This repository contains data schema and release notes deployed to neuronbridge data bucket - s3://janelia-neuronbridge-data-prod. The repository does not contain actual precomputed data for NeuronBridge.
- Root
- current.txt - pointer to the current production version
- next.txt - pointer to the next version, currently in testing
<version>
- DATA_NOTES.md - release notes for this version of the data
- config.json - data configuration (follows
schemas/DataConfig.json
) - publishedNames.txt - complete list of published names (EM and LM) included in this data version
- schemas
- DataConfig.json - JSON schema for the
config.json
file - ImageLookup.json - JSON schema for the
- PrecomputedMatches.json - JSON schema for precomputed matches in
cdsresults
andpppresults
- CustomMatches.json - JSON schema for custom search results
- DataConfig.json - JSON schema for the
- metadata
- by_body - image metadata for EM bodies
<publishedName>.json
- by_line - image metadata for LM lines
<publishedName>.json
- cdsresults - CDS match results
<imageId>.json
- pppresults - PPPM match results
<bodyId>.json
- by_body - image metadata for EM bodies
- Upload precomputed data see this repo for this step.
- Update DATA_NOTES.md
- Update config.json if needed with the new libraries and or versions
- Update next.txt to point to the corresponding version folder. Note: Only update current.txt when this is released in production
- Update data schema if there were changes to the schema
- Use sync-data.sh as explained below to synchronize the changes with the corresponding NeuronBridge data bucket
- Tag the repository using the following pattern for the tag
<data-bucket-suffix>-<data-version>
, e.g.dev-3.2.0
To sync data after everything was updated run
sh sync-data.sh <DATA_VERSION> <DATA_BUCKET_SUFFIX>
where:
- DATA_VERSION is the version prefixed with 'v' and '.' replaced with '_';
so for '3.1.0' use
v3_1_0
- DATA_BUCKET_SUFFIX is one of
dev
,prod
,val
,devpre
,prodpre