create the revision column in the file_source db
- fill the commit column during an analysis
- manage partial report of SCM from batch – search the missing SCM information in DB
During the current sprint to remove persistence from batch side, we have temporary disabled the feature to only send SCM data of updated files to the compute report, we're sending all SCM data, even if no files have been updated.
Technical details :
- In /batch/project WS, query DB to know if a blame is missing on a file (ProjectRepositoryLoader#addFileData)
- In the compute engine, load SCM from DB when no SCM exists in the report (source was not updated)
- => this is important in order to properly compute metrics like "coverage on new code" or "technical debt ratio on new code"