- Jun 13, 2022
-
-
Dan Villiom Podlaski Christiansen authored
Pushing to Dulwich itself is broken in recent releases; I've posted a PR that should fix that, so we assume the next release will fix it. https://github.com/jelmer/dulwich/issues/977
-
- Apr 13, 2022
-
-
Dan Villiom Podlaski Christiansen authored
* fix package name * don't refer to pip3 * don't suggest using a checkout directory The main motivation for not mentioning loading the extension from a checkout is that it doesn't fetch dependencies; users might miss out on `setuptools_scm` and `dulwich`.
-
- Apr 02, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Apr 01, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
The version we were using just broke against the latest click...
-
- Mar 25, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
- Mar 18, 2022
-
-
Dan Villiom Podlaski Christiansen authored
Git changed the exact wording and casing of the output of `--set-upstream`. This isn't important to our tests, so we suppress it.
-
Dan Villiom Podlaski Christiansen authored
-
- Mar 25, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
It's now totally unsupported, and the Alpine tests should ensure better version coverage.
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Mar 18, 2022
-
-
Dan Villiom Podlaski Christiansen authored
Not sure what to do here, so we just emit a somewhat unhelpful warning. Originally caused in cde57730faa7 which did not take history editing into account.
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
Our tests fail under Python 2.7 without this.
-
- Mar 14, 2022
-
-
Dan Villiom Podlaski Christiansen authored
We do this by moving the error handling out of git2hg.parse_gitmodules(), and adding ValueError — which Dulwich raises for invalid files — to list of caught exceptions.
-
- Dec 30, 2021
-
-
Dan Villiom Podlaski Christiansen authored
Indexing a byte string in Python 3 yields numbers...
-
- Jan 29, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
- Dec 30, 2021
-
-
Dan Villiom Podlaski Christiansen authored
Mercurial has a nice parser for that which doesn't break on weird lines. Use it.
-
- Mar 10, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Mar 04, 2022
-
-
Dan Villiom Podlaski Christiansen authored
This fixes the specific issue where we can prompt for authentication twice on push.
-
- Mar 03, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
* mark it as supported * stop testing the branches with Python 2.7
-
- Dec 30, 2021
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
This removes the need to set configure `receive.denyCurrentBranch` in Git, which is in part bad form, and also explicitly covered `test-push-to-head.t`. Pushing to a checked out branch in a non-bare repository is confusing, and leads to weird results in Git. More than anything, this is just a slight simplification.
-
Dan Villiom Podlaski Christiansen authored
This leads to unstable commit IDs.
-
- Jan 27, 2022
-
-
Dan Villiom Podlaski Christiansen authored
-
- Jan 26, 2022
-
-
Dan Villiom Podlaski Christiansen authored
The recent release failed to upload due to invalid ReStructured Text. Oops. While at it, I fixed another warning about a missing content type declaration.
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Dec 24, 2021
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
The upcoming release will be called 1.0 rather than 0.11.0. The main motivation for this is that I consider hg-git “ready”. In order for semantic versioning to make sense, something has to eventually be good enough. With the changes to user experience, I personally consider this good enough to actually use. And besides, if this isn't 1.0, when should it be? When development ceases? I've updated `CONTRIBUTING.rst` to specify an explicit compatibility policy.
-