almond-sh/almond

Lightweight Scala kernel for Jupyter / IPython 3

Lanking

@lanking520

This is tested on both Cent OS 7 and Mac with JDK 11, Scala 2.12.10, Almond (latest) + Almond Spark(0.10.9)

Joaquín Chemile

@jchemile

Hello! In which version of Jupyter Notebook Almond is worked? Because I'm facing this problem:

Traceback (most recent call last):
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\web.py", line 1699, in _execute
    result = await result
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 736, in run
    yielded = self.gen.throw(*exc_info)  # type: ignore
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\notebook\services\sessions\handlers.py", line 73, in post
    type=mtype))
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 729, in run
    value = future.result()
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 736, in run
    yielded = self.gen.throw(*exc_info)  # type: ignore
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\notebook\services\sessions\sessionmanager.py", line 79, in create_session
    kernel_id = yield self.start_kernel_for_session(session_id, path, name, type, kernel_name)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 729, in run
    value = future.result()
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 736, in run
    yielded = self.gen.throw(*exc_info)  # type: ignore
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\notebook\services\sessions\sessionmanager.py", line 92, in start_kernel_for_session
    self.kernel_manager.start_kernel(path=kernel_path, kernel_name=kernel_name)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 729, in run
    value = future.result()
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\tornado\gen.py", line 209, in wrapper
    yielded = next(result)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\notebook\services\kernels\kernelmanager.py", line 160, in start_kernel
    super(MappingKernelManager, self).start_kernel(**kwargs)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\jupyter_client\multikernelmanager.py", line 110, in start_kernel
    km.start_kernel(**kwargs)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\jupyter_client\manager.py", line 259, in start_kernel
    **kw)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\jupyter_client\manager.py", line 204, in _launch_kernel
    return launch_kernel(kernel_cmd, **kw)
  File "C:\Users\joaqchem\anaconda3\lib\site-packages\jupyter_client\launcher.py", line 138, in launch_kernel
    proc = Popen(cmd, **kwargs)
  File "C:\Users\joaqchem\anaconda3\lib\subprocess.py", line 775, in __init__
    restore_signals, start_new_session)
  File "C:\Users\joaqchem\anaconda3\lib\subprocess.py", line 1178, in _execute_child
    startupinfo)
FileNotFoundError: [WinError 2] The system cannot find the file specified

I've Python 3.7.3, Scala 2.11.12 and Almond 0.6.0

Anton Sviridov

@keynmol

I think there might be a subtle problem with Almond and 2.13.

It depends on ammonite's repl artifact:

[info] org.jline:jline-terminal:3.14.1
[info]   +-com.lihaoyi:ammonite-repl_2.13.3:2.2.0-4-4bd225e
[info]   | +-sh.almond:scala-interpreter_2.13.3:0.10.9 [S]

Which actually uses jline 3.14.1, but I think that's conflicting with Scala's own jline (which was changed to 3.15.0 in 2.13) which isn't binary compatible, so missing link barks:

[error] Category: FIELD NOT FOUND
[error]   In artifact: jline-3.15.0.jar
[error]     In class: org.jline.utils.Log
[error]       In method:  isEnabled(java.util.logging.Level):122
[error]       Access to: org.jline.utils.Log.logger
[error]       Problem: Field not found: logger
[error]       Found in: jline-terminal-3.14.1.jar

Even most recent releases of ammonite-repl use 3.14.1, which means merely updating almond won't help on 2.13

This is perhaps obscure, but just want to check with @alexarchambault that I'm understanding this correctly? And transitive exclusion might be the only way to fix it? Or silencing missinglink.

Joaquín Chemile

@jchemile

Thanks!!

Anton Sviridov

@keynmol

oh sorry @jchemile that's unrelated to your problem :)

with regards to which version - I sort of just get whatever pip3 installs, and it works. May be try using a newer almond version?

JasonVranek

@JasonVranek

Hello! I currently use the Scala embedded DSL 'chisel' to do hardware design and feel like a notebook environment lends well to this. I just had a couple questions: 1) is there a way to split a class across multiple cells? (some of these hardware modules end up being hundreds of lines). I'm currently tinkering around with implicit conversions to add methods but open to other options. 2) I am very excited about nbdev (https://github.com/fastai/nbdev) and am curious if there have been any attempts to port some of these features to scala. thanks!

