Merge pull request #113 from Materials-Data-Science-and-Informatics/feature/packages_support

Feature/packages support

Author: mustafasoylu

.github/workflows/ci.yml

@@ -21,18 +21,18 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install poetry
         run: pipx install poetry
-      - uses: actions/setup-python@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
 
       - name: Install poe, pre-commit and safety
         run: pip install poethepoet pre-commit safety
 
       # NOTE: using custom cache, to include pre-commit linters + deps
-      - uses: actions/cache@v3
+      - uses: actions/cache@v4
         with:
           path: |
             ~/.cache/pre-commit
@@ -54,10 +54,10 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install poetry
         run: pipx install poetry
-      - uses: actions/setup-python@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
           cache: "poetry"
@@ -78,10 +78,10 @@ jobs:
     runs-on: ${{ matrix.os }}
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install poetry
         run: pipx install poetry
-      - uses: actions/setup-python@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
           cache: "poetry"
@@ -100,10 +100,10 @@ jobs:
       contents: write
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install poetry
         run: pipx install poetry
-      - uses: actions/setup-python@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
           cache: "poetry"

.github/workflows/release.yml

@@ -27,7 +27,7 @@ jobs:
     permissions:
       contents: write # needed for creating a GH Release
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: softprops/action-gh-release@v1
 
   pypi:
@@ -37,12 +37,12 @@ jobs:
     permissions:
       id-token: write # needed for "trusted publishing" protocol
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: Install poetry
         run: pipx install poetry
 
-      - uses: actions/setup-python@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
           cache: "poetry"

CHANGELOG.md

@@ -9,6 +9,8 @@ Please consult the changelog to inform yourself about breaking changes and secur
 - make validation of output files, such as pyproject.toml, optional
 - make somesy project metadata input `version` optional
 - multiple output file support
+- enable having packages support
+- fix: package.json url set error on None value
 
 ## [v0.6.0](https://github.com/Materials-Data-Science-and-Informatics/somesy/tree/v0.6.0) <small>(2025-02-14)</small> { id="0.6.0" }
 

docs/manual.md

@@ -293,3 +293,17 @@ The example above shows a project with a `package.json` file in the root folder
 !!! note
 
     Let's remember `somesy` will create a `CITATION.CFF` and `codemeta.json` file in the root folder if it is set otherwise either by cli or by config.
+
+#### Packages (sub-modules) with their own somesy file
+
+A project with multiple packages inside can have completely different metadata such version and authors. Therefore, each package should have a separate metadata, in other words, separate `somesy` config and project metadata.
+
+```toml
+[project]
+...
+
+[config]
+packages = ['package1', 'package2', 'package3']
+```
+
+Each of these packages (sub-folders) are assumed to have separate somesy metadata in the folder. These packages could be a `somesy` supported language/framework or it could be any other language. `somesy` will create at least `CITATION.CFF` and `codemeta.json` regardless if not instructed otherwise.

src/somesy/cli/sync.py

@@ -159,6 +159,13 @@ def sync(
         "-P",
         help="Pass validation of metadata files (default: False)",
     ),
+    packages: Optional[List[Path]] = typer.Option(
+        None,
+        "--packages",
+        "-k",
+        help="Packages (subfolders) for monorepos with their own somesy config.",
+        **existing_file_arg_config,
+    ),
 ):
     """Sync project metadata input with metadata files."""
     somesy_input = resolved_somesy_input(
@@ -183,7 +190,9 @@ def sync(
         rust_file=rust_file,
         merge_codemeta=merge_codemeta,
         pass_validation=pass_validation,
+        packages=packages,
     )
+
     run_sync(somesy_input)
 
 

src/somesy/commands/sync.py

@@ -8,7 +8,8 @@
 
 from somesy.cff.writer import CFF
 from somesy.codemeta import CodeMeta
-from somesy.core.models import ProjectMetadata, SomesyInput
+from somesy.core.core import INPUT_FILES_ORDERED
+from somesy.core.models import ProjectMetadata, SomesyConfig, SomesyInput
 from somesy.core.writer import ProjectMetadataWriter
 from somesy.fortran.writer import Fortran
 from somesy.julia.writer import Julia
@@ -40,23 +41,114 @@ def _sync_file(
     logger.verbose(f"Saved synced '{file.name}'.\n")
 
