Dear All,

I have encountered a rather strange behavior of Stata:

Array

Basically after obtaining a legitimate error in my python code, subsequent behavior of Stata gets affected in a weird way. See the same command findfile first doesn't produce any output, then it does. Subsequent calls to findfile will work as expected, but the first one gets affected.

I can assure that there is no manipulation to the files in between 1. and 2. above, which could affect this, furthermore, by design findfile should either result in a path visible on the screen, or an error message visible on the screen. But not silent execution.
Additionally, if instead of running the findfile command for the second time, I inspect the results with return list, I find results of earlier commands.

If I try other commands, I find that Stata selectively ignores them, here is another sequence after I obtain the error:


Array



This is Stata/MP 16.1 on Windows (Revision 21 Jan 2021).

This issue may have been fixed already in Stata 17.

Within my session I could reproduce the same problem multiple times, but I am not sure if it gets past computer reboot, Stata restart or other manipulations.

PS: another strange observed symptom is that edits to the class file seem to get ignored at this point. Stata seems to obey clear all and discard and other commands, but fully removing portions of my class file still somehow not affect it's execution, as if it is cached elsewhere.

Thank you, Sergiy