deriva-ml 1.14.35__tar.gz → 1.14.37__tar.gz

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deriva-ml
-Version: 1.14.35
+Version: 1.14.37
 Summary: Utilities to simplify use of Dervia and Pandas to create reproducable ML pipelines
 Author-email: ISRD <isrd-dev@isi.edu>
 Requires-Python: >=3.10

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/pyproject.toml

@@ -28,6 +28,7 @@ deriva-ml-create-schema = "deriva_ml.schema_setup.create_schema:main"
 deriva-ml-alter-annotation = "deriva_ml.schema_setup.alter_annotation:main"
 deriva-ml-run-notebook = "deriva_ml.run_notebook:main"
 deriva-ml-install-kernel = "deriva_ml.install_kernel:main"
+bump-version = "deriva_ml.bump_version:main"
 
 deriva-ml-check-catalog-schema = "deriva_ml.schema.check_schema:main"
 

deriva_ml-1.14.37/src/deriva_ml/bump_version.py

@@ -0,0 +1,142 @@
+#!/usr/bin/env python3
+"""
+Release helper: seed initial semver tag if none exists, otherwise bump with bump-my-version.
+
+Usage:
+  python release.py [patch|minor|major]
+
+Env vars:
+  START   - initial version if no tag exists (default: 0.1.0)
+  PREFIX  - tag prefix (default: v)
+"""
+
+from __future__ import annotations
+
+import argparse
+import os
+import shutil
+import subprocess
+import sys
+from typing import Sequence
+
+
+def run(
+    cmd: Sequence[str], check: bool = True, capture: bool = False, quiet: bool = False
+) -> subprocess.CompletedProcess:
+    if not quiet:
+        print(f"$ {' '.join(cmd)}")
+    return subprocess.run(
+        cmd,
+        check=check,
+        capture_output=capture,
+        text=True,
+    )
+
+
+def in_git_repo() -> bool:
+    try:
+        run(["git", "rev-parse", "--is-inside-work-tree"], capture=True)
+        return True
+    except subprocess.CalledProcessError:
+        return False
+
+
+def has_commits() -> bool:
+    try:
+        run(["git", "log", "-1"], capture=True)
+        return True
+    except subprocess.CalledProcessError:
+        return False
+
+
+def latest_semver_tag(prefix: str) -> str | None:
+    # Use git's matcher to keep parity with Bash: prefix + x.y.z
+    pattern = f"{prefix}[0-9]*.[0-9]*.[0-9]*"
+    try:
+        cp = run(["git", "describe", "--tags", "--abbrev=0", "--match", pattern], capture=True, quiet=True)
+        tag = cp.stdout.strip()
+        return tag or None
+    except subprocess.CalledProcessError:
+        return None
+
+
+def seed_initial_tag(tag: str) -> None:
+    print(f"No existing semver tag found. Seeding initial tag: {tag}")
+    run(["git", "tag", tag, "-m", f"Initial release {tag}"])
+    # Push tags (ignore failure to keep parity with bash's simple flow)
+    run(["git", "push", "--tags"])
+
+
+def require_tool(name: str) -> None:
+    if shutil.which(name) is None:
+        print(f"Error: required tool '{name}' not found on PATH.", file=sys.stderr)
+        sys.exit(1)
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(
+        description="Set a new version tag for the current repository, and push to remote."
+    )
+    parser.add_argument(
+        "bump", nargs="?", default="patch", choices=["patch", "minor", "major"], help="Which semver part to bump."
+    )
+    args = parser.parse_args()
+
+    start = os.environ.get("START", "0.1.0")
+    prefix = os.environ.get("PREFIX", "v")
+
+    # Sanity checks
+    require_tool("git")
+    require_tool("uv")
+
+    if not in_git_repo():
+        print("Not a git repo.", file=sys.stderr)
+        return 1
+
+    # Ensure tags visible in shallow clones
+    try:
+        run(["git", "fetch", "--tags", "--quiet"], check=False)
+    except Exception:
+        pass  # non-fatal
+
+    if not has_commits():
+        print("No commits found. Commit something before tagging.", file=sys.stderr)
+        return 1
+
+    # Find latest semver tag with prefix
+    tag = latest_semver_tag(prefix)
+
+    if not tag:
+        seed_initial_tag(f"{prefix}{start}")
+        print(f"Seeded {prefix}{start}. Done.")
+        return 0
+
+    print(f"Bumping version: {args.bump}")
+
+    # Bump using bump-my-version via uv
+    # Mirrors: uv run bump-my-version bump $BUMP --verbose
+    try:
+        run(["uv", "run", "bump-my-version", "bump", args.bump, "--verbose"])
+    except subprocess.CalledProcessError as e:
+        print(e.stdout or "", end="")
+        print(e.stderr or "", end="", file=sys.stderr)
+        return e.returncode
+
+    # Push commits and tags
+    print("Pushing changes to remote repository...")
+    run(["git", "push", "--follow-tags"])
+
+    # Retrieve new version tag
+    try:
+        cp = run(["git", "describe", "--tags", "--abbrev=0"], capture=True)
+        new_tag = cp.stdout.strip()
+        print(f"New version tag: {new_tag}")
+    except subprocess.CalledProcessError:
+        print("Warning: unable to determine new tag via git describe.", file=sys.stderr)
+
+    print("Release process complete!")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml/execution/execution.py

@@ -360,6 +360,14 @@ class Execution:
             with Path(cfile).open("w", encoding="utf-8") as config_file:
                 json.dump(self.configuration.model_dump(), config_file)
 
+            lock_file = Path(self.configuration.workflow.git_root) / "uv.lock"
+            if lock_file.exists():
+                _ = self.asset_file_path(
+                    MLAsset.execution_metadata,
+                    lock_file,
+                    ExecMetadataType.execution_config.value,
+                )
+
             for parameter_file in self.configuration.parameters:
                 self.asset_file_path(
                     MLAsset.execution_asset,
@@ -896,9 +904,7 @@ class Execution:
                     asset_path.write_bytes(file_name.read_bytes())
 
         # Persist the asset types into a file
-        with Path(
-            asset_type_path(self._working_dir, self.execution_rid, asset_table)
-        ).open("a") as asset_type_file:
+        with Path(asset_type_path(self._working_dir, self.execution_rid, asset_table)).open("a") as asset_type_file:
             asset_type_file.write(json.dumps({file_name.name: asset_types}) + "\n")
 
         return AssetFilePath(

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml/execution/workflow.py

@@ -84,6 +84,7 @@ class Workflow(BaseModel):
     rid: RID | None = None
     checksum: str | None = None
     is_notebook: bool = False
+    git_root: Path | None = None
 
     _logger: logging.Logger = PrivateAttr(default=10)
 
@@ -125,6 +126,7 @@ class Workflow(BaseModel):
         if not self.url:
             path, self.is_notebook = Workflow._get_python_script()
             self.url, self.checksum = Workflow.get_url_and_checksum(path)
+            self.git_root = Workflow._get_git_root(path)
 
         self._logger = logging.getLogger("deriva_ml")
         return self
@@ -315,12 +317,9 @@ class Workflow(BaseModel):
             ]
             # Get the caller's filename, which is two up the stack from here.
             filename = Path(stack[-1])
-            if not filename.exists() or Workflow._in_repl():
+            if not (filename.exists() or Workflow._in_repl()):
                 # Being called from the command line interpreter.
                 filename = Path.cwd() / Path("REPL")
-            # Get the caller's filename, which is two up the stack from here.
-            else:
-                raise DerivaMLException("Looking for caller failed")  # Stack is too shallow
         return filename, is_notebook
 
     @staticmethod

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml/install_kernel.py

@@ -1,9 +1,13 @@
 # your_pkg/install_kernel.py
-import sys
 import re
+import sys
+from argparse import ArgumentParser
 from importlib import metadata
+from pathlib import Path
+
 from ipykernel.kernelspec import install as install_kernel
 
+
 def _dist_name_for_this_package() -> str:
     """
     Try to resolve the distribution name that provides this package.
@@ -20,6 +24,7 @@ def _dist_name_for_this_package() -> str:
     dist_name = dists[0] if dists else metadata.metadata(top_pkg).get("Name", top_pkg)
     return dist_name
 
+
 def _normalize_kernel_name(name: str) -> str:
     """
     Jupyter kernel directory names should be simple: lowercase, [-a-z0-9_].
@@ -28,19 +33,40 @@ def _normalize_kernel_name(name: str) -> str:
     name = re.sub(r"[^a-z0-9._-]+", "-", name)
     return name
 
+
+def _name_for_this_venv() -> str:
+    config_path = Path(sys.prefix) / "pyvenv.cfg"
+    with config_path.open() as f:
+        m = re.search("prompt *= *(?P<prompt>.*)", f.read())
+    return m["prompt"] if m else ""
+
+
 def main() -> None:
-    dist_name = _dist_name_for_this_package()               # e.g., "deriva-model-template"
-    kernel_name = _normalize_kernel_name(dist_name)         # e.g., "deriva-model-template"
+    parser = ArgumentParser()
+    parser.add_argument(
+        "--install-local",
+        action="store_true",
+        help="Create kernal in local venv directory instead of sys.prefix.",
+    )
+
+    dist_name = _name_for_this_venv()  # e.g., "deriva-model-template"
+    kernel_name = _normalize_kernel_name(dist_name)  # e.g., "deriva-model-template"
     display_name = f"Python ({dist_name})"
 
     # Install into the current environment's prefix (e.g., .venv/share/jupyter/kernels/..)
+    prefix_arg = {}
+    install_local = False
+    if install_local:
+        prefix_arg = {"prefix": sys.prefix}
+
     install_kernel(
-        user=False,                 # write under sys.prefix (the active env)
+        user=False,  # write under sys.prefix (the active env)
         kernel_name=kernel_name,
         display_name=display_name,
-        prefix=sys.prefix,
+        **prefix_arg,
     )
     print(f"Installed Jupyter kernel '{kernel_name}' with display name '{display_name}' under {sys.prefix!s}")
 
