Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    emeinhardt
    @emeinhardt
    Are there any guides / is there any documentation on the best way to have anaconda python and ihaskell installed on the same machine?
    Vaibhav Sagar
    @vaibhavsagar
    @emeinhardt what aspect of it are you having trouble with?
    emeinhardt
    @emeinhardt
    Running the macos install script results in the following error
    "Cannot remove entries from nonexistent file /Users/myusername/anaconda/lib/python3.5/site-packages/easy-install.pth"
    Vaibhav Sagar
    @vaibhavsagar
    what happens if you try to install it manually using stack?
    e.g. stack install; ihaskell install?
    emeinhardt
    @emeinhardt
    err
    the manual instructions on the github page (i.e. running "stack setup" from the IHaskell folder) seems to be proceeding fine so far
    emeinhardt
    @emeinhardt
    ok. so I ran "stack setup" in the cloned IHaskell folder, "brew install zeromq", and then "stack install", but now (following manual install instructions to the next step), "ihaskell install" yields
    zsh: command not found: ihaskell
    Vaibhav Sagar
    @vaibhavsagar
    is ~/.local/bin in your $PATH?
    emeinhardt
    @emeinhardt
    yes
    `cd ~/.local/bin
    *cd ~/.local/bin;ls yields
    compile disintegrate ghc-modi hkc maple-update momiji pretty stylish-haskell density ghc-mod hakaru hlint mh normalize simplify
    emeinhardt
    @emeinhardt

    ok. I burned everything to the ground, cloned the latest commit and followed the mac installation instructions. I can open an iHaskell notebook. \o/

    Two problems:

    1. :cd doesn't seem to support directory names with spaces. (If it's relevant, I'm on macOS.)
    2. Following the demo notebook, the cell that starts with import IHaskell.Display results in the error: <interactive>:1:1: error: Failed to load interface for ‘IHaskell.Display’
    Randall Britten
    @vertexcite
    :point_up: June 9, 2017 7:16 AM
    I’m also not having success with IHaskell on Mac.
    I followed Mac install instructions. Opening IHaskell.ipynb displays an error dialog: “Kernel not found. I couldn't find a kernel matching Haskell. …”.
    Selecting “new” from the jupyter home does not list Haskell as an option.
    Randall Britten
    @vertexcite
    Got IHaskell working on my Mac :smile:
    Randall Britten
    @vertexcite
    image.png
    Randall Britten
    @vertexcite
    (Did this to compare how it felt relative to the same thing in http://haskell.do/ )
    David Banas
    @capn-freako

    I'm having to supply the

    • --extra-include-dir=/usr/local/include, and
    • --extra-lib-dir=/usr/local/lib

    options to my stack build --fast command, in order to get around the libmagic issue under MacOS X.
    Aren't those two locations fairly standard search targets, for most build processes? Why am I having to provide them explicitly?

    Liberxue
    @Liberxue

    liberxue.github.io demo https://liberxue.github.io/

    Jekyll Themes

    uiliberxue
    Alexey Skoblikov
    @almays
    Hi all. I cannot get ihaskell to show svg plots/charts. I saw some notes in github wiki on that, but somehow no luck. Modules told to be imported are not found. I may show details later, but maybe there is some more uptodate hint? Macosx.
    Eric Landuyt
    @EricLanduyt
    Hi ! I was trying to run ihaskell inside its own (isolated) Docker container, connected through TCP to a separated Jupyter Lab docker instance.
    My first problem was that ihaskell was not using the "ip" field value from the kernel connection file (received as argument), but was always binding to 127.0.0.1 (I was passing "ip":"0.0.0.0" to allow the Jupyter Lab docker to connect to the ihaskell container). I fixed this by statically replacing all "127.0.0.1" strings to "0.0.0.0" in ZeroMQ.hs, and after this Jupyter Lab was able to run the kernel properly, at least in its "Console" mode, displaying output and so on. But in "notebook" mode (similar to the classical Jupyter notebook), it doesn't display me any output.
    Finally, I tried to run ihaskell under classical Jupyter notebook, and there the output was displayed properly, though I'm still observing a "Timeout waiting for kernel_info reply" message.
    Did someone else already try ihaskell under Jupyter Lab ?
    Randall Britten
    @vertexcite
    @EricLanduyt Perhaps someone on https://gitter.im/dataHaskell/Lobby could help.
    Eric Landuyt
    @EricLanduyt
    thanks Randall, will try there. ;)
    Jacqueline S. Homan
    @jacqueline-homan

    Hello, I'm trying to get through this tutorial for Haskell and Data Analysis but the author is using jupyter, which I never used before, so I had to install that plus the IHaskell thing so I have the option in jupyter to do a Haskell notebook but it's not working for me. I don't get any options of a Haskell notebook, and am having a LOT of difficulty with getting my dev environment set up to get this working. I am unable to get it to work right in either my Linux computer or my MBP and am having a lot of difficulty (was stuck for three days unable to get anywhere with it). Can somebody help me please?

