Add read-file action (#203)

* Add read-file

* Use YYYY.MM cache hash

* Correct inputs & outputs

* Cleanup

* Add default value for when file doesn't exist

* Convert positionals to options

* Invalidate cache based on workflows

* Correct README

* Improve caching

* Use Python >=3.8

* Initialize test suite

* Install all dependencies

* Upload coverage

* Add badges

* Update README.md

* Update test_combine_durations.py

* Add test_read_file.py

* Correct nested double quotes

* Use Python >=3.9

* Stringify

* Don't specify testpaths, allow pytest to autodetect

* Add comment regarding hashing

* Test GHA

* Debug

* Correct remote URL

* Fix quoting

* Setup python

* Standardize Python

* Enable dependabot updates for pip

* Add analyze step with automated error reports

* Minor cleanup

* Update existing failures

* Apply suggestions from code review

* Typing

* Update README.md

* Apply suggestions from code review

* Update .github/workflows/tests.yml

---------

Co-authored-by: Jannis Leidel <[email protected]>

.github/workflows/tests.yml

@@ -47,9 +47,46 @@ jobs:
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
 
+  read-file:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Source
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+
+      - name: Read Remote JSON
+        id: json
+        uses: ./read-file
+        env:
+          # `full_name` is the org/repo slug, extracting from `head` in case PR was opened via a fork
+          FULL_NAME: ${{ github.event.pull_request.head.repo.full_name || github.event.repository.full_name }}
+          # `head_ref` is the pull_request branch
+          # `ref_name` is the push/workflow_dispatch/schedule branch
+          REF: ${{ github.head_ref || github.ref_name }}
+        with:
+          path: https://raw.githubusercontent.com/${{ env.FULL_NAME }}/refs/heads/${{ env.REF }}/read-file/data/json.json
+          parser: json
+
+      - name: Read Local YAML
+        id: yaml
+        uses: ./read-file
+        with:
+          path: ./read-file/data/yaml.yaml
+          parser: yaml
+
+      - name: Setup Python
+        uses: actions/setup-python@82c7e631bb3cdc910f68e0081d67478d79c6982d # v5.1.0
+        with:
+          python-version: '>=3.9'
+
+      - name: Run Tests
+        shell: python
+        run: |
+          assert '''${{ steps.json.outputs.content }}''' == '''${{ steps.yaml.outputs.content }}'''
+          assert '''${{ fromJSON(steps.json.outputs.content)['foo'] }}''' == '''${{ fromJSON(steps.yaml.outputs.content)['foo'] }}'''
+
   # required check
   analyze:
-    needs: [pytest]
+    needs: [pytest, read-file]
     if: '!cancelled()'
     runs-on: ubuntu-latest
     steps:

pyproject.toml

@@ -3,7 +3,10 @@
 skip = '.git'
 
 [tool.coverage.report]
-exclude_also = ["if TYPE_CHECKING:"]
+exclude_also = [
+  "if TYPE_CHECKING:",
+  'if __name__ == "__main__":',
+]
 
 [tool.coverage.run]
 omit = [
@@ -15,6 +18,7 @@ omit = [
 addopts = [
   "--color=yes",
   "--cov=combine-durations",
+  "--cov=read-file",
   "--cov=template-files",
   "--cov-append",
   "--cov-branch",
@@ -46,5 +50,6 @@ select = [
 [tool.ruff.lint.isort]
 known-first-party = [
   "combine_durations",
+  "read_file",
   "template_files",
 ]

read-file/README.md

@@ -0,0 +1,52 @@
+# Read File
+
+A composite GitHub Action to read a local file or remote URL with an optional JSON/YAML parser.
+
+## Action Inputs
+
+| Name | Description | Default |
+|------|-------------|---------|
+| `path` | Local path or remote URL to the file to read. | **required** |
+| `parser` | Parser to use for the file. Choose json, yaml, or null (to leave it as plain text). | **optional** |
+| `default` | File contents to use if the file is not found. | **optional** |
+
+## Action Outputs
+
+| Name | Description |
+|------|-------------|
+| `content` | File contents as a JSON object (if a parser is specified) or the raw text. |
+
+## Sample Workflows
+
+```yaml
+name: Read File
+
+on:
+  pull_request:
+
+jobs:
+  read:
+    steps:
+      - id: read_json
+        uses: conda/actions/read-file
+        with:
+          path: https://raw.githubusercontent.com/owner/repo/ref/path/to/json.json
+          default: '{}'
+          parser: json
+
+      - id: read_yaml
+        uses: conda/actions/read-file
+        with:
+          path: https://raw.githubusercontent.com/owner/repo/ref/path/to/yaml.yaml
+          default: '{}'
+          parser: yaml
+
+      - id: read_text
+        uses: conda/actions/read-file
+        with:
+          path: path/to/text.text
+
+      - run: echo "${{ fromJSON(steps.read_json.outputs.content)['key'] }}"
+      - run: echo "${{ fromJSON(steps.read_yaml.outputs.content)['key'] }}"
+      - run: echo "${{ steps.read_file.outputs.content }}"
+```

