FAIR CLI readme

fair CLI outline #

DISCLAIMER:
The following document is largely conceptual and therefore does not represent a manual for the final interface. Statements within the following are likely to change, further details of possible changes are given throughout. Please either open an issue or pull request on the source repository raising any changes/issues.

FAIR-CLI forms the main interface for synchronising changes between your local and shared remote FAIR Data Pipeline registries, it is also used to instantiate model runs/data submissions to the pipeline.

The project is still under development with many features still to be implemented and checked. Available commands are summarised below along with their usage.

Installation #

The project makes use of Poetry for development which allows quick and easy mangement of dependencies, and provides a virtual environment exclusive to the project. Ultimately the project will be built into a pip installable module (using poetry build) meaning users will not need Poetry. You can access this environment by installing poetry:

pip install poetry

and, ensuring you are in the project repository, running:

poetry install

which will setup the virtual environment and install requirements. You can then either launch the environment as a shell using:

poetry shell

or run commands within it externally using:

poetry run <command>

Structure #

The layout of FAIR-CLI on a simplified system looks like this:

$HOME
├── .fair
│   ├── cli
│   │   ├── cli-config.yaml
│   │   └── sessions
│   ├── data
│   │   └── jobs
│   └── $REGISTRY_HOME
└─ Documents
   └─ my_project
      ├── config.yaml
      └── .fair
          ├── cli-config.yaml
          ├── logs
          └── staging

Global and Local Directories #

FAIR-CLI stores information for projects in two locations. The first is a global directory stored in the user’s home folder in the same location as the registry itself $HOME/.fair/cli, and the second is a local directory which exists within the model project itself $PROJECT_HOME/.fair.

The CLI holds metadata for the user in it’s own configuration file (not to be confused with the user modifiable config.yaml), cli-config.yaml, the global version of which is initialised during first use. In a manner similar to git, FAIR-CLI has repositories which allow the user to override these global configurations, this then forming a local variant.

Data Directory #

The directory $HOME/.fair/data is the default data store initialised by FAIR-CLI. During setup an alternative can be provided and this can be later changed on a per-run basis if the user so desires. The subdirectory $HOME/data/jobs contains timestamped directories of jobs.

Sessions Directory #

The directory $HOME/.fair/sessions is used to keep track of ongoing queries to the registry as a safety mechanism to ensure the registry is not shutdown whilst processes are still occuring.

Logs Directory #

The directory $PROJECT/.fair/logs stores stdout logs for jobs also giving information on who launched the job and how long it lasted.

Staging File #

The staging file, $PROJECT/.fair/staging, contains information of what jobs are being tracked, by default all jobs are added to this file after completion and are set to “unstaged”. Simply contains a dictionary of booleans where items for sync (staged) are marked true True and those to be held only locally False. The file uses paths relative to the local .fair folder as keys, to behave in a manner identical to git staging.

config.yaml #

This is the main file the user will interact with to customise their run. FAIR-CLI automatically generates a starter version of this file with everything in place. The only addition required is setting of either script or script_path (with the exception of running using fair run bash - see below) under run_metadata.

script
This should be a command callable by a shell for running a model/submitting data to the registry. This script is saved to a file prior to execution.
script_path
This is a direct path to an existing script to use for submission.

By default the shell used will be sh or pwsh for UNIX and Windows systems respectively, however this can be overwritten with the optional shell key which recognises the following values (where {0} is the script file):

Shell Command
bash bash -eo pipefail {0}
java java {0}
julia julia {0}
powershell powershell -command ". '{0}'"
pwsh pwsh -command ". '{0}'"
python2 python2 {0}
python3 python3 {0}
python python {0}
R R -f {0}
sh sh -e {0}
NOTE
This layout is subject to possible change depending on whether or not multiple aliases for the same user will be allowed in the registry itself. The main reason for having a local version is to support separate handling of multiple projects.

Registry Interaction #

Currently FAIR-CLI sets up the write data storage location on the local registry if it does not exist. Entries are created for the YAML file type, current user as an author, and object for a given run.