 
-def _sync_files(metadata, files, writer_class, **kwargs):
+def _sync_files(
+    metadata, files, writer_class, create_if_missing: bool = False, **kwargs
+):
+    """Sync metadata to files using the provided writer.
+
+    Args:
+        metadata: Project metadata to sync
+        files: Path or list of paths to sync
+        writer_class: Writer class to use
+        create_if_missing: Whether to create the file if it doesn't exist
+        **kwargs: Additional arguments passed to the writer
+
+    """
     if isinstance(files, Path):
         files = [files]
     for file in files:
-        if file.is_file():
+        if file.is_file() or create_if_missing:
             _sync_file(metadata, file, writer_class, **kwargs)
 
 
-def sync(somesy_input: SomesyInput):
-    """Sync selected metadata files with given input file."""
+def sync(somesy_input: SomesyInput, is_package: bool = False):
+    """Sync selected metadata files with given input file.
+
+    Args:
+        somesy_input: The input configuration and metadata to sync
+        is_package: Whether this is a package (subfolder) being synced
+
+    """
     conf, metadata = somesy_input.config, somesy_input.project
 
+    # Get the base directory from the input file's location
+    try:
+        base_dir = somesy_input._origin.parent
+    except AttributeError:
+        logger.warning(
+            "No origin found for somesy input, using current working directory."
+        )
+        base_dir = Path.cwd()
+
+    # Resolve all paths in the config relative to the base directory
+    conf.resolve_paths(base_dir)
+
+    if is_package:
+        logger.info("\n[bold green]Synchronizing package metadata...[/bold green]")
+    else:
+        logger.info("\n[bold green]Synchronizing root project metadata...[/bold green]")
+
     pp_metadata = pretty_repr(metadata.model_dump(exclude_defaults=True))
     logger.debug(f"Project metadata: {pp_metadata}")
 
+    # First sync the current project
+    _sync_root_project(conf, metadata)
+
+    # Then sync each package if defined
+    if conf.packages:
+        packages = [conf.packages] if isinstance(conf.packages, Path) else conf.packages
+        for package in packages:
+            logger.info(f"\n[bold blue]Processing package {package}...[/bold blue]")
+
+            # Try all possible input files in order of priority
+            config_files = [package / file for file in INPUT_FILES_ORDERED]
+            package_input = None
+
+            for config_file in config_files:
+                try:
+                    package_input = SomesyInput.from_input_file(config_file)
+                    logger.debug(f"Found config file: {config_file}")
+                    break
+                except (FileNotFoundError, RuntimeError):
+                    continue
+
+            if package_input is None:
+                logger.warning(
+                    f"No valid somesy config found in package {package} "
+                    f"(tried: {', '.join(str(f) for f in config_files)})"
+                )
+                continue
+
+            # Create new config with CLI options and package's input file
+            cli_options = {
+                "no_sync_pyproject": conf.no_sync_pyproject,
+                "no_sync_package_json": conf.no_sync_package_json,
+                "no_sync_julia": conf.no_sync_julia,
+                "no_sync_fortran": conf.no_sync_fortran,
+                "no_sync_pom_xml": conf.no_sync_pom_xml,
+                "no_sync_mkdocs": conf.no_sync_mkdocs,
+                "no_sync_rust": conf.no_sync_rust,
+                "no_sync_cff": conf.no_sync_cff,
+                "no_sync_codemeta": conf.no_sync_codemeta,
+                "merge_codemeta": conf.merge_codemeta,
+                "pass_validation": conf.pass_validation,
+                "packages": None,  # Don't pass packages to avoid recursive package handling
+            }
+            package_input.config = SomesyConfig(input_file=config_file, **cli_options)
+
+            # Set default CFF and CodeMeta paths in package directory if not specified
+            if not package_input.config.no_sync_cff:
+                package_input.config.cff_file = Path("CITATION.cff")
+            if not package_input.config.no_sync_codemeta:
+                package_input.config.codemeta_file = Path("codemeta.json")
+
+            # Recursively call sync on the package
+            sync(package_input, is_package=True)
+
+
+def _sync_root_project(conf: SomesyConfig, metadata: ProjectMetadata):
+    """Sync metadata files for the root project."""
     # update these only if they exist:
-
     if conf.pyproject_file and not conf.no_sync_pyproject:
         _sync_files(
             metadata,
@@ -119,6 +211,7 @@ def sync(somesy_input: SomesyInput):
             metadata,
             conf.cff_file,
             CFF,
+            create_if_missing=True,
             pass_validation=conf.pass_validation,
         )
 