Anton Sviridov

@keynmol

Always dreamt of having a job where I can use chisel :D 1) does defining multiple traits and then mixing them in in one final class work with chisel?

JasonVranek

@JasonVranek

It does! Thanks Anton, I just tried this and it does work. Much much cleaner way of achieving this. (btw chisel is awesome)

Anton Sviridov

@keynmol

Fantastic :) And I think your idea about chisel in notebooks makes a lot of sense, I should try it.

JasonVranek

@JasonVranek

Not quite my idea - they've already setup almond + chisel as an onboarding tool here you can mess around with https://github.com/freechipsproject/chisel-bootcamp but I'm unaware of anyone currently using this for more serious development

nicknn7

@nicknn7

Hi everyone, we are looking to increase the heap space available to the Scala kernel. When running "./almond --install --command "java -XX:MaxRAMPercentage=80.0 -jar almond" --copy-launcher true" on installation, the kernel.json file gets generated correctly (i.e. includes this XX:... line) but for some reason, inside the notebook, Runtime.getRuntime().maxMemory()/1000000000 always yields 32GB, no matter which RAMPercentage we pass to the installer. Any ideas what we're doing wrongly? The kernel is being installed globally. When we pass a small number, it seems to have an effect, but starting from a percentage which is above 32G, it always stays at 32GB. Any advice appreciated. Thanks!

2 replies

André L. F. Pinto

@andrelfpinto

Hello, I am trying to run this code:

case class Foo(id:Long, name:String)
val constructor = classOf[Foo].getConstructors()(0)
val args = Array[AnyRef](new java.lang.Integer(1), "Foobar")
val instance = constructor.newInstance(args:_*).asInstanceOf[Foo]

However, I am getting:

java.lang.IllegalArgumentException: wrong number of arguments
  sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
  sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
  sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
  java.lang.reflect.Constructor.newInstance(Constructor.java:423)
  ammonite.$sess.cmd1$Helper.<init>(cmd1.sc:2)
  ammonite.$sess.cmd1$.<clinit>(cmd1.sc:7)

I see that the constructor is:

java.lang.reflect.Constructor[?0] = public ammonite.$sess.cmd0$Helper$Foo(ammonite.$sess.cmd0$Helper,long,java.lang.String)

Where can I get this Helper?

Pham Nguyen

@akizminet

Hello, I'm new to this kernel. I don't know how to hide progress bar of almond-spark.

Kien Dang

@kiendang

Hi may I know how to get contextual help working? I passed both --metabrowse to the launcher and --sources --default=trueto coursier but it doesn't work for me.

Kien Dang

@kiendang

ah I was using scala 2.13.4 whereas metabrowse only supports 2.13.1

RndMnkIII

@RndMnkIII

How could I use the Jupyter widgets (IntSlider, TextBox, ...) in JupyterLab with the Almond kernel?

RndMnkIII

@RndMnkIII

drawing on canvas using javascript from Almond DisplayData API

Simon Parten

@Quafadas

Has anyone seen this before? Any hints on what I might have done wrong?
almond-sh/almond#784

Simon Parten

@Quafadas

Excitingly, I've now managed to dance through the minefield of my corporate setup, and got almond working. Yeyz...
I can't figure out how to enable "metabrowse" or any form of code completion though.

RUN ./almond --install --log info --metabrowse --id scala2_13 --display-name "Scala 2.13" --global --jupyter-path /opt/conda/share/jupyter/kernels

Should that work? Is there anything else to configure in Juypter itself?
ah almond-sh/almond#441

2 replies

Simon Parten

@Quafadas

Finally, the "variable inspector" is a fabulous tool, I have to say... however, if I add it to the kernel, then case class compilation appears to fail.

case class Something(i: Int)

cmd3.sc:20: not found: value res3_1
  .declareVariable("res3_1", res3_1); Iterator() }, 
                             ^Compilation Failed
Compilation Failed

It works wonderfully, for "fundamental" types though

Simon Parten

@Quafadas

Oh almond-sh/almond#799

Man I'm too slow to read github, sorry for spam

ZanyRumata

@ZanyRumata

hey guys!
I really stucked on how to hide progress bars in spark tasks?
Sometimes the info is really annoying.
I already tried - progressBars(false), but it seems doesn't work

okey. i just asked a question and finally found an answer.
NotebookSparkSession - doesn't work with ProgressBars (no errors, but also no effect)
AmmoniteSparkSession - allow to hide ProgressBars without any trouble.
I think it's a bug, but i'm not sure)

Simon Parten

@Quafadas

I've been trying to install the exciting 0.11.2 release ... but I get

 Cannot find default main class. Specify one with -M or --main-class.