Command Line Usage #

As mentioned, all of the subcommands within FAIR-CLI are still under review with many still serving as placeholders for future features. Running fair without arguments or fair --help will show all of these.

init #

Initialises a new FAIR repository within the given directory. This should ideally be the same location as the .git folder for the current project, although setup will ask if you want to use an alternative location. The command will ask the user a series of questions which will provide metadata for tracking run authors, and also allow for the creation of a starter config.yaml.

The first time this command is launched the global CLI configuration will be populated. In subsequent calls the global will provide default suggestions towards creating the CLI configuration for the repository (local).

A repository directory matching the structure above will be placed in the current location and a starter config.yaml file will be generated (see below).

Example: First call to fair init

This example shows the process of setting up for the first time. Note the default suggestions for each prompt, in the case of Full name and Default output namespace this is the hostname of the system and an abbreviated version of this name.

$ fair init
Initialising FAIR repository, setup will now ask for basic info:

Checking for local registry
Local registry found
Remote API URL: http://data.scrc.uk/api/
Remote Data Storage Root [http://data.scrc.uk/data/]: 
Remote API Token File: $HOME/scrc_token.txt
Local API URL [http://localhost:8000/api/]: 
Local registry is offline, would you like to start it? [y/N]: y
Default Data Store:  [/home/joebloggs/.fair/data]: 
Email: jbloggs@noreply.uk
ORCID [None]: 
Full Name: Joe Bloggs
Default input namespace [None]: SCRC
Default output namespace [jbloggs]: 
Project description: Test project
Local Git repository [/home/joebloggs/Documents/AnalysisProject]: 
Git remote name [origin]: 
Using git repository remote 'origin': git@notagit.com:jbloggs/AnalysisProject.git
Initialised empty fair repository in /home/joebloggs/Documents/AnalysisProject/.fair

Example: Subsequent runs

In subsequent runs the first time setup will provide further defaults.

$ fair init
Initialising FAIR repository, setup will now ask for basic info:

Project description: Test Project
Local Git repository [/home/joebloggs/Documents/AnalysisProject]: 
Git remote name [origin]: 
Using git repository remote 'origin': git@nogit.com:joebloggs/AnalysisProject.git
Remote API URL [http://data.scrc.uk/api/]: 
Remote API Token File [/home/kristian/scrc_token.txt]: 
Local API URL [http://localhost:8000/api/]: 
Default output namespace [jbloggs]: 
Default input namespace [SCRC]: 
Initialised empty fair repository in /home/joebloggs/Documents/AnalysisProject/.fair

Generated config.yaml

run_metadata:
  default_input_namespace: SCRC
  default_output_namespace: jbloggs
  description: Test Project
  local_data_registry: http://localhost:8000/api/
  local_repo: /home/joebloggs/Documents/AnalysisProject
  write_data_store: /home/joebloggs/.fair/data

the user then only needs to add a script or script_path entry to execute a code run. This is only required for run.

run #

The purpose of run is to execute a model/submission run to the local registry. The command fills any specified template variables of the form ${{ VAR }} to match those outlined below. Outputs of a run will be stored within the coderun folder in the directory specified under the data_store tag in the config.yaml, by default this is $HOME/.fair/data/coderun.

fair run

If you wish to use an alternative config.yaml then specify it as an additional argument:

fair run /path/to/config.yaml

You can also launch a bash command directly which will then be automatically written into the config.yaml for you:

fair run --script "echo \"Hello World\""

note the command itself must be quoted as it is a single argument.

pull #

Currently pull will update any entries within the config.yaml under the register heading creating external_object and data_product objects on the registry and downloading the data to the local data storage. For example:

