[Bug]: onecommander 3.95.2.0 not working

[Gitoffthelawn]

Prerequisites

• I have written a descriptive issue title.
• I have searched all issues/PRs to ensure it has not already been reported or fixed.
• I have verified that I am using the latest version of Scoop and corresponding bucket.

Package Name

onecommander

Expected/Current Behaviour

OneCommander v3.95.2.0 is not working. It was just updated in the Extras bucket 7 hours ago. Launching it causes a momentary spinner cursor, then nothing. OneCommander v3.95.1.0 worked well.

Please note that when OneCommander v3.95.2.0 is launched, Task Manager appears to show OneCommander's main process loading momentarily, then closing/crashing, then loading again momentarily, suspending, then closing/crashing again.

Steps to Reproduce

1. scoop install extras/onecommander
2. try to launch OneCommander
3. notice momentary spinner cursor, then nothing

Possible Solution

Revert to v3.95.1.0? Wait for next version? Contact upstream developer?

Scoop and Buckets Version

Current Scoop version:
859d1db5 (HEAD -> master, tag: v0.5.2, origin/master, origin/HEAD) chore(release): Bump to version 0.5.2 (#6080)

'main' bucket:
15c4fecb6 (HEAD -> master, origin/master, origin/HEAD) supabase: Update to version 2.9.6

'extras' bucket:
fe740d17f (HEAD -> master, origin/master, origin/HEAD) postman: Update to version 11.30.1

Scoop Config

last_update                       scoop_repo                              scoop_branch virustotal_api_key
-----------                       ----------                              ------------ ------------------
2025-01-28T11:27:39.1027912-00:00 https://github.com/ScoopInstaller/Scoop master       redacted

PowerShell Version

Name                           Value
----                           -----
PSVersion                      5.1.26100.2161
PSEdition                      Desktop
PSCompatibleVersions           {1.0, 2.0, 3.0, 4.0...}
BuildVersion                   10.0.26100.2161
CLRVersion                     4.0.30319.42000
WSManStackVersion              3.0
PSRemotingProtocolVersion      2.3
SerializationVersion           1.1.0.1

Additional Softwares

No response

[RichardLuo0]

I have the same issue. I then redownloaded from the onecommander website, but I can not launch either.

[Gitoffthelawn]

@RichardLuo0

I have the same issue. I then redownloaded from the onecommander website, but I can not boot either.

Thank you for the confirmation. That saves me a bunch of time in trying to ensure it wasn't something unique on my end.

Two questions:

1. When you mention that you "can not boot either", you mean after the Scoop-installed version failed, you then downloaded OneCommander from the OneCommander website, and had the same problem, correct?
2. You can still boot the system, just not launch OneCommander, correct?

[RichardLuo0]

1 yes
2 yes

[MonolithOrchids]

Developer already acknowledge the problem (check their discord for info) and fixed it in the version 3.95.3.0

[Gitoffthelawn]

Developer already acknowledge the problem (check their discord for info) and fixed it in the version 3.95.3.0

Thanks and LOL... the developer has 5 different ways of informing users, and I checked 4 of them, but did not check Discord as it seems to require an account, and I didn't want to spend the time to create yet another account for yet another service I don't use. Apparently the developer didn't mention this on the other 4 ways, but did on Discord... and they still haven't updated OneCommander's changelog. Remember developers: Update changelogs! That's where people look first!

[MonolithOrchids]

Leaving his words here just in case if anyone wanna know:

Hey everyone. Sorry about the issue. I have repackaged it as 3.95.3.0.
If you don't have a functional version that can auto-update, run first onecommanderupdater.exe from program folder, which will update to older stable version, and then on next start you'll see button again to update to .3 version ahead.
Please confirm that it works again
I am doing the difference between versions and I don't see anything different, so for now my only guess that the packaging process bugged out. Especially weird since the same version as msi built 2 minutes later doesn't seem to have this issue

[Gitoffthelawn]

Leaving his words here just in case if anyone wanna know:

Hey everyone. Sorry about the issue. I have repackaged it as 3.95.3.0.
If you don't have a functional version that can auto-update, run first onecommanderupdater.exe from program folder, which will update to older stable version, and then on next start you'll see button again to update to .3 version ahead.
Please confirm that it works again
I am doing the difference between versions and I don't see anything different, so for now my only guess that the packaging process bugged out. Especially weird since the same version as msi built 2 minutes later doesn't seem to have this issue

Thank you. That's very helpful. Scoop picked up the new update an hour ago. I'll try it after I perform some package analysis.

