Package Eat on NonGNU ELPA diverged

Package Eat on NonGNU ELPA diverged

[Akib Azmain Turja]

[-- Attachment #1: Type: text/plain, Size: 645 bytes --]


I mistakenly wrote a commit message with a lot of mistakes, and
unfortunately I also pushed it immediately.  After some 10-20 minutes,
I noticed the thing and quickly fixed that commit message (but not the
files) and force pushed it.  But within that 10-20 minutes, ELPA
machinery fetch from my repository, and now the two repositories
diverged.

Is there any chance to do something on ELPA side, or do I have to do
something on my side?

-- 
Akib Azmain Turja, GPG key: 70018CE5819F17A3BBA666AFE74F0EFA922AE7F5
Fediverse: akib@hostux.social
Codeberg: akib
emailselfdefense.fsf.org | "Nothing can be secure without encryption."

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

Re: Package Eat on NonGNU ELPA diverged

[Philip Kaludercic]

Akib Azmain Turja <akib@disroot.org> writes:

> I mistakenly wrote a commit message with a lot of mistakes, and
> unfortunately I also pushed it immediately.  After some 10-20 minutes,
> I noticed the thing and quickly fixed that commit message (but not the
> files) and force pushed it.  But within that 10-20 minutes, ELPA
> machinery fetch from my repository, and now the two repositories
> diverged.
>
> Is there any chance to do something on ELPA side, or do I have to do
> something on my side?

IIRC You could add the ELPA mirror as a remote, then merge the diverged
branch onto your own and push that.

Re: Package Eat on NonGNU ELPA diverged

[Akib Azmain Turja]

[-- Attachment #1: Type: text/plain, Size: 1033 bytes --]

Philip Kaludercic <philipk@posteo.net> writes:

> Akib Azmain Turja <akib@disroot.org> writes:
>
>> I mistakenly wrote a commit message with a lot of mistakes, and
>> unfortunately I also pushed it immediately.  After some 10-20 minutes,
>> I noticed the thing and quickly fixed that commit message (but not the
>> files) and force pushed it.  But within that 10-20 minutes, ELPA
>> machinery fetch from my repository, and now the two repositories
>> diverged.
>>
>> Is there any chance to do something on ELPA side, or do I have to do
>> something on my side?
>
> IIRC You could add the ELPA mirror as a remote, then merge the diverged
> branch onto your own and push that.

Does that mean that there is really nothing doable on a server-side?  I
wish there was something.

Anyway, I'll merge if the above is true.

-- 
Akib Azmain Turja, GPG key: 70018CE5819F17A3BBA666AFE74F0EFA922AE7F5
Fediverse: akib@hostux.social
Codeberg: akib
emailselfdefense.fsf.org | "Nothing can be secure without encryption."

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

Re: Package Eat on NonGNU ELPA diverged

[Stefan Monnier]

> files) and force pushed it.  But within that 10-20 minutes, ELPA
> machinery fetch from my repository, and now the two repositories
> diverged.
>
> Is there any chance to do something on ELPA side, or do I have to do
> something on my side?

The problem goes much faster that the `nongnu.git` repository on
`git.sv.gnu.org`, because that repository is tracked by several others.
Can it be fixed?
Yes, technically, but it's a pain in the rear to go and fix them all
by hand.  It's only justified if there's a *real* problem (like the
faulty commit contains something we *really* don't want to keep).
Here, we're talking about including a commit with a suboptimal commit
message.  Nobody will suffer from it,


        Stefan

Re: Package Eat on NonGNU ELPA diverged

[Akib Azmain Turja]

[-- Attachment #1: Type: text/plain, Size: 1091 bytes --]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> files) and force pushed it.  But within that 10-20 minutes, ELPA
>> machinery fetch from my repository, and now the two repositories
>> diverged.
>>
>> Is there any chance to do something on ELPA side, or do I have to do
>> something on my side?
>
> The problem goes much faster that the `nongnu.git` repository on
> `git.sv.gnu.org`, because that repository is tracked by several others.
> Can it be fixed?
> Yes, technically, but it's a pain in the rear to go and fix them all
> by hand.  It's only justified if there's a *real* problem (like the
> faulty commit contains something we *really* don't want to keep).
> Here, we're talking about including a commit with a suboptimal commit
> message.  Nobody will suffer from it,
>
>
>         Stefan
>

Thanks for the explanation, I'll merge from the diverged branch.

-- 
Akib Azmain Turja, GPG key: 70018CE5819F17A3BBA666AFE74F0EFA922AE7F5
Fediverse: akib@hostux.social
Codeberg: akib
emailselfdefense.fsf.org | "Nothing can be secure without encryption."

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]