run_metadata:
  default_input_namespace: SCRC
  default_output_namespace: jbloggs
  description: Test project
  local_data_registry: http://localhost:8000/api/
  local_repo: /home/joebloggs/Documents/SCRC/FAIR-CLI
  write_data_store: /home/joebloggs/.fair/data
register:
- external_object: records/SARS-CoV-2/scotland/human-mortality
  namespace_name: Scottish Government Open Data Repository
  namespace_full_name: Scottish Government Open Data Repository
  namespace_website: https://statistics.gov.scot/
  root: https://statistics.gov.scot/sparql.csv?query=
  path: |-
    PREFIX qb: <http://purl.org/linked-data/cube#>
    PREFIX data: <http://statistics.gov.scot/data/>
    PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
    PREFIX dim: <http://purl.org/linked-data/sdmx/2009/dimension#>
    PREFIX sdim: <http://statistics.gov.scot/def/dimension/>
    PREFIX stat: <http://statistics.data.gov.uk/def/statistical-entity#>
    PREFIX mp: <http://statistics.gov.scot/def/measure-properties/>
    SELECT ?featurecode ?featurename ?areatypename ?date ?cause ?location ?gender ?age ?type ?count
    WHERE {
     ?indicator qb:dataSet data:deaths-involving-coronavirus-covid-19;
       mp:count ?count;
       qb:measureType ?measType;
       sdim:age ?value;
       sdim:causeOfDeath ?causeDeath;
       sdim:locationOfDeath ?locDeath;
       sdim:sex ?sex;
       dim:refArea ?featurecode;
       dim:refPeriod ?period.

       ?measType rdfs:label ?type.
       ?value rdfs:label ?age.
       ?causeDeath rdfs:label ?cause.
       ?locDeath rdfs:label ?location.
       ?sex rdfs:label ?gender.
       ?featurecode stat:code ?areatype;
         rdfs:label ?featurename.
       ?areatype rdfs:label ?areatypename.
       ?period rdfs:label ?date.
    }    
  title: Deaths involving COVID19
  description: Nice description of the dataset
  unique_name: Scottish deaths involving COVID19
  file_type: csv
  release_date: ${{DATETIME}}
  version: 0.${{DATE}}.0       
  primary: True

if run on 10/10/2021 would download the data from the given root/path URL and store in a file:

/home/joebloggs/.fair/data/records/SARS-CoV-2/scotland/human-mortality/0.20211010.0.csv

and register all required objects into the local registry.

purge #

Removes the local .fair (FAIR repository) folder by default so the user can reinitialise:

fair purge

You can remove the global configuration and start again entirely by running:

fair purge --glob

you will be asked if you wish to erase the data store, do not do this unless you intend on reinstalling the registry itself.

registry #

By default the CLI will launch the registry whenever a synchronisation or run is called. The server will only be halted once all ongoing CLI processes (in the case of multiple parallel calls) have been completed.

However the user may also specify a manual launch that will override this behaviour, instead leaving the server running constantly allowing them to view the registry in the browser.

The commands:

fair registry start

and

fair registry stop

will launch and halt the server respectively.

log #

Runs are logged locally within the local FAIR repository. A full list of runs is shown by running:

fair log

This will present a list of runs in a summary analogous to a git log call:

run 0db35c20946a1ebeaafdc3b30103cd74a57eb6b6
Author: Joe Bloggs <jbloggs@noreply.uk>
Date:   Wed Jun 30 09:09:30 2021
NOTE
The SHA for a job is not yet related to a registry code run identifier as multiple code runs can be executed within a single job.

view #

To view the stdout of a run given its SHA as shown by running fair log use the command:

fair view <sha>

you do not need to specify the full SHA but rather the first few characters:

--------------------------------
 Commenced = Wed Jun 30 09:09:30 2021 
 Author    = Joe Bloggs <jbloggs@noreply.uk>
 Namespace = jbloggs
 Command   = bash -eo pipefail /home/jbloggs/.fair/data/coderun/2021-06-30_09_09_30_721358/run_script
--------------------------------
0
1
2
3
4
5
6
7
8
9
10
------- time taken 0:00:00.011910 -------

Template Variables #

Within the config.yaml file, template variables can be specified by using the notation ${{ VAR }}, the following variables are currently recognised:

Variable Description
DATE Date in the form %Y%m%d
DATETIME Date and time in the form %Y-%m-%sT%H:%M:S
DATETIME-%Y%H%M Date and time in custom format (where %Y%H%M can be any valid form)
USER The current user as defined in the CLI
REPO_DIR The FAIR repository root directory
CONFIG_DIR The directory containing the config.yaml after template substitution
SOURCE_CONFIG Path of the user defined config.yaml
GIT_BRANCH Current branch of the git repository
GIT_REMOTE The URI of the git repository specified during setup
GIT_TAG The latest tag on git