by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jul 07 20:03
    MichaelMure commented #419
  • Jul 07 19:39

    MichaelMure on code-analysis

    (compare)

  • Jul 07 19:39

    MichaelMure on master

    fix github action Merge pull request #422 from Mi… (compare)

  • Jul 07 19:37

    MichaelMure on code-analysis

    fix github action (compare)

  • Jul 07 19:35

    MichaelMure on code-analysis

    fix github action (compare)

  • Jul 07 19:30

    MichaelMure on code-analysis

    (compare)

  • Jul 07 19:30

    MichaelMure on master

    code analysis Merge pull request #421 from Mi… (compare)

  • Jul 07 19:28

    MichaelMure on code-analysis

    code analysis (compare)

  • Jul 07 19:15

    MichaelMure on code-analysis

    code analysis (compare)

  • Jul 07 19:14

    MichaelMure on code-analysis

    code analysis (compare)

  • Jul 07 18:50

    MichaelMure on code-analysis

    code anlysis (compare)

  • Jul 07 18:40
    sornas commented #420
  • Jul 07 18:16
    MichaelMure commented #420
  • Jul 07 18:15

    MichaelMure on master

    commands: fix segfault with bad… (compare)

  • Jul 06 04:32
    bqv commented #315
  • Jul 05 23:51
    bqv commented #315
  • Jul 05 23:39
    bqv commented #315
  • Jul 03 17:49
    MichaelMure commented #417
  • Jul 03 17:47

    MichaelMure on remove-legacy-identity

    identity: remove support for le… (compare)

  • Jul 03 17:05

    MichaelMure on master

    repo: fix test chocking on rand… (compare)

