This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Summary: | .HtmlBrowser$1.run blocks awt queue | ||
---|---|---|---|
Product: | platform | Reporter: | pzajac <pzajac> |
Component: | Window System | Assignee: | _ tboudreau <tboudreau> |
Status: | VERIFIED FIXED | ||
Severity: | blocker | CC: | jglick, jtulach, rkubacki |
Priority: | P3 | Keywords: | PERFORMANCE, THREAD |
Version: | 3.x | ||
Hardware: | All | ||
OS: | All | ||
Issue Type: | DEFECT | Exception Reporter: | |
Attachments: | awt-que stacktrace |
Description
pzajac
2003-06-12 13:01:35 UTC
Created attachment 10669 [details]
awt-que stacktrace
I put the code into HtmlBrowser to avoid #9470. This was caused by wrong implementation of java.net.URL.equals. Maybe java.net.URI.equals can be used now but I am afraid that we need to resolve host names anyway and thus reschedule the part of work out of AWT as Petr suggests. Tim, are you sure that responsivness or performance issue is enhancement? Okay, it's a defect. It's definitely not a TASK. Fixed in trunk - moved it to default request processor. Doesn't seem to cause any trouble. Checking in HtmlBrowser.java; /cvs/openide/src/org/openide/awt/HtmlBrowser.java,v <-- HtmlBrowser.java new revision: 1.86; previous revision: 1.85 done Processing log script arguments... Mailing the commit message to cvs@openide.netbeans.org (from tboudreau@netbeans. org) There is concern over whether the fix is correct; see nbdev & cvs@openide. I am attempting to improve the fix, but getting a strange compiler error (even after cleaning) about a duplicate HtmlBrowser$1.class and the build will not complete. I tried moving the second runnable that schedules the change back to the event queue (which had been created inside an anonymous inner class declared in a method body) into its own named class, but the problem persists. I'm not sure what's wrong. I know about the compiler problem. Please do not use annonymous innerclasses for now. I'll try to fix it. Second runnable? Why not use invokeLater (this)? Please also address the problem with RP. You're right, I could use the same inner class for both. What *is* the compiler problem, anyway? Fixed. Again. it looks fine |