Does anyone have an idea how to suppress extra output with Print[] when running in parallel kernels?

Sorry, extra output when running outside of a notebook (IE, wolframscript)

Immediately before the output I get: StringForm[From 1:, ParallelKernelskernel[ParallelKernelsPrivatebk[SubKernelsLocalKernelsloca‌​lKernel[SubKernelsLocalKernelsPrivatelk[LinkObject['/Applications/Mathematica.‌​app...

(just partial output)

I'm assuming that this is used to get the kernel ID into cells in the notebook interface, but breaks my (otherwise clean) output

I am trying to expand the compatibility of IGraph/M with various Linux distros. If you use an older Linux distro, or have access to one (with Mathematica) through your institutional HPC cluster, please ping me. The test I need you to do takes only 4 commands (evaluated in turn in Mathematica).

Ah, I wish I could help @Szabolcs, great package!

:)

Unfortunately I only have newer Linux distros and macos available.

But while you're here, is there a way of suppressing the output from loading the package? (IGraph/M 0.3.100 (June 26, 2018) Evaluate IGDocumentation[] to get started.)

@Richard I see this:

@Richard I was planning to remove that eventually. But before I do that, can you tell me why you want to remove it?

@Szabolcs, I'm assuming the "From KernelObject... bits is what is being translated into the big stuff I'm getting in script mode

Could it be due to woflramscript? Let me try.

@Szabolcs, actually for the same reason as my other problem, this is a script and the output is used computationally later as well

(I'm trying to do parallel computations on some pretty big graphs)

@Richard OK. If IGraph/M is used as part of another package (specifically, if IGraph/M detects a BeginPackage in the Stack) then it already does not print it. If you want to suppress it in other cases, do Block[{Print}, Needs["IGraphM`"]].

However, I might just completely remove it soon.

cool, thanks :)

I'm finding the notebook interface to get upset when there is a long running computation, so I only use it for graphics generation

I can reproduce the StringForm thing, not only with wolframscript, but also with math -script

I might come up with a super hacky alternative

I would report this to support ...

although, I'm pleased to report that IGraph/M + Mathematica is much faster in my implementation than IGraph/R + R

@Richard Here's the solution for your problem: mathematica.stackexchange.com/a/651/12

I tried it, it works.

(To suppress that ugly StringForm)

ah, thats great!, thank you

it doesn't quite suppress output altogether, but at least it gives consistent output with everything else

aha, mathematica.stackexchange.com/questions/118511/…

clever workaround

Does anyone know of a documented way to achieve the functionality of DistributeDefinitions without the "DistributeDefinitions applies itself recursively to any symbols appearing in the definitions" part?

SetSharedFunction[Print]; works perfectly for my use case

@Szabolcs, that answer is actually yours :D

It seems so. But I have no memory of writing it.

Has anyone seen on the Geo* livestrems whether 12 will support EPSG codes for projecctions?

I have now released version 1.0 of jsoupLink. I didn't change a lot, except that I added the shorthand notation for assigning attributes that I and @b3m2a1 talked about. I also started using paclets which previously I hadn't used for this project.

Since it was two years since I had last updated the package and since I didn't feel like it was missing something, I felt that it was appropriate to say that this is 1.0.

I think there may be coming functions that do things like what jsoupLink does with version 12. It seems like they are working in that direction, so this is also a good time for that reason to wrap up this project.

@C.E. looks good but the README might need an edit. See this for example: github.com/cekdahl/jSoupLink#installing-objimporter

Just a switch from OBJImporter to jsoupLink of course

@b3m2a1 Thank you for telling me this, fixed now.

I might say also that I introduced the repo structure that @Kuba showed me for the wp package. Very nice.