I have been pulling my hair out for the past two days trying to figure
out a bug. I was getting exceptions at runtime telling me that the
WOHTML parser could not find a component or class named: "someClass"
when in fact, there was a perfectly good webobject tag in the html
with the name "someClass" and a perfectly good someClass declared and
bound in the WOD.
I changed it to an inline binding, and it would not cause that
exception, but would blow up the same way on some other webobject
named "somethingElse" despite the fact that that also existed in the
WOD. Again, I could change it to inline and it worked fine (and blew
up on yet another one).
I had not changed these objects or their declarations in the wod in a
week or more, and they worked fine before. I also hadn't changed my
preferences in weeks.
After all this time, I discovered that they seem to be reliably
reproducible if I add a comment after a binding in the wod (note: the
exception never points to this particular declared webobject, but some
other perfectly good one). Basically, I was changing the list binding
on a WORepetition to a method returning an array from a keypath which
returned an array. Just to be safe <?> I commented the keypath and
added the method as the key before it. e.g. list = mymethod; // my
old keypath
I removed the comment and the exception goes away and everything
works. I add some other comment to that line and the problem
returns. Meanwhile, the syntax coloring makes these comments look
like comments should, but the comment apparently causes psychosis (and
not just mine).
Any insight?
Andrew Kinnie
This archive was generated by hypermail 2.0.0 : Wed Nov 19 2008 - 15:38:44 EST