> >>>>>> currently the GML/Base.js parser ignores attribute nodes like: >>>>>> >>>>>> <rws:ELEMENTCODE></rws:ELEMENTCODE> >>>>>> >>>>>> This is because of the following check: >>>>>> >>>>>> if(node.childNodes.length == 1 && node.firstChild.nodeType == 3) { >>>>>> >>>>>> However, I would like to parse this type of attribute node (because >>>>>> otherwise I will have to protect all code that does something with >>>>>> the >>>>>> features against this case, e.g. JSON encoding the features and >>>>>> exporting >>>>>> them to Excel results in a shifting columns problem), so I've >>>>>> modified >>>>>> Base.js in the following way: >>>>>> >>>>>> if(node.childNodes.length == 1 && >>>>>> node.firstChild.nodeType == 3) { >>>>>> if(this.extractAttributes) { >>>>>> name = "_attribute"; >>>>>> } >>>>>> } else if (node.childNodes.length == 0) { >>>>>> name = "_attribute"; >>>>>> } else { >>>>>> name = "_attribute"; >>>>>> } >>>>>> >>>>>> >>>>>> Is this something general, or very specific to my use-case? Should >>>>>> I open >>>>>> up a ticket with a patch? >>>>>> >>>>>> Best regards, >>>>>> Bart >>>>>> >>>>>> _______________________________________________ >>>>>> Dev mailing list >>>>>> [email protected] >>>>>> http://openlayers.org/mailman/listinfo/dev >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> Dev mailing list >>>>> [email protected] >>>>> http://openlayers.org/mailman/listinfo/dev >>>>> >>>> -- >>>> Tim Schaub >>>> OpenGeo - http://opengeo.org >>>> Expert service straight from the developers. >>>> _______________________________________________ >>>> Dev mailing list >>>> [email protected] >>>> http://openlayers.org/mailman/listinfo/dev >>>> >> >> >> > > -- Tim Schaub OpenGeo - http://opengeo.org Expert service straight from the developers. _______________________________________________ Dev mailing list [email protected] http://openlayers.org/mailman/listinfo/dev
