Remove bugs in TableFileParser
|Reported by:||mkommend||Owned by:||gkronber|
Description (last modified by mkommend)
During modeling and importing CSV files in HeuristicLab I discovered three bugs in the current TableFileParser:
- Converting the data types of columns could change the values (test_csv_importer_conversion). The first missing values are replaced with double.NaN while the last ones with string.Empty.
- Errors in files after the look a head limit (per default 500 lines) are not reported, the progress dialog hangs and occurring exceptions are swallowed.
- test_csv_importer_conversion2: column y contains a character after line 500 -> Exception when setting y as the target variable.
- test_csv_importer_columns: row 500 has one additional value -> Exception while importing
I am unsure whether these bugs were introduced with #2071 or if they have already been present.