Huge News!Announcing our $40M Series B led by Abstract Ventures.Learn More
Socket
Sign inDemoInstall
Socket

datalad-metalad

Package Overview
Dependencies
Maintainers
3
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

datalad-metalad

DataLad extension for semantic metadata handling

  • 0.4.22
  • PyPI
  • Socket score

Maintainers
3

DataLad extension for semantic metadata handling

Build_status codecov.io GitHub release PyPI version fury.io Documentation

Overview

This software is a DataLad extension that equips DataLad with an alternative command suite for metadata handling (extraction, aggregation, filtering, and reporting).

Command(s) currently provided by this extension
  • meta-extract -- run an extractor on a file or dataset and emit the resulting metadata (stdout).

  • meta-filter -- run an filter over existing metadata and return the resulting metadata (stdout).

  • meta-add -- add a metadata record or a list of metadata records (possibly received on stdin) to a metadata store, usually to the git-repo of the dataset.

  • meta-aggregate -- aggregate metadata from multiple local or remote metadata-stores into a local metadata store.

  • meta-dump -- reporting metadata from local or remote metadata stores. Allows to select metadata by file- or dataset-path matching patterns including dataset versions and dataset IDs.

  • meta conduct -- execute processing pipelines that consist of a provider which emits objects that should be processed, e.g. files or metadata, and a pipeline of processors, that perform operations on the provided objects, such as metadata-extraction and metadata-adding.Processors are usually executed in parallel. A few pipeline definitions are provided with the release.

Commands currently under development:
  • meta-export -- write a flat representation of metadata to a file-system. For now you can export your metadata to a JSON-lines file named metadata-dump.jsonl:

     datalad meta-dump -d <dataset-path> -r >metadata-dump.jsonl
    
  • meta-import -- import a flat representation of metadata from a file-system. For now you can import metadata from a JSON-lines file, e.g. metadata-dump.jsonl like this:

     datalad meta-add -d <dataset-path> --json-lines -i metadata-dump.jsonl
    
  • meta-ingest-previous -- ingest metadata from metalad<=0.2.1.

Additional metadata extractor implementations
  • Compatible with the previous families of extractors provided by datalad and by metalad, i.e. metalad_core, metalad_annex, metalad_custom, metalad_runprov

  • New metadata extractor paradigm that distinguishes between file- and dataset-level extractors. Included are two example extractors, metalad_example_dataset, and metalad_example_file

  • metalad_external_dataset and metalad_external_file, a dataset- and a file-extractors that execute external processes to generate metadata allow processing of the externally created metadata in datalad.

  • metalad_studyminimeta -- a dataset-level extractor that reads studyminimeta yaml files and produces metadata that contains a JSON-LD compatible description of the data in the input file

Indexers
  • Provides indexers for the new datalad indexer-plugin interface. These indexers convert metadata in proprietary formats into a set of key-value pairs that can be used by datalad search to search for content.

  • indexer_studyminimeta -- converts studyminimeta JSON-LD description into key-value pairs for datalad search.

  • indexer_jsonld -- a generic JSON-LD indexer that aims at converting any JSON-LD descriptions into a set of key-value pairs that reflect the content of the JSON-LD description.

Installation

Before you install this package, please make sure that you install a recent version of git-annex. Afterwards, install the latest version of datalad-metalad from PyPi. It is recommended to use a dedicated virtualenv:

# create and enter a new virtual environment (strongly recommended)
virtualenv --system-site-packages --python=python3 ~/env/datalad
. ~/env/datalad/bin/activate

# install from PyPi
pip install datalad-metalad

Support

For general information on how to use or contribute to DataLad (and this extension), please see the DataLad website or the main GitHub project page. The documentation is found here: http://docs.datalad.org/projects/metalad

All bugs, concerns and enhancement requests for this software can be submitted here: https://github.com/datalad/datalad-metalad/issues

If you have a problem or would like to ask a question about how to use DataLad, please submit a question to NeuroStars.org with a datalad tag. NeuroStars.org is a platform similar to StackOverflow but dedicated to neuroinformatics.

All previous DataLad questions are available here: http://neurostars.org/tags/datalad/

Acknowledgements

This DataLad extension was developed with support from the German Federal Ministry of Education and Research (BMBF 01GQ1905), and the US National Science Foundation (NSF 1912266).

FAQs


Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts

SocketSocket SOC 2 Logo

Product

  • Package Alerts
  • Integrations
  • Docs
  • Pricing
  • FAQ
  • Roadmap
  • Changelog

Packages

npm

Stay in touch

Get open source security insights delivered straight into your inbox.


  • Terms
  • Privacy
  • Security

Made with ⚡️ by Socket Inc