https://github.com/marianoguerra/pynerl

run an embeded python interpreter in erlang
https://github.com/marianoguerra/pynerl

Last synced: 3 months ago
JSON representation

run an embeded python interpreter in erlang

• Host: GitHub
• URL: https://github.com/marianoguerra/pynerl
• Owner: marianoguerra
• Created: 2010-05-18T23:38:49.000Z (about 14 years ago)
• Default Branch: master
• Last Pushed: 2012-08-09T07:52:06.000Z (almost 12 years ago)
• Last Synced: 2024-01-17T04:01:01.913Z (5 months ago)
• Language: C
• Homepage:
• Size: 103 KB
• Stars: 21
• Watchers: 4
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.rst

Lists

• AwesomeInterpreter - pynerl

README

        
pynerl

======

status

------

I would really like to finish this, but for now it's on hold, if you would like to help, fork it and send pull requests.

about

-----

pynerl is a set of NIFs that allow to run python code from erlang

the NIFs embed python and allow to run eval and call functions.

pynerl does the transformation of data types.

an example session::

        1> pynerl:eval("t = 1", "t").

        1

        2> pynerl:eval("t = 1 * 2", "t").

        2

        3> pynerl:eval("import time;t = time.time()", "t").            

        1274236802.877999

        4> pynerl:eval("import random;t = random.random()", "t").         

        0.45102117275294684

        5> pynerl:eval("t = print('hello erlang or python')", "t").                                 

        hello erlang or python

                              none

        6> pynerl:call("time", "time", []).                        

        1274236859.510859

        7> pynerl:call("random", "random", []).

        0.9623136682858975

        8> pynerl:eval("t = True", "t").                           

        true

        9> pynerl:eval("t = 2.3", "t"). 

        2.3

Information

-----------

This is just an experiment, some of the code may be usable for other things but

as the documentation on NIFs says::

        Avoid doing lengthy work in NIF calls as that may degrade the

        responsiveness of the VM. NIFs are called directly by the same

        scheduler thread that executed the calling Erlang code. The calling

        scheduler will thus be blocked from doing any other work until the NIF

        returns.

That means that you shouldn't run a really long python method with it since it would

block the scheduler.

I may try to make something similar to `erlectricity`__

or erlv8_

__ http://github.com/mojombo/erlectricity/

.. _erlv8: https://github.com/beamjs/erlv8/wiki/Using-erlv8

building

--------

you will need a compiler, erlang dev files and python dev files, in debian

based systems you get them with::

    sudo apt-get install python3-dev erlang-dev build-essential

after that you may have to check that the paths in build.sh are the corrects

for you, specially this two:

* /usr/lib/erlang/erts-5.8.5/include/

* /usr/include/python3.2/ 

if you have another version of erlang or python just change them on build.sh

if it gives some linker error check the *-lpython3.2mu* and change it to fit

(maybe -lpython3.2)