Open VladKlimen opened 1 month ago
Hi @VladKlimen,
thanks for this PR and for signaling the issue. I was wandering if ignoring the non-utf-8 characters is a good idea in general: Python offers other ways such as replacing: https://docs.python.org/3/library/codecs.html#error-handlers. Having pieces of outputs magically disappear from the logs might not be great, maybe replacing is a better option?
@arbimo @roeger Any thoutghs on this?
Hi, I am not particularly familiar with this part of the code but in general in would agree with Andrea and favor replacing the character with UTF-8 ? (�), as supported by the python function. For me that immediately makes the problem obvious and suggests an encoding problem.
…in solve_
I tried to run some solver and the only problem was the exception that was thrown because of encoding failure of out_err_bytes, that anyway goes to logs and not critical to lack some non-unicode characters