Optionally disable console by feature

[marcomq]

Going to fix
#5690

Optionally disable stdin, stdout and stderr for example for windows GUI apps, where there is no console and where the previous code panicked.

[youknowone]

Thanks! Did this patch solved your problem?

One change request. Rust features are expected to be additive. In other words, feature stdio is better than disable-stdio.

[marcomq]

Rust features are expected to be additive. In other words, feature stdio is better than disable-stdio.

Oh, thx. You are completely right! The docs actually mention this special case
A consequence of this is that features should be additive. That is, enabling a feature should not disable functionality, and it should usually be safe to enable any combination of features. A feature should not introduce a SemVer-incompatible change. For example, if you want to optionally support [no_std](https://doc.rust-lang.org/reference/names/preludes.html#the-no_std-attribute) environments, do not use a no_std feature. Instead, use a std feature that enables std.

Always good to learn something new :)
I will modify the PR.

[marcomq]

Tbh - I think
#5420
actually looks better.
It looks like it would allow calling "print" and just prevents writing to stdout. My version would raise an error when calling "print".

[marcomq]

I guess I will try to resolve the conflict in #5420 and check if it solves my issue.
Also, I'm probably going to name the feature quiet-stdio and it should disable the stdio if active.
I would prefer this for following reasons:

• Many projects have just included RustPython-vm and stdio would just be a standard feature of RustPython but not of the VM. In those projects, they would loose stdio.
• Some projects have already a "disable-default" and define all features that are needed manually. Those would also loose stdio.
• The "disable-default" doesn't guarantee that a feature is actually disabled. This is usually no problem, but it might cause issues here, as the stdio might be still active due to a foreign dependency.

[marcomq]

So - I tried to implement #5420 but it somehow didn't solve my issue. I still got these errors locally during initialization.
I also had a short look on the original issue - the non-UTF8 error message. But I wasn't able to locate where the error message actually originated and what the proper handling would be.

So - this is would currently be my approach of this issue.
Let me know if this is ok or if I still should change it to stdio instead of quiet-stdio

[arihant2math]

Works for me.

[coolreader18]

Sorry for the delay and for the mess that this is in general - I just merged #5420; does that approach work for your problem?