1) Using the sfi module to display messages instead of print statements
2) Outputting stata error messages and exiting with stata exit codes instead of raising python exceptions. This means that for caught exceptions, the user sees a familiar looking stata message and exit code, rather than an unexpected python traceback.
3) Checking and exiting with stata's usual "dataset has changed" error message if we try to read an arrow file and there is data already in the stata dataset (this happens if you try to load an arrow file without clearing the stata dataset first); this should be expected behaviour for stata users. The import delimited command for importing csv files works the same way,
4) Adds a verbosity arg so users can choose to suppress the output of the arrowload command
Makes things nicer for users by:
1) Using the sfi module to display messages instead of print statements 2) Outputting stata error messages and exiting with stata exit codes instead of raising python exceptions. This means that for caught exceptions, the user sees a familiar looking stata message and exit code, rather than an unexpected python traceback. 3) Checking and exiting with stata's usual "dataset has changed" error message if we try to read an arrow file and there is data already in the stata dataset (this happens if you try to load an arrow file without clearing the stata dataset first); this should be expected behaviour for stata users. The
import delimitedcommand for importing csv files works the same way, 4) Adds averbosityarg so users can choose to suppress the output of the arrowload command