calkit-python 0.28.3__tar.gz → 0.28.4__tar.gz

{calkit_python-0.28.3 → calkit_python-0.28.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: calkit-python
-Version: 0.28.3
+Version: 0.28.4
 Summary: Reproducibility simplified.
 Project-URL: Homepage, https://calkit.org
 Project-URL: Issues, https://github.com/calkit/calkit/issues

{calkit_python-0.28.3 → calkit_python-0.28.4}/calkit/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.28.3"
+__version__ = "0.28.4"
 
 from .core import *  # noqa: F403, I001
 from . import git  # noqa: F401

{calkit_python-0.28.3 → calkit_python-0.28.4}/calkit/cli/new.py

@@ -157,10 +157,16 @@ def new_project(
                 repo.git.commit(["-m", "Initialize DVC"])
     ck_info_fpath = os.path.join(abs_path, "calkit.yaml")
     if os.path.isfile(ck_info_fpath) and not overwrite:
-        raise_error(
+        ck_info = calkit.load_calkit_info(wdir=abs_path)
+        name = ck_info.get("name", name)
+        title = ck_info.get("title", title)
+        description = ck_info.get("description", description)
+        typer.echo(
             "Destination is already a Calkit project; "
-            "use --overwrite to continue"
+            "will use existing project info where possible"
         )
+    else:
+        ck_info = {}
     if os.path.isdir(abs_path) and os.listdir(abs_path) and repo is None:
         warn(f"{abs_path} is not empty")
     if name is None:
@@ -224,7 +230,29 @@ def new_project(
                     )
                 repo.git.add("calkit.yaml")
                 if not no_commit:
-                    repo.git.commit(["-m", "Create calkit.yaml"])
+                    repo.git.commit(
+                        ["calkit.yaml", "-m", "Create calkit.yaml"]
+                    )
+            else:
+                # Merge with existing project info in calkit.yaml
+                typer.echo("Updating existing calkit.yaml file")
+                ck_info = (
+                    dict(
+                        owner=resp["owner_account_name"],
+                        name=resp["name"],
+                        title=resp["title"],
+                        description=resp["description"],
+                        git_repo_url=resp["git_repo_url"],
+                    )
+                    | ck_info
+                )
+                with open(calkit_fpath, "w") as f:
+                    ryaml.dump(ck_info, f)
+                repo.git.add("calkit.yaml")
+                if not no_commit and repo.git.diff("--staged"):
+                    repo.git.commit(
+                        ["calkit.yaml", "-m", "Update calkit.yaml"]
+                    )
         try:
             calkit.dvc.configure_remote(wdir=abs_path)
         except Exception:

{calkit_python-0.28.3 → calkit_python-0.28.4}/docs/notebooks.md

@@ -322,4 +322,4 @@ pipeline:
         - results/param1={param1}/param2={param2}/results.csv
 ```
 
-For more information, see [pipeline iteration](/pipeline#iteration).
+For more information, see [pipeline iteration](pipeline/index.md#iteration).

{calkit_python-0.28.3 → calkit_python-0.28.4}/docs/quickstart.md

@@ -25,7 +25,7 @@ If you're using Conda for environment management,
 e.g., with an `environment.yml` file,
 you can use the `calkit new conda-env` command.
 
-Next, we can start building our [pipeline](pipeline.md).
+Next, we can start building our [pipeline](pipeline/index.md).
 Let's say we have a Jupyter notebook called `collect-data.ipynb`
 that produces raw data at `data/raw.h5`.
 We can add a pipeline stage to run this notebook in the `main` environment

calkit_python-0.28.4/docs/tutorials/github-actions.md

@@ -0,0 +1,83 @@
+# Running Calkit in GitHub Actions
+
+A project can be set up to automatically run the pipeline every time a
+change is pushed, either to the main branch or on a pull request,
+using GitHub Actions.
+The latter allows for inspection of outputs before merging into main.
+
+To get started, generate a DVC token and set it in your GitHub Actions
+secrets as `CALKIT_DVC_TOKEN`,
+either at your account or project level.
+On calkit.io, there are shortcuts on the project page for managing both
+Calkit tokens and GitHub Actions secrets:
+
+![Project quick actions links](img/quick-actions.png)
+
+![DVC token](img/latex-codespaces/new-token.png)
+
+![Actions repo secrets](img/actions-repo-secrets.png)
+
+Next, add a workflow to the project in the `.github/workflows` folder.
+For example, we can put the content below into `.github/workflows/run.yml`:
+
+```yaml
+name: Run pipeline
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+
+permissions:
+  contents: write
+
+# Make sure we only ever run one per branch so we don't have issues pushing
+# after running the pipeline
+concurrency:
+  group: calkit-run-${{ github.ref }}
+  cancel-in-progress: false
+
+jobs:
+  main:
+    name: Run
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          # For PRs, checkout the head ref to avoid detached HEAD
+          ref: ${{ github.head_ref || github.ref_name }}
+          token: ${{ secrets.GITHUB_TOKEN }}
+      - name: Configure Git credentials
+        run: |
+          git config user.name github-actions[bot]
+          git config user.email 41898282+github-actions[bot]@users.noreply.github.com
+      - name: Setup uv
+        uses: astral-sh/setup-uv@v5
+      - name: Install Calkit
+        run: uv tool install calkit-python
+      - name: Run Calkit
+        uses: calkit/run-action@v1
+        with:
+          dvc_token: ${{ secrets.CALKIT_DVC_TOKEN }}
+```
+
+This particular example installs Calkit with uv,
+meaning `uv` and `uv-venv` environment types will work without any additional
+configuration.
+Docker is also installed by default on the `ubuntu-latest` machine.
+If other environment types are used in the project,
+setup steps may be necessary for those, e.g.,
+[`setup-miniconda`](https://github.com/marketplace/actions/setup-miniconda)
+or [`install-juliaup`](https://github.com/marketplace/actions/install-juliaup).
+
+By default, the Calkit GitHub Action will run the pipeline and save results.
+This is why the workflow needs write permissions and configures Git
+credentials to act as the GitHub Actions bot.
+The workflow also limits concurrency so multiple jobs don't attempt to push
+to the same branch at the same time.
+
+It's possible to configure the action to not save results, e.g., if you
+just want to check that the pipeline can run without errors.
+See the [documentation](https://github.com/marketplace/actions/run-calkit)
+for all available options.

{calkit_python-0.28.3 → calkit_python-0.28.4}/docs/tutorials/index.md

@@ -10,3 +10,4 @@
 - [Adding a new LaTeX-based publication with its own Docker build environment](adding-latex-pub-docker.md)
 - [A reproducible workflow using Microsoft Office (Word and Excel)](office.md)
 - [Using Calkit with MATLAB](matlab.md)
+- [Automation with GitHub Actions](github-actions.md)

{calkit_python-0.28.3 → calkit_python-0.28.4}/mkdocs.yml

@@ -67,6 +67,7 @@ nav:
       - tutorials/conda-envs.md
       - tutorials/office.md
       - tutorials/matlab.md
+      - tutorials/github-actions.md
   - CLI reference: cli-reference.md
   - Help and support: help.md
 markdown_extensions: