Skip to content

CLAUDE.md

Latest commit

 

History

History
76 lines (53 loc) · 4.06 KB

CLAUDE.md

File metadata and controls

76 lines (53 loc) · 4.06 KB
Raw

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Overview

Operational scripts and utilities for the PDS Engineering Node (EN). Scripts automate tasks across two primary GitHub organizations: NASA-PDS (main PDS repos) and pds-data-dictionaries (Discipline LDD repos).

Setup

python3 -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"          # installs packaged code + dev tools
pip install -r requirements.txt  # additional deps for legacy scripts
pre-commit install                # install commit hook
pre-commit install --hook-type pre-push  # install pre-push hook (runs tests)

Required environment variables (set before running most scripts):

  • GITHUB_TOKEN — GitHub personal access token with repo access
  • PDSEN_OPS_TOKEN — Used by shell scripts targeting LDD repos
  • EMAIL_PASSWORD — For NSSDCA status email notifications (pds-operator@jpl.nasa.gov)

Commands

# Run all tests
pytest --verbose tests/ test/

# Run tox (full matrix + lint)
tox

# Run a single test file
pytest test/context/test_check_duplicate_identifiers.py -v

# Format packaged code
black src/

# Lint packaged code
flake8 src/

# Type-check packaged code
mypy src/

Key Scripts

Script Purpose
scripts/ldds/ldd-corral.py Generates the PDS4 data dictionaries web page and stages all Discipline LDD releases; outputs to /data/tmp/ldd-release/
scripts/ldds/update-ldd-actions.py Propagates GitHub Actions CI/CD workflow files from ldd-template to all Discipline LDD repos in pds-data-dictionaries org
scripts/ldds/prep_for_ldd_release.sh Creates release branches in all Discipline LDD repos for a given PDS4 IM version
scripts/repos/repo-corral.py Bulk-updates repos in the NASA-PDS org (e.g., propagating template changes)
scripts/repos/update_templates.sh / update_action.sh Shell helpers for repo template propagation
scripts/pds-stats.py Fetches GitHub release download metrics for PDS software tools
scripts/context/check_duplicate_identifiers.py Scans a directory of PDS4 context XML files for duplicate logical_identifier values
scripts/portal/pds-sync-api.py Thin wrapper — logic lives in src/pds/en_ops_utils/portal/pds_sync_api.py; after install use the pds-sync-api console script
scripts/sitemap/ds-view.py Sitemap/data set view utilities
scripts/solr/deprecated_solr_registry_pds3_export.py Deprecated Solr export for PDS3 registry

LDD Configuration

conf/ldds/config.yml — Controls which Discipline LDDs are included in a release. Each entry maps a GitHub repo name (e.g., ldd-img) to an optional name override and description for the generated web page. Repos not listed here but present in the pds-data-dictionaries org are still included unless they appear in each script's SKIP_REPOS list.

Architecture

The repo has a dual structure that will converge as scripts are migrated one by one:

Packaged code (src/pds/en_ops_utils/) — scripts that have been converted to a proper pds.en_ops_utils Python package. Install with pip install -e ".[dev]". Entry points are declared in setup.cfg under [options.entry_points]. Tests live under tests/pds/en_ops_utils/ mirroring the package layout.

Legacy scripts (scripts/) — standalone CLI tools not yet packaged. Each imports directly from third-party packages (github3, lxml, PyGithub, pystache, pds.ldd_manager). Install deps with pip install -r requirements.txt. Tests for these live under test/ and use sys.path manipulation to import directly from scripts/.

The ldd-corral.py script is the most complex legacy script: it uses github3 to enumerate repos, lxml to parse IngestLDD XML, pystache templates to render the output HTML, and lasso to clone/checkout branches.

Packaging files: setup.cfg (metadata + deps + entry points), pyproject.toml (build system + black config), tox.ini (test matrix + lint env), MANIFEST.in (package data).