@Dennis can I get an RProgN2 Pull please?

@Dennis Pyth is configured incorrectly on TIO. In particular, the field labeled "Argument" contains the code that actually gets run, "Input" is the input, and "Code" is ignored. Here's an example demonstrating this: tio.run/nexus/pyth#@2/4/7/xf21TAA

@isaacg Hm, I just my standard language template for Pyth. tio.run/nexus/bash#@5@cWKKgX1xUpl9elFhQkFpUrF9QWZLx/z8A Looks like there's a problem with mixing a filename and extra arguments. It works just fine if there aren't any.

Pyth runs the last argument by default, that's probably why.

That would explain it, yes. I'll remove the "$@" part then.

@isaacg Done. Thanks for making me aware of this.

@ATaco Done.

@Pavel Tomorrow, hopefully.

Thanks

@Dennis @MistahFiggins and @DestructibleWatermelon ask that you ping them when you do.

Will do. Do you now what they're cooking?

@Dennis Discussion in TNB has been slow today, we've mostly been thinking of ways to approach codegolf.stackexchange.com/questions/111310/….

This lead to a CMC to solve it in Flobnar, but no one can be bothered to download the interpreter.

@Dennis Pyth setup issue on TIO: codegolf.stackexchange.com/questions/111439/…

seems arguments get treated as code.

Ah cool

@Dennis Speaking of that challenge, it would be nice to have Jolf and SOGL, when you have time.

For SOGL, input needs to be added to the end of the source code, and newlines replaced with ¶ in the source.

@Pavel Jolf wasn't written for the command line, so you'd have to talk to Conor first. I'll look into SOGL, but this doesn't sound promising.

> note: this language is very much in developement, and can, and probably will change a lot in the future, so if used, always keep the version number a program was made for.

@Pavel That will require some work. It prints things like Result: 4 and errors for some of the examples.

@ConorO'Brien ^^^

@Pavel I'm sorry, but Jolf bears many features that would be lost in the translation from browser to command line. It will therefore remain as it is.

For the record:

> the convention of the result of each program being printed after "Result: ", in the tests here, is merely a convention. What the implementation does with the result of the main Flobnar expression is outside the domain of Flobnar proper.

@Dennis Did you try anything new on the OVH test server? Seems like something broke all remote monitoring (CPU, RAM and network activity)

Not that that's a problem.

@mınxomaτ Huh. Not in the last week, no. CentOS seems to be a bigger problem for TIO than I initially thought.

Missing packages?

Mostly, yes. It took some time making 200+ languages work with Fedora 24, some more work to get from 24 to 25, and changing to CentOS would basically require starting from scratch.

I got through the whole TIO setup on C7, minus the packages I mentioned in my gist. It might actually be worth to give the compilebox approach a try, given that the container can be any OS, including Fedora with stock kernel - no matter if the cloud provider supports it.

That's the Docker thing, yes?

Ye

Hm, that would solve pretty much all distro-specific problems.

It's more complicated to explain than it actually is, especially if you are not that into different sandboxing techniques. I've got it setup on a few local test servers. We can have a quick Skype call (next week+) if you want so I can give you a quick tour through what I have in mind.

@Dennis We talked about maybe adding alphuck to Tio the other week, but then holidays/the weekend/life happened, so I just wanted to bring it up again. This was the interpreter: gitlab.com/PoroCYon/bfc/tree/master

@Chance Yes, sorry. In the last four days, I haven't been able to get much work on TIO done.

@Dennis Cool, thanks. Just wanted to confirm that it was still a possibility. :)