This documents the file format that is required by the new post-processing pipeline as defined in Bridge Downstream. This covers files that are present in all zip files for all assessments. Assessment-specific file formats are handled on a per-assessment basis.

metadata.json

Flat map of key-value pairs representing upload metadata.

Fields

• startDate & endDate - ISO8601 timestamps for when the assessment was started and finished on the participant’s device

• deviceInfo - Describes the participant’s device

Open Questions:

• Can you confirm the existence and format of the fields listed above?

• Are there any other keys that must be present in this file? Or is this just freeform key-value pairs?

• Is this redundant with the metadata submitted in the Upload API as documented in https://developer.sagebridge.org/articles/data/bundled_zip_file_uploads.html#call-the-upload-request-api? Are there fields in this file that are specifically expected to not be part of the Synapse annotations?

  • Note that because these aren’t part of Synapse annotations, researchers will not be able to add these fields to a view and query on them.

assessmentResult.json

Open Questions:

• Not sure what this is. Is this survey-specific? Assessment-specific? Do we expect all assessments to submit a file called assessmentResult.json?

• If validation is keyed off of filename, does this mean all assessmentResult.json will have the same schema? Or will we need assessment-specific schemas for this?

data.json

This is the result file for an Arc assessment. Schema is defined in the metadata.json file but is different depending upon the assessment.

answers.json

Flat map of key-value pairs, used for generating the CSV.

Open Questions:

• If we’re computing this map on server-side, does this still need to be defined as part of our file format?