pydata / numexpr

Fast numerical array expression evaluator for Python, NumPy, Pandas, PyTables and more
https://numexpr.readthedocs.io/en/latest/user_guide.html
MIT License
2.25k stars 212 forks source link

It is not possible to disable sanitization via `NUMEXP_SANITIZE` when using `numexpr.NumExpr` #450

Closed loichuder closed 9 months ago

loichuder commented 1 year ago

While encountering #449, I tried to get rid of it by disabling the sanitization.

Unfortunately, the code I am calling is using numexpr.NumExpr so that setting NUMEXP_SANITIZE to 0 does not work. (NUMEXP_SANITIZE is only checked when using validate.)

Is this the intended behaviour ?

27rabbitlt commented 11 months ago

It seems that sanitize works only before evaluation but not when NumExpr is created. I don't know if there is any benefit allowing NumExpr not being sanitized. I guess it's unintended.

github-actions[bot] commented 9 months ago

Message to comment on stale issues. If none provided, will not mark issues stale