    1. I followed these intructions for installing jupyter here (and that seemed to work): https://jupyter.readthedocs.io/en/latest/install.html

    2. Then I tried doing this part to get IHaskell installed: https://github.com/gibiansky/IHaskell but that did not work for me. It failed at this step where you install this pip3 install -r requirements.txt

    and here's the results of that: https://pastebin.com/KMj3RsxB

    I can now run jupyter notebook (or ipython notebook) but I get no options for creating a Haskell notebook, only Python.

    Help me PLEASE!

    I'm not able to get this to work for me in either my Mac or my Debian Linux desktop. :'(
    Eric Landuyt
    @EricLanduyt
    @jacqueline-homan
    The requirements.txt file is in the IHaskell repo, so you have to first 'git clone' IHaskell itself, then 'pip3 install -r requirements.txt'.
    Your pastebin shows you used 'cabal', but the instructions are using 'stack', I would suggest to try it this way.
    Also, don't forget the 'ihaskell install --stack' step at the end of the installation procedure, as it's this step which will install the IHaskell kernel in practice through the configuration of some Jupyter files: without this step, the Jupyter UI don't know at all about the IHaskell kernel.
    Jacqueline S. Homan
    @jacqueline-homan
    I did all that but it was not working for me. So I tried Cabal as a last ditch effort
    I finally with the help of a friend got the problem sorted out with the requirements.txt part, and then went on to install it using haskell stack (which I also have). So then the final problem was that with stack, it needs the enrivonment variable path thing sorted. Which again, finally got that sorted.
    So now I can create a notebook w/ Haskell
    However, I can't get it to work correctly. When I was following this Lynda.com tutorial for starting a new Haskell notebook, and had to do the part where you enter import Text.CSV, it did not work for me.
    Jacqueline S. Homan
    @jacqueline-homan
    <interactive>:1:1: error: Failed to load interface for ‘Text.CVS’
    Now what?
    Jacqueline S. Homan
    @jacqueline-homan
    I wish there was a way to upload a screenshot here.
    If there is, I don't know how to do it.
    Randall Britten
    @vertexcite
    You can paste into the browser version of Gitter, but not the app (I think?)
    Arvind Devarajan
    @arvindd
    Hello, is there somebody here trying to port IHaskell to Windows? I just ambitiously wanted to check what prevents IHaskell to work on Windows, and noticed that it depends on the "unix" package: probably a big reason for not being supported on Windows. Anything else prevents IHaskell being supported on Windows?
    B. Kerim Tshimanga
    @Tshimanga
    hello all
    a question
    is there a way to connect ihaskell to a greater set of latex functions (\mathcal, \mathbb, etc.)? they don't seem to work out of the box, and the usual %load_ext latex doesn't work either
    Michał J. Gajda
    @mgajda
    @Tshimanga If You find a translation of these to HTML... Or make a pull request?
    Randall Britten
    @vertexcite
    Vaibhav Sagar
    @vaibhavsagar
    Maybe we should add the mybinder badge to the README
    Vaibhav Sagar
    @vaibhavsagar
    okay, done
    MMesch
    @MMesch
    @vaibhavsagar for the ihaskell jupyterlab extension, could we setup a npmjs.org account to be able to download and install it without the source code? If I use my own name, it is tied to me ... better would be to use ihaskell or something and also a mail address that comes with it ...
    MMesch
    @MMesch
    ah probably it works without ...
    Arvind Devarajan
    @arvindd
    I've been trying very hard to get IHaskell on Windows without installing a third-party software such as VirtualBox.
    Finally, the best I could do was to get it working with WSL on Windows 10. I used OpenSUSE Leap 15. I've documented the whole procedure here : https://blog.ramdoot.in/ihaskell-on-windows-c549e6442262. Go there to get yourself a working copy of IHaskell on Windows!
    MMesch
    @MMesch
    nice!