this post was submitted on 23 Oct 2024
-3 points (36.4% liked)

Firefox

4081 readers
14 users here now

A community for discussion about Mozilla Firefox.

founded 1 year ago
MODERATORS
 

Firefox 131, btw.

To replicate this bug,

go to https://www.st.com/en/microcontrollers-microprocessors/stm32f091rc.html#documentation

tap-hold on one of the documents there,

download link..

Where is it?

It isn't in Downloads..

Firefox says it was downloaded successfully..

OK, so what happens if I Open it?

It opens!!

So, it's saving it to its cache,

& calling that "downloaded", or something??

It's bullying/wrong/abuse, is what it is.

BTW, it doesn't matter if I've tossed all my apps, then rebooted the tablet,

nor does it matter if I toss Firefox & on restarting it just try opening the document in a different private-tab ( open link in new private-tab ), then use the Download button in the top-left:

FF's still fucking me over for the documentation I'm trying to get.

It does seem to be a Heisenbug, intermittent & .. not obvious what the hell the key is, to make it predictable..

Last boot-session it had allowed me to get the Specs .pdf,

but then none of the Application Notes ( damn ALL of them ) appeared in the Downloads folder, when I checked..

Maybe it's seeing if the file has EVER been downloaded, or tried-to-download, & if so, it's vetoing my getting my hands on it?


( if anybody cares:

the Nucleo-64 boards that "Bare Metal C", a NoStarch Press book, works with, apparently include the Nucleo-F091RC board, which may be more powerful than the board that "Bare Metal C" was written for.

https://nostarch.com/bare-metal-c

I can't remember the details, but this particular board should be compatible, and have more memory, at least.

This microcontroller is the one on the Nucleo-F091RC board.

That is why I want to be reading the Reference Manual & the Programmer's Manual, & all the Application Notes for it..

Someday I'll be able to afford the hardware, too.. : )


PS: If this community wants my bug-report deleted from here,

( & obviously Mozilla removed their bug-reporting feature from Firefox/Android, to eradicate bug-reports from Android FF users )

then remove this, & that's fine.

I've no problem with either way:

world wants it gone,

XOR world permits the bug-report to exist.


_ /\ _

top 4 comments
sorted by: hot top controversial new old
[–] [email protected] 6 points 1 week ago (1 children)

I can't follow your train of thought completely, but downloaded PDFs show in the download folder for me on Firefox mobile. What version do you have?

[–] [email protected] 1 points 1 week ago* (last edited 1 week ago)

As I said, 131.

Also, I didn't remember to mention, that FF isn't putting entries in the Downloads listing it keeps,

either for the .pdf's successfully-downloaded, OR for the ones it hides.

Oh, another point..

It seems that some tabs ( not on the same site: haven't tested that, yet ) still download files, but other tabs only fake it..

_ /\ _

[–] [email protected] 2 points 1 week ago (1 children)

I went to that site and downloaded something. It opened in Firefox, and shows in my downloads. From downloads I was able to open it with Librera. There was also a 'download completed' notification. All working as expected.

[–] [email protected] 1 points 4 days ago

I now know that it's a Heisenbug:

Some tabs in my still-running browser-instance fake downloading stuff:

if I try downloading the same file multiple-times, instead of showing the

filename(1).pdf

( the "(1)" part that FF applies, to prevent a filename-collision )

doesn't appear in those tabs, as it isn't really saving the file in Downloads,

but does appear when doing the same download in other tabs, tabs which are saving the files in Downloads.

This means that something in the tab-instance code is being obliterated/broken, somehow,

and it isn't affecting all tabs.

Obviously a coder familiar with the codebase would possibly be able to use that, as leverage, to dig-in & get the bug.

The fact that the FF "Downloads" list isn't being updated properly also is a clue ( the whole subsystem seems to be corrupted-code? )

I don't consider there to be any point in anybody bothering with this discussion, anymore, however,

since it seems to require some ??whatever??, to activate the bug, and if it isn't guaranteed to be always-appearing, then .. what proof is there that it's real, to anyone else?

Why is it that it's happening in some tabs, some of the time??

Why is it happening at all?

Since people just randomly trying to replicate the bug, & NOT seeing evidence, "proves" it isn't real,

therefore .. everybody should just not bother considering it.

Why waste effort.

I've been forced to switch to Brave to get reliable-downloads, & I hate Brave.

Cheers, people.

_ /\ _