Code currently doesn't account for controlled vocabulary fields with multiple values

[SamuelJennings]

There are a number of controlled vocabulary columns that now accept multiple values separated by a semi-colon ";".

At the moment, this codebase does not take this into account. For example, there are many conditional checks in the code that take this form:

if df.loc[id,'C44'] in ["[dry measured]","[unspecified]","[other (specify in comments)]"]:
    p1 = -0.2 

If a user enters multiple values into C44 (e.g. "[dry measured];[other (specify in comments)]") this condition will fail when it should pass.

Unfortunately, there are lots of similar occurrences of this type of check which will be affected, leading to incorrect quality scores.

@Elifbp @MarineHeatFlow @GeoFux

[Elifbp]

If multiple entries are given by semicolons, the highest penalty value among them is used for the score.
Additionally, if "[unspecified]" or "[other (specify in comments)]" is one of the multiple entries, the highest penalty for the corresponding column is applied @GeoFux @MarineHeatFlow

[MarineHeatFlow]

Thats sounds correct @Elifbp.

[SamuelJennings]

Okay, well at the moment the code is not doing as expected so will need to be fixed. Something along the lines of:

value = "[dry measured];[other (specify in comments)]"
check_against = ["[dry measured]","[unspecified]","[other (specify in comments)]"]

if any(item in check_against for item in value.split(";")):
    # do something here.

This sort of condition will be required anytime a multi-value field is checked.