Ok I fixed that bug. See the log. Still wondering if we should force a stack trace on some of the null pointer checks.
I'm back on track pursuing my innerHTML stuff. I didn't answer Adam's questions because I'm not sure yet. Probably what I'm doing now will still not be the final answer. But it should help in the short run, and maybe bring us closer in the long run. It's really a zoo. html can run javascript wich can crank out more html, through document.write or through innerHTML, or, javascript can create html like elements using createElement(), and those should be rendered as though they were html. Not easy to come up with a design that handles all these possibilities, then code, then execute, then test. I do agree jsrt has to run many more tests. We've added functionality already, and jsrt is not always kept up to date. Kinda like documentation. Karl Dahlke _______________________________________________ Edbrowse-dev mailing list [email protected] http://lists.the-brannons.com/mailman/listinfo/edbrowse-dev
