Download xyz.hash.part filenaming is sometimes "backwards"
Categories
(Firefox :: File Handling, defect)
Tracking
()
People
(Reporter: me, Unassigned)
References
(Regression)
Details
(Keywords: regression)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:128.0) Gecko/20100101 Firefox/128.0
Steps to reproduce:
Download a file containing multiple "." in the filename
e.G. affinity-publisher-msi-2.5.3.exe
Actual results:
File while downloading is named
affinity-publisher-msi-2.R6eOAc9P.5.3.exe.part
Expected results:
File should be named
affinity-publisher-msi-2.5.3.exe.R6eOAc9P.part or
affinity-publisher-msi-2.5.3.R6eOAc9P.exe.part
So the Hash should be placed after the last occuring "." not after the first from the left.
Comment 1•1 year ago
|
||
The Bugbug bot thinks this bug should belong to the 'Firefox::File Handling' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.
Hello,
I was able to reproduce this issue on Firefox Nightly 130.0a1 (2024-07-28), using Windows 11, while following the steps described in Comment 0 (the file can be downloaded from https://store.serif.com/en-us/update/windows/publisher/2/).
I have looked for the regression range, and the issue is reproducible starting with Firefox Nightly 98.0a1 (2022-02-03).
Using Firefox Firefox Nightly 98.0a1 (2022-02-02), the name of the file is displayed as 'affinity-publisher-msi-2.5.3.exe.part' during the download.
Pushlog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=82085e71004296ecde5f16d6f5463fc6224678a4&tochange=0714a788f670287b2e1a7b53dc3dd84637501a4e
Unfortunately I am not able to identify the regressor from the pushlog and Mozregression is unable to bisect more data. Hope the pushlog helps!
Setting as NEW so the developing team can have a look.
Comment 4•1 year ago
|
||
In that range there is Bug 1752996 that introduced the random component in .part files, so I think this may have been broken from the beginning.
Comment 5•1 year ago
|
||
Setting Bug 1752996 as the regressor, please correct if needed
Comment 6•1 year ago
|
||
Set release status flags based on info from the regressing bug 1752996
Updated•1 year ago
|
Comment 7•1 year ago
|
||
I think provided that the final file has the correct name, it's not worth trying to fix this.
Even if to a human, it's obvious that "2.5.3" is not an extension, it's not obvious to the computer. Moving where the random bit goes in the filename requires a bunch of people to page a bunch of code back in and make sure that doing that doesn't break anything. It seems easier to just leave this be - the .part file name is not really expected to be exposed to humans for very long (or at all).
Updated•1 year ago
|
Updated•1 year ago
|
Description
•