Studies

Study objects are central to Mediumroast for GitHub. Studies rely on Interactions and Companies to function. At this time only the Foundation study is implemented; a future update will add support for user defined studies. To enable the Foundation study run mrcli study --init_foundation, which will create the basic structures of the study in the repository. Each of the major functions for mrcli study are described in this document.

Notice(s)

• Some of the command line options and switches may not yet be implemented; therefore, if a switch or option is not yet implemented the CLI will inform the user and then exit.
• Study deletion is not yet implemented and not present.
• The only update possible to the Foundation study is to reset its status to cause Caffeine to reprocess the study.
• Populating the Foundation study requires a run of the Mediumroast for GitHub Caffeine Machine Intelligence service. If you're interested in running the Caffeine Machine Intelligence service please contact the Mediumroast for GitHub team via Discord or email us at [hello@mediumroast.io], (mailto:hello@mediumroast.io).

Help

Prints the usage for the study sub-command and exits.

Command(s) run

• mrcli study --help

Screenshot with output

List study objects

Print out one or more Studies to the command line or an alternative output mechanism like a CSV file. Filtering can be applied to find Studies with specific attributes.

List all study objects in a table format

This is the default output when running mrcli s or mrcli study which prints a text table to STDOUT.

Command(s) run

• mrcli s
• mrcli study

Screenshot with output

List all study objects in JSON format

Output a list of study objects in properly formatted JSON to STDOUT which can be viewed, redirected to a file, or piped to another command.

Command(s) run

• mrcli s --output=json
• mrcli study --output=json

Screenshot with output

List all study objects and output to a CSV

Study data can be output in a CSV file to enable consumption in common tools like Microsoft Excel. The resulting file will be stored in $HOME/Documents directory as Mr_Studies.csv.

Command(s) run

• mrcli s --output=csv
• mrcli study --output=csv

Screenshot of commands being run

Screenshot of CSV imported into MacOS numbers

Filter study outputs

The CLI offers the ability to filter outputs by almost any study attribute. This is manifest by two switches on the study sub-command one specific to finding studies by name, --find_by_name and the other by an arbitrary attribute, --find_by_x. Note all output format options, like JSON, CSV, etc., are available when the outputs are filtered. Finally, only exact matches are supported, meaning if you want to search for a study using any attribute you have to fully provide the attribute's value (i.e., "Med" would not match "Mediumroast Study", but "Mediumroast Study" would).

Filter a study by name

To zero in on a specific study using the find by name switch is provided.

Command(s) run

• mrcli s --find_by_name="Foundation"

Screenshot with output

Filter a study by attribute

Find a specific study by a particular attribute in the example below the switch filters on the attribute status.

Command(s) run

• mrcli s --find_by_x='{"status": "0"}'

Screenshot with output

Initialize the Foundation study

Initializes the Foundation study with core attributes and data. This is the first step in setting up the Foundation study.

Command(s) run

• mrcli s --init_foundation

Screenshot with output

Report on a study

Produce a report on a study. The report provides the most important insights per company in Microsoft Excel format as of the current release. The report is stored in the $HOME/Documents directory as <study_name>.xlsx.

Command(s) run

• mrcli s --report="Foundation"

Screenshot of the report

Reset the status of a study

Reset the status of a study will cause Caffeine to reprocess the study on its next run. This is useful when the study has been updated and needs to be reprocessed.

Command(s) run

• mrcli s --reset_by_name="Foundation"

Screenshot with output