DCIC Snovault

Important

DCIC Snovault is a FORK of snovault created at the ENCODE DCC project at Stanford. Our fork supports other projects of the 4D Nucleome Data Coordination and Integration Center (4DN-DCIC). Although this software is available as open source software, its primary function is to support our layered projects, and we are not at this time able to offer any active support for other uses. In particular, this fork does not purport to supersede the original snovault. we just have a different use case that we are actively exploring.

Overview

DCIC Snovault is a JSON-LD Database Framework that serves as the backend for the 4DN Data portal and CGAP. Check out our full documentation here.

Note

This repository contains a core piece of functionality shared amongst several projects in the 4DN-DCIC. It is meant to be used internally by the DCIC team in support of Fourfront, the 4DN data portal, and at this point in time it is not expected to be useful in a standalone/plug-and-play way to others.

Installation in 4DN components

DCIC Snovault is pip installable as the dcicsnovault package with:

$ pip install dcicsnovault``

However, at the present time, the functionality it provides might only be useful in conjunction with other 4DN-DCIC components.

NOTE: If you'd like to enable Elasticsearch mapping with type=nested, set the environment variable "MAPPINGS_USE_NESTED" or set the registry setting "mappings.use_nested".

Installation for Development

Currently these are for Mac OSX using homebrew. If using linux, install dependencies with a different package manager.

Step 0: Install Xcode

Install Xcode (from App Store) and homebrew: http://brew.sh

Step 1: Verify Homebrew Itself

Verify that homebrew is working properly:

$ brew doctor

Step 2: Install Homebrewed Dependencies

Install or update dependencies:

$ brew install libevent libmagic libxml2 libxslt openssl postgresql graphviz python3
$ brew install freetype libjpeg libtiff littlecms webp  # Required by Pillow
$ brew cask install adoptopenjdk8
$ brew install elasticsearch@5.6

NOTES:

• If installation of adtopopenjdk8 fails due to an ambiguity, it should work to do this instead:

  $ brew cask install homebrew/cask-versions/adoptopenjdk8
  

• If you try to invoke elasticsearch and it is not found, you may need to link the brew-installed elasticsearch:

  $ brew link --force elasticsearch@5.6
  

• If you need to update dependencies:

  $ brew update
  $ rm -rf encoded/eggs
  

• If you need to upgrade brew-installed packages that don't have pinned versions, you can use the following. However, take care because there is no command to directly undo this effect:

  $ brew update
  $ brew upgrade
  $ rm -rf encoded/eggs
  

Step 3: Running Poetry

To locally install using versions of Python libraries that have worked before, use this:

$ poetry install

Updating dependencies

To update the version dependencies, use:

$ poetry update

This command also takes space-separated names of specific packages to update. For more information, do:

$ poetry help update

Managing poetry.lock after update

There may be situations where you do this with no intent to check in the resulting updates, but once you have checked that the updates are sound, you may wish to check the resulting poetry.lock file.

Publishing

Normally, a successful build on a tagged branch (including a branch tagged as a beta) will cause publication automatically. The process begins by obtaining the version. You might do

$ head pyproject.toml

to see the first few lines of pyproject.toml, which will contain a line like version = 100.200.300, which is the snovault version. You should prepend the letter v to that version, and create the tag and push it to the GitHub server:

$ git tag v100.200.300 $ git push origin v100.200.300

Please do NOT use some other syntax for git push that pushes all of your tags. That might pick up tags that do not belong on the server and can generally cause tag pollution. Push specifically the tag you intend to publish.

Pushing such a tag should trigger publication automatically within a few minutes.

Manual Publication

There might be rare occasions where you need to do the publication manually, but normally it is not necessary or desirable. In most cases, it will either fail or will cause the automatic publication step to fail. The main case where this is known to be needed is where publication has failed on a tagged branch for reasons other than the fact of that tag being already published (e.g., a network interruption or a premature shutdown of the GitHub Actions task). An incomplete publication on GitHub Actions cannot be easily retried, so only in that case you may need to do:

$ make publish

However, to do this command locally, you would need appropriate credentials on PyPi for such publication to succeed. As presently configured, these credentials need to be in the environment variables PYPI_USER and PYPI_PASSWORD. The script that runs if you manually attempt make publish checks that you have properly declared credentials before it attempts to publish. Note that GitHub Actions is already configured with credentials, so you do not need to worry about them if you just push a tag and let the pre-defined action do the publication.

Running tests

To run specific tests locally:

$ bin/test -k test_name

To run with a debugger:

$ bin/test --pdb

Specific tests to run locally for schema changes:

$ bin/test -k test_load_workbook

Run the Pyramid tests with:

$ bin/test