+
 if __name__ == "__main__":
-    main()
+    main()

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml/schema/create_schema.py

@@ -298,7 +298,7 @@ def initialize_ml_schema(model: Model, schema_name: str = "deriva-ml"):
             },
             {
                 "Name": "Runtime_Env",
-                "Description": "Information about the execution environment",
+                "Description": "Information about the runtime environment",
             },
             {
                 "Name": "Execution_Metadata",

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deriva-ml
-Version: 1.14.35
+Version: 1.14.37
 Summary: Utilities to simplify use of Dervia and Pandas to create reproducable ML pipelines
 Author-email: ISRD <isrd-dev@isi.edu>
 Requires-Python: >=3.10

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml.egg-info/SOURCES.txt

@@ -3,7 +3,6 @@ LICENSE
 README.md
 mkdocs.yml
 pyproject.toml
-release.sh
 uv.lock
 .github/release-drafter.yml
 .github/workflows/publish-docs.yml
@@ -42,6 +41,7 @@ docs/user-guide/install.md
 docs/user-guide/notebooks.md
 docs/user-guide/overview.md
 src/deriva_ml/__init__.py
+src/deriva_ml/bump_version.py
 src/deriva_ml/demo_catalog.py
 src/deriva_ml/feature.py
 src/deriva_ml/install_kernel.py

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/src/deriva_ml.egg-info/entry_points.txt

@@ -1,4 +1,5 @@
 [console_scripts]
+bump-version = deriva_ml.bump_version:main
 deriva-ml-alter-annotation = deriva_ml.schema_setup.alter_annotation:main
 deriva-ml-check-catalog-schema = deriva_ml.schema.check_schema:main
 deriva-ml-create-schema = deriva_ml.schema_setup.create_schema:main

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/tests/conftest.py

@@ -5,7 +5,6 @@ Pytest configuration and shared fixtures.
 import os
 
 import pytest
-from test_utils import MLCatalog, MLDatasetCatalog, create_jupyter_kernel, destroy_jupyter_kernel
 
 from deriva_ml import DerivaML
 from deriva_ml.demo_catalog import (
@@ -14,6 +13,8 @@ from deriva_ml.demo_catalog import (
     populate_demo_catalog,
 )
 
+from .test_utils import MLCatalog, MLDatasetCatalog, create_jupyter_kernel, destroy_jupyter_kernel
+
 
 @pytest.fixture(scope="session")
 def catalog_host():

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/tests/execution/test_execution.py

@@ -134,7 +134,10 @@ class TestExecution:
             }
             for a in metadata_path.entities().fetch()
         ]
-        assert 2 == len(execution_metadata)
+        assert "uv.lock" in [d["Filename"] for d in execution_metadata]
+        assert "configuration.json" in [d["Filename"] for d in execution_metadata]
+
+        assert 3 == len(execution_metadata)
 
     def test_execution_upload(self, test_ml):
         ml_instance = test_ml

{deriva_ml-1.14.35 → deriva_ml-1.14.37}/tests/test_utils.py

@@ -3,7 +3,6 @@ import shutil
 from tempfile import TemporaryDirectory
 from urllib.parse import quote as urlquote
 
-from demo_catalog import create_demo_features
 from deriva.core.datapath import DataPathException
 from ipykernel.kernelspec import install
 from jupyter_client.kernelspec import KernelSpecManager
@@ -13,6 +12,7 @@ from deriva_ml.demo_catalog import (
     DatasetDescription,
     create_demo_catalog,
     create_demo_datasets,
+    create_demo_features,
     populate_demo_catalog,
 )
 

deriva_ml-1.14.35/release.sh

@@ -1,37 +0,0 @@
-#!/usr/bin/env bash
-set -e  # Exit immediately if any command fails
-
-# Check for GitHub CLI (gh)
-#if ! command -v gh &> /dev/null; then
-#    echo "Error: GitHub CLI (gh) is not installed. Please install it and log in."
-#    exit 1
-#fi
-
-# Default version bump is patch unless specified (patch, minor, or major)
-VERSION_TYPE=${1:-patch}
-
-echo "Bumping version: $VERSION_TYPE"
-
-# Bump the version using bump-my-version.
-# This command should update version files, commit the changes, and create a Git tag.
-uv run bump-my-version bump "$VERSION_TYPE" --verbose
-
-# Push commits and tags to the remote repository.
-echo "Pushing changes to remote repository..."
-git push --follow-tags
-
-# Create a GitHub release with auto-generated release notes.
-#echo "Creating GitHub release for $NEW_TAG..."
-#gh release create "$NEW_TAG" --title "$NEW_TAG Release" --generate-notes
-
-# Build the package.
-# During the build, setuptools_scm will derive the version from Git tags.
-#echo "Building the package..."
-#uv build
-
-# Retrieve the new version tag (latest tag)
-NEW_TAG=$(git describe --tags --abbrev=0)
-echo "New version tag: $NEW_TAG"
-#uv publish  dist/*${NEW_TAG/v/}*
-
-echo "Release process complete!"