[Nix-dev] force-push to master

Bjørn Forsman bjorn.forsman at gmail.com
Sun Mar 9 14:22:17 CET 2014


On 9 March 2014 13:56, Bjørn Forsman <bjorn.forsman at gmail.com> wrote:
> On 9 March 2014 13:48, Bjørn Forsman <bjorn.forsman at gmail.com> wrote:
>> On 9 March 2014 13:37, Domen Kožar <domen at dev.si> wrote:
>>> Hi all,
>>>
>>> I managed to force-push master and delete a couple of commits. I have no
>>> idea how that happened, but let's first fix the mess. Does anyone commit
>>>
>>> nixpkgs/master 6cfbc75 Michael Raskin: Fix CL-Launch source link
>>>
>>> If so, please force-push it to master (check before that if any new commits
>>> are there since "fa7ef9c pythonPackages.sqlalchemy9: add"
>>>
>>> I apologize for inconveniences. Domen
>>
>> A few minutes ago I did this:
>>
>> $ git pull --rebase
>> remote: Counting objects: 12, done.
>> remote: Compressing objects: 100% (12/12), done.
>> remote: Total 12 (delta 1), reused 0 (delta 0)
>> Unpacking objects: 100% (12/12), done.
>> From https://github.com/NixOS/nixpkgs
>>  + 5729457...fa7ef9c master     -> origin/master  (forced update)
>>  + c5fbbd4...4434040 modernize_nixos_generate_config ->
>> origin/modernize_nixos_generate_config  (forced update)
>>  + c5fbbd4...4434040 refs/pull/1900/head -> origin/pr/1900  (forced update)
>> First, rewinding head to replay your work on top of it...
>>
>> Above you can see the commit hashes that origin/master moved between.
>> Your tree should still have those same commit hashes / ids (assuming
>> you didn't "git gc" just now :-) )
>
>
> I *think* this is the history that was lost (but still retrievable),
> based on "5729457...fa7ef9c master     -> origin/master  (forced
> update)" when I did git pull:
>
> $ git log --pretty=oneline 5729457...fa7ef9c
> fa7ef9c16c837a1453ef2c10d3c7092b5b12e2fd pythonPackages.sqlalchemy9: add
> 5729457d50350fae4c2a0426e7e7574cc6a2b5d4 Merge: remove poppler_0_18
> from libreoffice
> 6cfbc75bc1cea45a586a2316f860beb9d411f962 Fix CL-Launch source link
> 727634b0e4ef7d233e75eff9c2c223e3cb3165a4 BadVPN moved to GitHub
> 28e28acc36441c9966ddb4d794604ec26eaa92c7 gqrx: new package
> 24d9bc576aed512962429acd6b1da8064f398f05 gnuradio-osmosdr: new package
> b9d7780ced1ea6a921d8d75316309b60a42796c7 saleae-logic: install udev rules
> c33495b74961d82470f3d43cc7f109449f3d63f5 saleae-logic: align attr name
> with pkg name (saleaeLogic -> saleae-logic)
> 12e00b5eebeb4996c447e1e41a9c18fd6636c960 openocd: install udev rules
> f0af1e7512cf40d02b6e6232950edab5784bfb48 rtl-sdr: new package
> 73eead9f13157aa2c3b3b84457e7d982ba5220d5 libreoffice: add deps needed
> now (libXdmcp, libpthreadstubs)
> 8db9fe47fec72ebecf75f6c8a8eb2cd0864786f6 poppler_0_18: remove, only
> used by libreoffice
> 39cf7bd19a4e330b286a3ff58458ffa4f2d85b3f merge master into x-updates
> (would-be fast-forward)

Domen and I had a little chat on IRC. We first decided to do
cherry-picking, but I bailed out due to cherry-picking merge commits
isn't straight forward.

I force pushed the master branch to commit commit
5729457d50350fae4c2a0426e7e7574cc6a2b5d4 (Merge: remove poppler_0_18
from libreoffice) which has the current binary channel commit as
parent (so it was the correct history at least at some point :-) ).

And then I cherry-picked my own commit that I pushed *just* before the
first force-push.

Please check history of origin/master now and see if you're missing something.

Best regards,
Bjørn Forsman


More information about the nix-dev mailing list