I can't find how to import AnalysisException
in PySpark so I can catch it. For example:
df = spark.createDataFrame([[1, 2], [1, 2]], ['A', 'A'])
try:
df.select('A')
except AnalysisException as e:
print(e)
Error message:
NameError: name 'AnalysisException' is not defined
You can import it here:
from pyspark.sql.utils import AnalysisException
This is shown in the error traceback like
Traceback (most recent call last):
...
File "<string>", line 3, in raise_from
pyspark.sql.utils.AnalysisException: cannot resolve ...