read-file/action.yml

@@ -0,0 +1,59 @@
+name: Read File
+description: Read a local file or remote URL.
+author: Anaconda Inc.
+branding:
+  icon: book-open
+  color: green
+
+inputs:
+  path:
+    description: Local path or remote URL to the file to read.
+    required: true
+  parser:
+    description: Parser to use for the file. Choose json, yaml, or null (to leave it as plain text).
+  default:
+    description: File contents to use if the file is not found.
+outputs:
+  content:
+    description: File contents as a JSON object (if a parser is specified) or the raw text.
+    value: ${{ steps.read.outputs.content }}
+
+runs:
+  using: composite
+  steps:
+    # `hashFiles` only works on files within the working directory, since `requirements.txt`
+    # is not in the working directory we need to manually compute the SHA256 hash
+    - name: Compute Hash
+      id: hash
+      shell: bash
+      run: echo hash=$(sha256sum ${{ github.action_path }}/requirements.txt | awk '{print $1}') >> $GITHUB_OUTPUT
+
+    - name: Pip Cache
+      uses: actions/cache@0c45773b623bea8c8e75f6c82b208c3cf94ea4f9 # v4.0.2
+      with:
+        path: ~/.cache/pip
+        key: ${{ github.workflow }}-read-file-${{ steps.hash.outputs.hash }}
+
+    - name: Setup Python
+      uses: actions/setup-python@82c7e631bb3cdc910f68e0081d67478d79c6982d # v5.1.0
+      with:
+        python-version: '>=3.9'
+
+    - name: Pip Install
+      shell: bash
+      run: pip install --quiet -r ${{ github.action_path }}/requirements.txt
+
+    - name: Pip List
+      shell: bash
+      run: pip list
+
+    - name: Read JSON
+      id: read
+      shell: bash
+      run: >
+        python ${{ github.action_path }}/read_file.py
+        ${{ inputs.path }}
+        ${{ inputs.parser && format('"--parser={0}"', inputs.parser) || '' }}
+        ${{ inputs.default && format('"--default={0}"', inputs.default) || '' }}
+      env:
+        GITHUB_TOKEN: ${{ github.token }}

read-file/data/json.json

@@ -0,0 +1 @@
+{"foo": "bar"}

read-file/data/yaml.yaml

@@ -0,0 +1 @@
+foo: bar

read-file/read_file.py

@@ -0,0 +1,97 @@
+"""Read a local file or remote URL."""
+
+from __future__ import annotations
+
+import json
+import os
+from argparse import ArgumentParser
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+import requests
+import yaml
+from requests.exceptions import HTTPError, MissingSchema
+
+if TYPE_CHECKING:
+    from argparse import Namespace
+    from collections.abc import Sequence
+    from typing import Literal
+
+
+def parse_args(argv: Sequence[str] | None = None) -> Namespace:
+    # parse CLI for inputs
+    parser = ArgumentParser()
+    parser.add_argument(
+        "file",
+        type=str,
+        help="Local path or remote URL to the file to read.",
+    )
+    parser.add_argument(
+        "--parser",
+        choices=["json", "yaml"],
+        help=(
+            "Parser to use for the file. "
+            "If not specified, the file content is returned as is."
+        ),
+    )
+    parser.add_argument(
+        "--default",
+        type=str,
+        help=(
+            "Default value to use if the file is not found. "
+            "If not specified, an error is raised."
+        ),
+    )
+    return parser.parse_args(argv)
+
+
+def read_file(file: str | os.PathLike[str] | Path, default: str | None) -> str:
+    try:
+        response = requests.get(file)
+        response.raise_for_status()
+    except (HTTPError, MissingSchema):
+        # HTTPError: if the response status code is not ok
+        # MissingSchema: if the URL is not valid
+        try:
+            return Path(file).read_text()
+        except FileNotFoundError:
+            if default is None:
+                raise
+            return default
+    else:
+        return response.text
+
+
+def parse_content(content: str, parser: Literal["json", "yaml"]) -> str:
+    # if a parser is defined we parse the content and dump it as JSON
+    if parser == "json":
+        content = json.loads(content)
+        return json.dumps(content)
+    elif parser == "yaml":
+        content = yaml.safe_load(content)
+        return json.dumps(content)
+    else:
+        raise ValueError("Parser not supported.")
+
+
+def get_output(content: str) -> str:
+    return f"content<<GITHUB_OUTPUT_content\n{content}\nGITHUB_OUTPUT_content\n"
+
+
+def dump_output(content: str) -> None:
+    if output := os.getenv("GITHUB_OUTPUT"):
+        with Path(output).open("a") as fh:
+            fh.write(get_output(content))
+
+
+def main() -> None:
+    args = parse_args()
+
+    content = read_file(args.file, args.default)
+    if args.parser:
+        content = parse_content(content, args.parser)
+    dump_output(content)
+
+
+if __name__ == "__main__":
+    main()

read-file/requirements.txt

@@ -0,0 +1,2 @@
+pyyaml==6.0.2
+requests==2.32.3