Previously, invoking the divergence function (and any functions that invoke the divergence function, like snapshot) required the P3 and coverage information to be stored in separate tables and connected by a key.
This setup is convenient for data drawn from a database that uses this representation already, but is inconvenient for users reading in data that directly associates a coverage string with each performance result.
This commit makes the cov DataFrame optional: if it exists, then behavior is the same as before; if it does not exist, then the library looks for coverage information stored alongside the P3 data.
Related issues
N/A
Proposed changes
Make cov optional when calling divergence.
Make cov optional when calling snapshot (which calls divergence internally).
Previously, invoking the divergence function (and any functions that invoke the divergence function, like snapshot) required the P3 and coverage information to be stored in separate tables and connected by a key.
This setup is convenient for data drawn from a database that uses this representation already, but is inconvenient for users reading in data that directly associates a coverage string with each performance result.
This commit makes the cov DataFrame optional: if it exists, then behavior is the same as before; if it does not exist, then the library looks for coverage information stored alongside the P3 data.
Related issues
N/A
Proposed changes
covoptional when callingdivergence.covoptional when callingsnapshot(which callsdivergenceinternally).