On Saturday, 03. December 2005 18:09, Kjetil Kjernsmo wrote: > The first turn-rtsub.diff, I think fixes a bug, since $rtsub is first > assigned a value there, it can't be meaningful as it is.
Indeed. My test code worked due to that loop, pure coincidence. > The second turns most of Moe's commented warns into AxKit::Debug > statements. I think that's useful, at least they are useful to me now > that I try to debug. I have assigned levels as I think is nice, but I > don't know about that, really. I don't think this is a good idea, as it is really debugging output for STL developers, not useful to STL users IMHO. > Finally, there's a patch that turns die's into Axkit exceptions. That's > more elegant, right? I agree, mostly. I left the attribute parsing code as-is, as those are compile-time errors, they usually happen outside of AxKit's exception handling code. > Finally, I think there is something wrong in end_element: $longtag is > found on line 813, but gets no value before it is first used on line > 848. Now, it is Moe-magic, but still... It doesn't work for me... > Perhaps people wants to have a look? I had a look, and it seems a few lines went missing somewhere. There's also a problem report regarding my session plugin in 1.7, I will prepare a patch soon. -- CU Joerg
