User-Agent: Opera/9.80 (Windows NT 5.1; U; pl) Presto/2.6.30 Version/10.61 Build Identifier: Mozilla/5.0 (X11; Linux armv7l; rv:2.0b5pre) Gecko/20100818 Namoroka/4.0b5pre Fennec/2.0a1pre Maemo 5 fennec nightly (20100818) Reproducible: Always Steps to Reproduce: 1. Go to http://pidgin.im 2. Tap 'Download Pidgin' button 3. On the next page tap again on 'Download Pidgin' link 4. Tap 'Save' button 5. Open browser preferences and tap on downloads icon 6. Tap 'Pause' button Actual Results: It's not possible to pause the download, tapping on 'Pause' button has no effect Expected Results: Downloading should be paused
We need to support nsIResumableChannel in e10s.
nsIResumableChannel should be fully supported from the suspend/resume work I did already. test_resumable_channel_wrap.js passes completely, at least.
We need to support nsIResumableChannel in ExternalHelperAppParent, is what mfinkle means. jdm: Feel free to drop a patch on me, if you like. If not I'll probably be harassing you for review of my own in the next few days.
Created attachment 468248 [details] [diff] [review] Patch Piece of cake. Pausing and resuming downloads works like a charm.
Some Notes: - Pause does work if you Cancel/Retry the download first. - This is seen on Android and Maemo on 8/23 nightly builds.
Wow, thanks jdm!
Reopening bug. This is still broken when trying to pause directly. I can confirm it pauses only after you cancel/retry as aakash noted. Tested against: Mozilla/5.0 (X11; U; Linux armv7l; en-US; rv:2.0b4pre) Gecko/201000901 Namoroka/4.04pre Fennec/2.0a1pre Mozilla/5.0 (Android; Linux armv7l; rv:2.0b5pre) Gecko/20100901 Firefox/4.0b5pre Fennec/2.0a1pre
If you cancel/resume, you're actually restarting the download on the chrome side, so that basically means pause doesn't work right for the e10s case. jdm, do you have thoughts on this?
I definitely remember it working when testing my changes; I'll see if I can reproduce the problem on my local desktop.
Created attachment 471910 [details] [diff] [review] Add missing interface reference to ExternalHelperAppParent.
Comment on attachment 471910 [details] [diff] [review] Add missing interface reference to ExternalHelperAppParent. I feel pretty dumb. Apparently I wasn't actually testing it properly, because it couldn't have worked without this change. Anyways, we should be golden with this patch.
Comment on attachment 471910 [details] [diff] [review] Add missing interface reference to ExternalHelperAppParent. Gee gee.
We should have unit tests for this bug, so flagging in-testsuite?. Otherwise, verified FIXED on builds: Mozilla/5.0 (X11; U; Linux armv71; Nokia N900; en-US; rv:2.0b6pre) Gecko/20100907 Namoroka/4.0b6pre Fennec/2.0b1pre and Mozilla/5.0 (Android; Linux armv71; Nokia N900; en-US; rv:2.0b6pre) Gecko/20100907 Namoroka/4.0b6pre Fennec/2.0b1pre