@@ -127,6 +220,7 @@ def sync(somesy_input: SomesyInput):
             metadata,
             conf.codemeta_file,
             CodeMeta,
+            create_if_missing=True,
             merge_codemeta=conf.merge_codemeta,
             pass_validation=conf.pass_validation,
         )

src/somesy/core/models.py

@@ -173,7 +173,7 @@ def at_least_one_target(cls, values):
     ] = False
 
     input_file: Annotated[
-        Path, Field(description="Project metadata input file path.")
+        Optional[Path], Field(description="Project metadata input file path.")
     ] = Path("somesy.toml")
 
     no_sync_pyproject: Annotated[
@@ -233,9 +233,9 @@ def at_least_one_target(cls, values):
     no_sync_codemeta: Annotated[
         bool, Field(description="Do not sync with codemeta.json.")
     ] = False
-    codemeta_file: Annotated[Path, Field(description="codemeta.json file path.")] = (
-        Path("codemeta.json")
-    )
+    codemeta_file: Annotated[
+        Union[Path, List[Path]], Field(description="codemeta.json file path.")
+    ] = Path("codemeta.json")
     merge_codemeta: Annotated[
         bool,
         Field(
@@ -249,6 +249,14 @@ def at_least_one_target(cls, values):
         Field(description="Pass validation for all output files."),
     ] = False
 
+    # packages (sub-folders) for monorepos with their own somesy config
+    packages: Annotated[
+        Optional[Union[Path, List[Path]]],
+        Field(
+            description="Packages (sub-folders) for monorepos with their own somesy config."
+        ),
+    ] = None
+
     def log_level(self) -> SomesyLogLevel:
         """Return log level derived from this configuration."""
         return SomesyLogLevel.from_flags(
@@ -272,6 +280,37 @@ def get_input(self) -> SomesyInput:
         somesy_input.config = SomesyConfig(**dct)
         return somesy_input
 
+    def resolve_paths(self, base_dir: Path) -> None:
+        """Resolve all paths in the config relative to the given base directory.
+
+        Args:
+            base_dir: The base directory to resolve paths against.
+
+        """
+
+        def resolve_path(
+            paths: Optional[Union[Path, List[Path]]],
+        ) -> Optional[Union[Path, List[Path]]]:
+            if paths is None:
+                return None
+            if isinstance(paths, list):
+                return [base_dir / p for p in paths]
+            return base_dir / paths
+
+        # Resolve all file paths
+        resolved_input = resolve_path(self.input_file)
+        self.input_file = resolved_input if isinstance(resolved_input, Path) else None
+        self.pyproject_file = resolve_path(self.pyproject_file)
+        self.package_json_file = resolve_path(self.package_json_file)
+        self.julia_file = resolve_path(self.julia_file)
+        self.fortran_file = resolve_path(self.fortran_file)
+        self.pom_xml_file = resolve_path(self.pom_xml_file)
+        self.mkdocs_file = resolve_path(self.mkdocs_file)
+        self.rust_file = resolve_path(self.rust_file)
+        self.cff_file = resolve_path(self.cff_file)
+        self.codemeta_file = resolve_path(self.codemeta_file)
+        self.packages = resolve_path(self.packages)
+
 
 # --------
 # Project metadata model (modified from CITATION.cff)
@@ -735,6 +774,13 @@ class SomesyInput(SomesyBaseModel):
         ),
     ]
 
+    # if config.input_file is set, use it as origin
+    @model_validator(mode="after")
+    def set_origin(self):
+        """Set the origin of the input file."""
+        if self.config and self.config.input_file:
+            self._origin = self.config.input_file
+
     def is_somesy_file(self) -> bool:
         """Return whether this somesy input is from a somesy config file.
 

src/somesy/package_json/writer.py

@@ -202,4 +202,7 @@ def repository(self) -> Optional[Union[str, Dict]]:
     @repository.setter
     def repository(self, value: Optional[Union[str, Dict]]) -> None:
         """Set the repository url of the project."""
-        self._set_property(self._get_key("repository"), dict(type="git", url=value))
+        if value is None:
+            self._set_property(self._get_key("repository"), None)
+        else:
+            self._set_property(self._get_key("repository"), dict(type="git", url=value))