1770683 - linux download filename has extra .tar

Status: VERIFIED FIXED

(Firefox :: File Handling, defect, P1)

Description

[[e-mail]]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Firefox/102.0

Steps to reproduce:

firefox-102.0a1.en-US.linux-x86_64.tar.bz2 is posted as:
firefox-102.0a1.en-US.linux-x86_64.tar.bz2.tar <====<<<

This started after May 8. I always use the tar command out of my history, so I've been running the 05-08 version since then. Tonight, it was obvious something had to be wrong... tracked it down to my issuing:
tar xf firefox-102.0a1.en-US.linux-x86_64.tar.bz2
all this time until noticing the newer files have .tar added which makes no sense.

Actual results:

Someone updated a script..?

Expected results:

More careful editing... :)

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Core::Widget: Gtk' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 2

[Alice0775 White]

I can reproduce the issue on Nightly102.0a1 Windows10 as well as on ubuntu20.04.

Regression window:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=b32ccff079ea010ad4adc466d76b6e99552fee58&tochange=b91422aa4ba10c7068a0b941f382d4c2ff423400

Comment 3

[BugBot [:suhaib / :marco/ :calixte]]

:enndeakin, since you are the author of the regressor, bug 1746052, could you take a look?
For more information, please visit auto_nag documentation.

Comment 4

[Alice0775 White]

[Tracking Requested - why for this release]:Downloading the following .tar.gz , .tar.bz2 creates .tar.gz.tar , .tar.bz2.tar.

• https://ftp.mozilla.org/pub/firefox/nightly/2022/05/2022-05-24-16-17-18-mozilla-central/firefox-102.0a1.en-US.win64.xpcshell.tests.tar.gz
• https://ftp.mozilla.org/pub/firefox/nightly/2022/05/2022-05-24-16-17-18-mozilla-central/firefox-102.0a1.en-US.linux-x86_64.tar.bz2

Bug 1768898 does not fix this issue.

Comment 5

[Neil Deakin]

Attachment: Bug 1770683, don't replace file extensions that are for known compressed extensions, when both the proper extension and the compressed extension appear in sequence, r=mtigley

Comment 6

[:Gijs (he/him)]

Neil, between this and bug 1768898 I had another look at the patch in https://phabricator.services.mozilla.com/D135951 . My impression is that in GetFilenameAndExtensionFromChannel, we used to only overwrite the extension if there wasn't an extension present at all, whereas here we are (I think) always overwriting it if the mimetype and extension don't match. Given that 102 is also an ESR release, can we relax this overwriting of extensions more generically, without losing all the gains from bug 1746052? I'm worried that the compression thing and JSON/XML/ZIP (from bug 1768898) are just the tip of the iceberg.

Comment 7

[Neil Deakin]

Attachment: Bug 1770683, only replace or append the correct extension for media types, and some selected content types, or if the extension is not present, r=gijs

This is closer to the behaviour before bug 1746052. Also adds some tests for compressed types.

Comment 8

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/7ddfc8dd77aa
only replace or append the correct extension for media types, and some selected content types, or if the extension is not present, r=Gijs

Comment 9

[Cristina Cozmuta (:CrissCozmuta)]

https://hg.mozilla.org/mozilla-central/rev/7ddfc8dd77aa

Comment 10

[Ardelean Oana, Desktop Test Engineering [:oardelean]]

I managed to reproduce this issue on a 2022-05-22 Nightly build on Ubuntu 20.04. Verified as fixed on Firefox 102.0b4(20220605185654) and Nightly 103.0a1(20220606212503) on Windows 10 x64, macOS 11 and Ubuntu 20.04.

Comment 11

[Andrej (:aglavic)]

== Change summary for alert #34418 (as of Sat, 11 Jun 2022 02:56:33 GMT) ==

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)
6%	perf_reftest_singletons parent-basic-singleton.html	macosx1015-64-shippable-qr	e10s fission stylo webrender	69.66 -> 65.60
6%	perf_reftest_singletons attr-selector-1.html	macosx1015-64-shippable-qr	e10s fission stylo webrender	124.35 -> 117.20

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=34418

Comment 12

[:Gijs (he/him)]

(In reply to Andrej (:andrej) from comment #11)

== Change summary for alert #34418 (as of Sat, 11 Jun 2022 02:56:33 GMT) ==

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)
6%	perf_reftest_singletons parent-basic-singleton.html	macosx1015-64-shippable-qr	e10s fission stylo webrender	69.66 -> 65.60
6%	perf_reftest_singletons attr-selector-1.html	macosx1015-64-shippable-qr	e10s fission stylo webrender	124.35 -> 117.20

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=34418

FWIW I'm fairly sure this is misattributed...

Comment 13

[Andrej (:aglavic)]

It is, I seem to have copied the wrong alert summary thank you for bringing this to my attention