Wide Check returns False for every Column in a row

[typkrft]

When I do a wide check referencing two columns, the DataFrameSchema reports that every column in my data frame for that row failed. What am I doing wrong and how should I avoid this. This stack overflow question from another user asked 7 months ago appears to be looking for the same thing.

My goal would be to denote some, single related column as the failure, or should I be going about this differently. Effectively, I am trying to keep track of invalid data in a row, so I can handle issues, instead of failing a whole row.

Example:

import pandas as pd
import pandera as pa

def custom_check(df):
    check_output: pd.Series = pd.Series(True, index=df.index)
    return check_output.where(filter_2 & filter_1, other=False)

df = pd.DataFrame({
        "name": ["BOB", "BOB"],
        "pizza": ["PEPPERONI", "CHEESE"],
        "age": [6, 7],
        "ERRORS DATA": [[],[]]
    })

schema = pa.DataFrameSchema(
    columns={
        "name": pa.Column(str),
        "pizza": pa.Column(str),
        "age": pa.Column(int)
    },
    checks=[pa.Check(custom_check)]
)

def df_validate(df: pd.DataFrame, schema: pa.DataFrameSchema) -> pd.DataFrame:
    try: 
        schema.validate(df, lazy=True)

    except pa.errors.SchemaErrors as schema_errors: 
        print("Schema errors and failure cases:")
        print(schema_errors.failure_cases)


filter_1 = df["name"] == "BOB"
filter_2 = df["pizza"].str.fullmatch("CHEESE")

df_validate(df, schema)

output:

Schema errors and failure cases:
    schema_context       column         check  check_number failure_case  index
0  DataFrameSchema         name  custom_check             0          BOB      0
1  DataFrameSchema        pizza  custom_check             0    PEPPERONI      0
2  DataFrameSchema          age  custom_check             0            6      0
3  DataFrameSchema  ERRORS DATA  custom_check             0           []      0

Update:
This seems to get me pretty close, but I don't get an index of the row where it failed.

def custom_check(data):
    for k, v in data.items():
        check_output: pd.Series = pd.Series(True, index=v.index)
        data[k] = check_output.where(v == "CHEESE", other=False)

    results = pd.concat(data.values())
    return results

schema = pa.DataFrameSchema(
    columns={
        "name": pa.Column(str),
        "pizza": pa.Column(str, pa.Check(
            custom_check, groupby="name")),
        "age": pa.Column(int)
    },
    # checks=[pa.Check(custom_check)]
)

Schema errors and failure cases:
  schema_context column         check  check_number  failure_case index
0         Column  pizza  custom_check             0         False  None

[typkrft]

This appears to be being worked on. A workaround is posted here.