Adrien D.
@dtrckd
I tried with the last commit, v0.7.1 and v0.7.0 from the git repo. I still get the same error when typing git bug bridge (the second one, installation is ok) . When doing git bug bridge configure, the prompt launch, by when entering the project name, the same error occurs.
Adrien D.
@dtrckd
Finally, I solved it by removing the git-bug config in ./.git/config, the following config was making git-bug panicing (I shuffle the IDs):
[git-bug]
    identity = 03659aeb4abf08er78353828f3g2168f8ba48
[git-bug "bridge.my_name"]
    target = gitlab
    project-id = 1
    base-url = https://mysite.site/
    default-login =
Adrien D.
@dtrckd
Hi again :),
Is there a known status for the feature that consist of updating an issue that were modified on remote gitlab/github (e.g a added comment) and already imported, with git bug pull ?
ps: I apology if I am over posting here, tell me if i am causing tensions.
Michael Muré
@MichaelMure
no worries
so, the new version is not backward compatible regarding the bridge configuration, hence why you needed to remove the config and do it again
once you imported a bug through a bridge, you can import again and get all the new updates on the the remote tracker
you can also edit the bug locally, and push to the bugtracker, it will replicate your changes there
you can also push your bugs in a git remote, but be aware that you can have trouble if more than one person does the import from the tracker
git-bug is not aware that the remote tracker is imported multiple times from different machines and you will have conflicts when those version are merged together
Adrien D.
@dtrckd
@MichaelMure thanks for the clarification. And when we do git bug bridge pull several times, weI got duplicate issues.. this is the expected behavior right ? Is there any plan/strategy for merging imported with local issues ?
Michael Muré
@MichaelMure
it should not happen. If you import multiple times, the bridges should see that a bug already exist locally and update it if necessary.
the problem is if that import happen in a different machine with a different bug state
Adrien D.
@dtrckd
I believe it happens in my case (single machine, single remote (gitlab).
Adrien D.
@dtrckd
..and an other thing (not sure if its related) when I remove a bug (as in #66), it get removed locally, but when I push then pull, the local repo get back the issue I deleted.
Michael Muré
@MichaelMure
The bug is still present on the remote so you pull it back again
Adrien D.
@dtrckd
@MichaelMure and can we delete a bug on the remote? (or if not, how can we completely clear the remote ?)
Michael Muré
@MichaelMure
Yeah, look at the makefile, there is commands to clean everything
Michael Muré
@MichaelMure
@dtrckd the new go-gitlab version used as a dependency has an API change that break git-bug. I think your go get command got that version instead of the one specified in the go.mod file. Would you have go modules support turned off for some reason?
Adrien D.
@dtrckd
I checked the version of go-gitlab and indeed, I got the v0.31, my GO11MODULE was unset, I don't know why I got this version. Anyway, I removed it, and reinstall git-bug (0.7.1). Then, I used the command from the makefile to remove all references before running git bug bridge pull. It imported the issue.
But unfortunately, when I re-enter the command, it creates duplicates issues every times . Furthermore, it seems that created issues (with git-bug add) are not exported with git bug bridge push. (I got a exported 0 issues with my_bridge bridge) .
Gitlab version where that occurs: GitLab 12.9.2
(it works fine with Github.)
ps: go-gitlab is at v.0.29
Michael Muré
@MichaelMure
And you have those problems starting with a clean repo as well ?
That might be a bug if so
Michael Muré
@MichaelMure
Can you check that and open an issue if it's still a problem?
Sorry you have been dealing with this for 3 days now >_<
Adrien D.
@dtrckd
No worries, I still think git-bug is a great piece of code :)
I've tested on a fresh repo and summarized the results here: #366
As i mention in the issue, the fact that is a self hosted gitlab instance may play a role in this bug...
mdv
@melnikovdv
Hi. Does GPL license somehow limits usage in non open-source projects? As I see it, GPL is about sources and not about binary usage. Am I right?
matrixbot
@matrixbot
strk GPL is about distribution
strk mostly
strk the license allows YOU a number of usages
strk but ENFORCES you to grant the same rights to people you further distribute the software to
Michael Muré
@MichaelMure
GPL absolutely allows you to use git-bug for whatever you want including private project, the same way you can use Linux or GCC (both are under GPL)
the main thing GPL prevent is for a company to take git-bug, make its own version and distribute/sell that without making the source code public as well
also note that the GPL applies to git-bug itself, absolutely not on the bugs created/managed by git-bug
those are yours and yours only
mdv
@melnikovdv
😂 Thank you
Michael Muré
@MichaelMure
git-bug is in github's trending again :)
Sladyn
@sladyn98
@MichaelMure that is absolutely awesome :0
Michael Muré
@MichaelMure
thanks :)
also featured in a podcast in russian: https://radio-t.com/p/2020/04/11/podcast-697/
Alex Băluț
@aleb

@MichaelMure I added these commands to list/add/remove keys:

git-bug key [<user-id>]
git-bug key add [<user-id>] [flags]
git-bug key rm <key-id> [<user-id>]

Do you think it would be better with some other commands?

Alex Băluț
@aleb
Regarding using them.. what's the simplest functionality you'd want in git-bug?
Michael Muré
@MichaelMure
@aleb hey sorry I completely missed your messages
I'll reply directly in the PR
Alex Băluț
@aleb
Can it happen for one of the commits in git log <identity-id> to be a merge commit, to have more than one parent?
Michael Muré
@MichaelMure
@aleb not from git-bug, it should be a purely linear chain
I'm thinking on the side about how to migrate over a data model that support branches (mainly for bugs) to get a better conflict resolution but it's far from done
mitchyme
@mitchyme
Hey folks, had a small question. What was the reasoning for storing the data in internal git trees and not as objects that are stored as regular files in the repo, maybe in a separate branch? Feels like they should be equivalent and will be less blackbox-ish to users
Michael Muré
@MichaelMure
In short, to not conflict with the normal code workflow
mitchyme
@mitchyme
Makes sense, but separate branch should cover that too right? Is there any technical reason why it wouldn't? (Not saying the current one is wrong, just want to understand some of the reasoning)