This is a big perk of Scoop.. because the app was broken, its in-app auto-updater was broken. But Scoop is independent of such auto-updaters, and thus doesn't get affected by such Catch-22's.

It's a little concerning that the OneCommander dev doesn't know what went wrong... packaging processes don't generally just "bug out"... this is binary, people, not approximate states of reality. This is why open source is so preferential.

[Gitoffthelawn]

I tried 3.95.3.0 via Scoop, and it looks good.

On the OneCommander Google Groups site, there is now this post:

Issue on 3.95.2 update

There was an issue packaging portable version of this "version ahead" update, so if you can't run OC after the update
Run onecommanderupdater.exe from program folder (it will update to stable track version)
Then on next start you'll see button again to update to 3.95.3 version ahead.

If you are not on "version ahead" track, or using msi installer version, you shouldn't be affected.

If I'm reading this correctly (and feel free to correct me if I'm not), it sounds like Scoop is using the "version ahead" of OneCommander. To me, this sounds like an, um, unique, way of saying "alpha version" or "beta version". If so, would it be a better choice to have Scoop use a non-alpha/beta version?

If that's in everyone's best interest, how does the Scoop manifest for OneCommander need to be modified to effectuate that change?

[MonolithOrchids]

I think we're fine and should stick with the portable version.

Until a couple years ago they only had 2 versions on the website: Portable and Microsoft Store, and Portable was the recommended version, just recently that the developer introduced that .msi version.

I might be reading this wrong, but from what I understood reading on his Discord server, "version ahead" is indeed a version that is not considered as official/stable, but that the version 3.95.3 is a onetime thing and does not mean that the portable version is a "version ahead" in every update.
It does sound like the "onecommanderupdater.exe" is what make this as "version ahead". So, if for example, he makes a similar mistake but within the .msi version, he would take the same treatment as he did today with the 3.95.3 on portable, thus, .msi version having a "version ahead" from portable.

Looking some more into his Discord, I found this message from him:

MSI version is the same code as portable, just packaged in an installer.

So .msi and portable are the same and the Microsoft Store version is clearly the worst version out of it, so yeah, it is pretty much what I said, he probably only does the "version ahead" when he faces some problem with that specific version and not treating portable as a beta version.

[Gitoffthelawn]

I think we're fine and should stick with the portable version.

I definitely agree that the portable version is the way to go, especially for Scoop.

I might be reading this wrong, but from what I understood reading on his Discord server, "version ahead" is indeed a version that is not considered as official/stable, but that the version 3.95.3 is a onetime thing and does not mean that the portable version is a "version ahead" in every update. It does sound like the "onecommanderupdater.exe" is what make this as "version ahead". So, if for example, he makes a similar mistake but within the .msi version, he would take the same treatment as he did today with the 3.95.3 on portable, thus, .msi version having a "version ahead" from portable.

Ah, that's a different reading than I had, and certainly a good possibility. To paraphrase, are you thinking that "version ahead" is referring to being a version ahead of the other release types (e.g. Microsoft Store & .msi)? That actually makes sense linguistically.

If that reading is correct, do you understand how/why the portable version got a version ahead (which turned out to be a dud)?

[MonolithOrchids]

Ah, that's a different reading than I had, and certainly a good possibility. To paraphrase, are you thinking that "version ahead" is referring to being a version ahead of the other release types (e.g. Microsoft Store & .msi)? That actually makes sense linguistically.

Yeah, I'm pretty sure "version ahead" is a version ahead of regular releases, including Portable...

If that reading is correct, do you understand how/why the portable version got a version ahead (which turned out to be a dud)?

...which goes with this, he probably pushes "version ahead" if any of the regular releases has any problem, you can also find "version ahead" option on Settings > Advance, so is literally a toggle, I would guess that this option doesn't even exist/work on the Microsoft Store version since it already uses an older version compared to the others.
I will ask him on Discord if he wants to jump in here to clarify it for us.