CEGUI (Crazy Eddie's GUI) Forums

Good afternoon,
I'm working on porting the new 0.7 version of CEGUI and have a question.
For testing I'm using the configuration xml files from the source code samples release.

The code is throwing an assert here:



When it tries to parse a second <ImagerySection> tag. The documentation shows multiple sections as being valid
http://www.cegui.org.uk/wiki/index.php/The_Beginners_Guide_to_Falagard_skinning_-_Part_I
Am I using the wrong docs and sample configuration files?

Thanks

I have to post the obligatory link to the Forum Guidelines and ask you to post your CEGUI.log file. While I have not heard of this issue before, there are several revisions of v0.7 with numerous bugs and it is important to clarify which revision you are using. Also, when posting xml-related questions it is important to actually include the xml in question (even if from the samples)

In any case, I am far from being a Falagard expert, so I cannot be of much help. However, I will confirm that multiple ImagerySection elements are legal. My first assumption would be that you have used the same name (attribute) for both elements? Again, it would help if we could see the actual xml in question. Thanks.

I'm on version 0.7.1.

Here's my log:

The xml file is too large to post in the forum. It's the unmodified file "CEGUI-0.7.1\datafiles\looknfeel\TaharezLook.looknfeel" from the 0.7.1 download. I can post any portion of it you'd like to see.

Reading the code it doesn't seem that the name is checked before the assert is thrown.

Thanks for trying to help

Thanks for the extra info, I'm sure someone (CE) will be able to use that. Looking at my own log, I noticed that none of the Falagard code is logged by default. So, it may be helpful to turn on additional logging output prior to initializing CEGUI via:



I also noticed you are using a custom xml-parser, among other things. Have you tried any of the supported parsers (tinyxml, expat, etc.)? Ruling out the parser would be one of the first things to try. If the supported parsers work, then it is most certainly an issue with the RapidXML wrapper.

Thanks for the logging code, I'll certainly give that a whirl.
I'm testing the rapidxml as part of the port but it's probably a good idea to minimize the variables.

You will find this issue is another instance of the copy/paste issue from before (but for end element).

For info, a tested, debugged and working version of your RapidXML parser now exists in the CEGUI svn tree - both branches/v0-7 and trunk

CE.

Wow

I'll drop my contribution in the box later tonight...

Thanks very much, it's most appreciated

CE.