Firefox downloads only last file in case of consecutive download
Categories
(Core :: DOM: Core & HTML, defect)
Tracking
()
Tracking | Status | |
---|---|---|
firefox-esr60 | --- | unaffected |
firefox65 | --- | wontfix |
firefox66 | --- | fix-optional |
firefox67 | --- | fix-optional |
firefox68 | --- | fix-optional |
People
(Reporter: mohamed.bgb, Unassigned)
References
(Regression)
Details
(Keywords: regression)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.109 Safari/537.36
Steps to reproduce:
Firefox version 65.0.1 (64-bit) & Nightly 67.0a1 (2019-02-15) (64-bit) on Ubuntu LTS and Window 10.
Please use the JSFiddle to reproduce the issue. Note, it is working in Chrome (Chromium & Canary), IE11 and Edge.
https://jsfiddle.net/mboughaba/wake4mbt/12/
Open the JSFiddle and trigger the download button, which will trigger using Javascript the download of 3 files.
Very dirty workaround:
Surround the call to the download function in a requestAnimationFrame or setTimeout.
Also, user need to confirm and check Always saveAs similar files.
As you can see the workaround is not something we can easily tell to the users.
Actual results:
Downloading multiple files one after the other leads to Firefox taking only the last download. User is only promoted once for the last file (save/saveAs/open).
Expected results:
All files should be downloaded and user should be promoted for all files.
Reporter | ||
Comment 1•6 years ago
|
||
Last worked on 2-5-2019.
Reporter | ||
Updated•6 years ago
|
Comment 2•6 years ago
|
||
5:09.52 INFO: No more inbound revisions, bisection finished.
5:09.52 INFO: Last good revision: 722bd413de72d9422924c052648a098ae48f3d4d
5:09.52 INFO: First bad revision: 1d6b361c337ad7b5b4fa016030f2f2aa84a76f83
5:09.52 INFO: Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=722bd413de72d9422924c052648a098ae48f3d4d&tochange=1d6b361c337ad7b5b4fa016030f2f2aa84a76f83
Andrea Marchesini — Bug 1495363 - Abort the previous request, if a form is submitted twice - tests, r=smaug
Updated•6 years ago
|
Updated•6 years ago
|
Assignee | ||
Updated•6 years ago
|
Comment 5•6 years ago
|
||
These 2 are duplicate, also if it doesn't seem. The algorithm we are using here is wrong. See my comments in bug 1517667.
Updated•6 years ago
|
![]() |
||
Comment 6•6 years ago
|
||
Andrea, how is this a duplicate of bug 1517667? This bug was regressed by bug 1495363, but bug 1517667 was filed before bug 1495363 was fixed!
I'm going to mark this duplicate of bug 1542912, which has more information about what's going on, I guess.
![]() |
||
Updated•6 years ago
|
Comment 8•6 years ago
|
||
Bug 1517667 shows an interesting issue related to the discussion we had on IRC yesterday. Bz and I talked about how the 'planned navigation task' in HTML Form Element is (or is not) implemented in gecko.
This bug, the double form submission, bug 1517667, they belong to the same root cause.
![]() |
||
Comment 9•6 years ago
|
||
This bug, the double form submission, bug 1517667, they belong to the same root cause.
Not really. They have very different root causes, though in some cases they could all be wallpapered by adding an extra event loop trip during form submission...
Updated•3 years ago
|
Description
•