Issues with the round method on SASDataFrame objects

[pestyld]

There seems to be an issue with the round method on a SASDataFrame object.

For example, i'll load the following file as a CAS Table

castbl = conn.read_csv(r'https://support.sas.com/documentation/onlinedoc/viya/exampledatasets/cars.csv', casout = {'name':'hmeq', 'caslib':'casuser', 'replace':True})

Then i'll execute the describe and round methods on the CASTable objects. The describe method returns a SASDataFrame, and the round method should round that DataFrame.

castbl.describe().round()
The results show it makes all the values NaN

I also tested this with the head method on a CASTable object. I get this error:
castbl.head().round()

If you use traditional Pandas and create a traditional DataFrame, the round method works each time.

df = pd.read_csv(r'https://support.sas.com/documentation/onlinedoc/viya/exampledatasets/cars.csv')
df.describe().round()
df.head().round()

[bkemper24]

This appears to be a problem with pandas. The problem was introduced in pandas version 1.2.0.
The problem occurs when a class that subclasses pandas.DataFrame ( in this case the SASDataFrame class ) contains an original property named "name" that is included in the _metadata list.

The problem is fixed in pandas version 2.0.0 that was released last month.