Skip to content

README.md

Latest commit

 

History

History
160 lines (112 loc) · 5.64 KB

README.md

File metadata and controls

160 lines (112 loc) · 5.64 KB
Raw

Nanopublications

This directory contains nanopublication-related files including:

  • Nanopub wave files (.trig format)
  • Hypothesis mapping files
  • Scripts for processing and canonicalizing nanopublications

Directory Structure

  • waves/: Complete set of nanopublication wave files (np_levin_nanopubs_wave*.trig, 19 files including waves A-R)
  • npi-waves/: Subset of wave files with different structure (10 files, appears to be alternative/processed versions)
  • waves_fixed/: Fixed/canonicalized versions of wave files (empty)
  • *.trig: Hypothesis mapping files and other nanopub files

Note: The batch publisher defaults to publishing waves/ only (the complete set). Use --include npi-waves/ if you want to publish those files as well, but they appear to be duplicates/alternatives of files already in waves/.

Scripts

Processing Scripts

  • np_canonicalize_slash_uris.py: Canonicalize nanopubs and normalize graph IRIs
  • nanopub_restructure.py: Restructure nanopublication files
  • planform_to_nanopubs.py: Convert PlanformDB data to nanopublications

Publishing Scripts

  • publish_all_nanopubs.py: Batch script to sign and publish all nanopublication files
  • publish_batch.sh: Convenient wrapper script for batch publishing

Utility Scripts

  • np or np_utils.py: Command-line utilities for nanopub operations
    • check <file.trig> - Check if a nanopub is valid
    • mktrusty <file.trig> [-o output] - Create trusty URI version
    • sign <file.trig> [-o output] - Sign a nanopub
    • publish <file.trig> - Publish a nanopub

Files

  • np_hypothesis_mappings_A_to_Q.trig: Hypothesis mappings
  • hypothesis_mappings_A_to_Q.trig: Alternative hypothesis mappings format

Publishing Nanopublications

Setup

Before publishing, you need to install the nanopub library and set up your profile:

pip install nanopub rdflib
python3 -m nanopub setup --newkeys  # Interactive setup - generates RSA keys and asks for ORCID

Important: Use the --newkeys flag to automatically generate RSA keys. The setup will:

  1. Ask for your ORCID iD (e.g., https://orcid.org/0000-0000-0000-0000)
  2. Ask for your name
  3. Generate RSA keys automatically
  4. Store configuration in ~/.nanopub/

Note: Make sure your virtual environment is activated before running the setup command.

See SETUP_GUIDE.md for detailed setup instructions and troubleshooting.

Batch Publishing

Important Notes:

  1. Bulk uploading is not officially supported by nanopub registries by design. The batch publishing script publishes one nanopub at a time with delays between publications. For future bulk publishing needs, consider setting up your own Registry node.

  2. Production server restrictions: The production registry (registry.petapico.org) may have stricter validation or reject publications that the test server accepts. If you encounter HTTP 400 errors on production, try:

    • Publishing to the test server first (--test)
    • Checking if the nanopub URI already exists (duplicate)
    • Contacting the registry administrators about specific rejection reasons
    • Publishing manually one at a time to identify problematic nanopubs

  3. Only signed nanopubs are accepted: All nanopubs must be signed before publishing. The script handles this automatically.

To sign and publish all nanopublication files:

# Preview what would be published (dry-run)
./publish_batch.sh --dry-run

# Publish to TEST server first (recommended)
./publish_batch.sh --test

# Publish to PRODUCTION (be careful!)
./publish_batch.sh --prod                    # Default: Petapico
./publish_batch.sh --prod-kp                 # KnowledgePixels production
./publish_batch.sh --prod-petapico           # Petapico production
./publish_batch.sh --prod-trusty             # TrustyURI production

# Publish only npi-waves files
./publish_batch.sh --test --include npi-waves/

# Test with first 5 files
./publish_batch.sh --test --limit 5

Or use the Python script directly:

python3 publish_all_nanopubs.py --publish test --include npi-waves/ waves/

The script will:

  • Find all publishable .trig files in npi-waves/ and other directories
  • Skip files already listed in a manifest
  • Sign each nanopublication (only signed nanopubs are accepted)
  • Publish each nanopublication one at a time with delays between requests
  • Track published files in separate manifests for test and production

PlanformDB Nanopubs

To generate and publish nanopubs from PlanformDB:

# Generate .trig files only
python3 planform_to_nanopubs.py --db planformDB_2.5.0.edb --out ./planform_nanopubs

# Generate and publish to test server
python3 planform_to_nanopubs.py --db planformDB_2.5.0.edb --out ./planform_nanopubs --publish test --limit 10

Then use the batch publisher to publish all generated files:

python3 publish_all_nanopubs.py --publish test --include planform_nanopubs/

Command-Line Utilities

The np script provides convenient utilities for working with nanopubs:

# Check if a nanopub file is valid
./np check file.trig

# Create trusty URI version
./np mktrusty file.trig -o trusty/file.trig

# Sign a nanopub (produces signed.file.trig)
./np sign trusty/file.trig

# Check signed nanopub
./np check signed.file.trig

# Publish a nanopub
./np publish signed.file.trig

Or use the Python script directly:

python3 np_utils.py check file.trig
python3 np_utils.py mktrusty file.trig -o trusty/file.trig
python3 np_utils.py sign trusty/file.trig
python3 np_utils.py publish signed.file.trig

Usage

See individual script files for detailed usage instructions and documentation.