Against the command

/tmp/cs bootstrap almond:0.11.2 --scala 2.13.4 -i user --default=true --sources -o almond

Can anyone see what might be wrong here?

Simon Parten

@Quafadas

Should someone else bump into the same, coursier docs appear to lay out a solution by adding a main class directly.

 -M almond.ScalaKernel

Simon Parten

@Quafadas

@alexarchambault To my understanding you are the protagonist behind coursier and almond. Thankyou for your work. They are wonderful.

Simon Parten

@Quafadas

When working with large variables, is there a way to view the entire variable, currently it gets truncated after some number of lines?

4 replies

Simon Parten

@Quafadas

Feedback on almond-sh/almond#813 welcome...

Anyone else tried this?

notmike

@notmike-5

Hi, I am trying to install almond for jupyter notebook running on archliunux. I get the error Cannot find default main class. Specify one with -M or --main-class. Using the suggestion of adding -M almond.ScalaKernel It appears to install and I see the scala kernel option in ipython notebook. However, when I select this kernel I get a 500 internal server error.

notmike

@notmike-5

plz to helping me :(

Simon Parten

@Quafadas

A couple of questions: would macros work in almond?

object Report {
    implicit val pickler : ReadWriter[Report] = upickle.default.macroRW[Report]
}

Should that work? I get

cmd24.sc:2: trait CaseObjectContext is a trait; does not take constructor arguments
    implicit val pickler : ReadWriter[LossReport] = upickle.default.macroRW[LossReport]

Also, I can't get tab completion to work... has anyone got it working successfully, any tips?

Simon Parten

@Quafadas

Have followed the docs here, https://jupyterlab-lsp.readthedocs.io/en/latest/Configuring.html... but can't find any diagnostics on why it isn't working :-(.

xuansontrinh

@xuansontrinh

Hello @alexarchambault , can I somehow interact with almond kernel to get some information like brief info about defined variables/classes?

21 replies

nasirkhanpak

@nasirkhanpak

I am running Jupiter notebook in visual studio code. HOW CAN I ADD ALMOND SCALA KERNEL TO JUPITER NOTEBOOK IN VISUAL STUDIO CODE. KINDLY GUIDE ME PLEASE...

4 replies

heksesang

@heksenlied:matrix.org

[m]

When I try to install almond with 2.13.4 as argument for scala version, I get Cannot find default main class. Specify one with -M or --main-class..

1 reply

heksesang

@heksenlied:matrix.org

[m]

But I can find sh.almond:scala-kernel_2.13.4 on Maven, so not sure why it should fail?

Matt Turner

@mt-digital

@heksenlied:matrix.org I got the same error on macOS Cannot find default main class. I then tried the install command example that includes version numbers and it succeeded: ./coursier launch --fork almond:0.10.0 --scala 2.12.11 -- --install

Enrique Mandujano

@EnriqueM_gitlab

Any workaround to install almond con ARM64 (Apple M1)?... I got following error exception in thread "main" java.lang.exception: unrecognized cpu architecture: x86.

Matt Turner

@mt-digital

Sorry for the delay, I do not know. Did you figure it out?

ankusharya

@ankusharya

Any idea on how to run python magic in almond?

Orellabac

@orellabac

No answers for python magic questions :( i was wondering the same. For now i will propanle just add some helpers into a library, but it will be great to have magics. I see that Toree has some support for magics

Simon Parten

@Quafadas

Does anyone know if it' spossible to set a compiler flag when running almond.

I'd love to be able to use a scala 3 library in the current release...

Where communities thrive

People

Repo info

Activity