there's a known issue if you destroy a view while it's loading. we leak open requests.
A On May 22, Henry Minsky wrote: > I will take a look. One question, the timeout for a load request is 30 > seconds by default, does anything seem to unfreeze if you wait 30 seconds? > > There are only two loader network connections available to the app (thanks, > Microsoft) and if something is holding them locked then no other load > requests will be permitted to start. > > > On 5/22/06, kathryn aaker <[EMAIL PROTECTED]> wrote: > > > > I'm about to submit a bug on this and I can't find any open relevant > > ones in JIRA, but was wondering if anyone else had come across it. > > > > ----------------- > > Summary: > > I have views that replicate according to their bound data- they have a > > subview with a datapath bound to an image source attribute and call > > setSource() from the applyData method. I change that bound data based > > on user action- in my test case, a button click. If that data changes > > too quickly(If you double/triple click), setSource() is called, but it > > doesn't actually go and request the image. Not only that, but from that > > point forward, *ALL* outbound requests fail silently. > > > > ----------------- > > Test files: > > test-loadingbug.lzx - embededed data and laszlo gui > > images/ - directory of images to load > > > > ----------------- > > To reproduce: (after unzipping to a working laszlo directory) > > 1. Run test-loadingbug.lzx > > 2. Single click "apply a new dataset" to see views replicate and load > > images > > 3. Single click "apply a new dataset" to see views replicate again with > > different data > > 4. Click "try sending data" to see a request send (and fail: expected > > behavior) in the debugger our output field > > 5. Click "apply a new dataset" several times quickly, until the > > replicated views show as gray boxes with no images > > 6. Click "try sending data" to see a request send, but not show > > response (fail) from server. > > > > ---------------- > > You can verify with an external debugger like liveHttpHeaders, Charles, > > or Fiddler that the calls are or are not being sent. > > > > --------------- > > Notes: > > * Reproduces with both proxied and unproxied applications > > * Reproduces with FP7 in IE, FP8 in Firefox, and the standalone SWF > > player with compiled lzx.swf > > * Reproduces with setSource() called from ondata event instead of > > applyData method > > * Reproduces with OpenLaszlo 3.2 and 3.3 > > * Reproduces with images located in local filesystem or loaded remotely > > > > ------------------ > > My guess is that there's something going on when a replicated view is > > removed while an image is still loading into it. I'm not sure where to > > look from here. > > > > Any thoughts or ideas? > > > > Thanks much, > > Kathryn > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > Laszlo-user mailing list > > [email protected] > > http://www.openlaszlo.org/mailman/listinfo/laszlo-user > > > > > > > > > > > _______________________________________________ Laszlo-user mailing list [email protected] http://www.openlaszlo.org/mailman/listinfo/laszlo-user
