A Python 3.6+ package to run .many files, where many programs written in many languages may exist in one file.

Overview

PyPI - Python Version Test Coverage

RunMany

Intro | Installation | VSCode Extension | Usage | Syntax | Settings | About

A tool to run many programs written in many languages from one file.

Suppose you want to practice multiple programming languages at once. Normally you'd have to juggle multiple files or multiple projects, perhaps multiple IDEs. RunMany lets you write multiple programs in the same file using any programming languages you want, and then run them all at once.

For example, give RunMany this simple file

Python:
    print("Hi")
JavaScript:
    console.log("Hi")
Rust:
    fn main() {
        println!("Hi");
    }

and it will number and run each program, giving this output:

************************************************************
1. Python
-------------------- output from line 1 --------------------
Hi


************************************************************
2. JavaScript
-------------------- output from line 3 --------------------
Hi


************************************************************
3. Rust
-------------------- output from line 5 --------------------
Hi


************************************************************
3/3 programs successfully run!
3/3 had the exact same stdout!
************************************************************

In general, RunMany can be used for:

  • Chrestomathy - Writing identically behaving programs in many languages, like on Rosetta Code. (example/output)
  • Performance Testing - Timing different implementations of a program, even across languages. (example/output)
  • Input Testing - Easily giving many combinations of argv or stdin to programs. (example/output)
  • Polyglots - Making esoteric code that can be executed in multiple languages at once. (example/output)

Overall it is hopefully a good tool for anyone who wants to play with multiple programming languages at once.

Installation (supports Python 3.6+)

pip install runmany

If that doesn't work try pip3 install runmany or python -m pip install runmany or python3 -m pip install runmany.

PyPI Package Page | Bleeding edge version on TestPyPI

VSCode Extension

The RunMany VSCode extension adds syntax highlighting to RunMany files and makes them runnable with one button. It is highly recommended for use with RunMany.

With and without syntax highlighting:

syntax highlighting example

Get VSCode here and get the RunMany extension here, or install it directly once you have VSCode with:

code --install-extension discretegames.runmany

Usage

Running From Command Line

runmany myfile.many

More generally:

runmany [-h --help] [-j --json <settings-file>] [-o --output <output-file>] <input-file>
  • <input-file> is the required .many file to run.
  • <settings-json> is the optional .json file that defines how languages are run and how the output is formatted.
  • <output-file> is the optional file to send the output to. When omitted output goes to stdout.

When a settings JSON file is not provided, the hardcoded settings JSON at the top of the .many file is used. If neither is present, or for any missing settings, default_settings.json is used as a fallback.

See the examples folder for some .many files to try. Note that RunMany expects the system to already have the necessary interpreters and compilers installed for the programming languages it runs. RunMany runs them internally with normal console commands.

RunMany has preset commands for a number of languages:

Ada, C, C#, C++, Dart, Fortran, Go, Groovy, Haskell, Java, JavaScript, Julia, Kotlin, Lisp, Lua, Pascal, Perl, PHP, Python, Python 2, R, Racket, Ruby, Rust, Scala, TypeScript, VBScript, and Visual Basic

But these presets were made for a Windows machine and may fail depending on OS and system configuration. However, commands can be overridden and new languages can be added by modifying the settings JSON. See more below.

Running From Python

from runmany import runmany, runmany_to_s, runmany_to_f

# Run to stdout
runmany('path/to/input.many', 'path/to/settings.json') # settings JSON is always optional

# Run to output file
runmany('path/to/input.many', 'path/to/settings.json', 'path/to/output.txt')

# Run to string
string = runmany_to_s('path/to/input.many', 'path/to/settings.json')

# Run to file object
with open('output.txt', 'w') as file_obj:
    runmany_to_f(file_obj, 'path/to/input.many', 'path/to/settings.json')

As with the command line, the settings JSON provided as an argument takes precedence over the one that may be at the top of the .many file, and default_settings.json is used as a fallback for all settings.

In each of the 3 runmany functions, the settings JSON argument may be given as a path to the .json file or a JSON-like Python dictionary.

Additionally, the .many file contents may be given as a string rather than a file path with from_string=True.

The function runmany.cmdline, which takes a list of command line arguments, is also present as an alternative to using the command line directly.

.many Syntax

The .many file format is what RunMany expects when given a file to run. (Though, of course, ".many" is not required as an extension.)

Principally, a .many file consists of unindented lines which are section headers that define the languages and context for the lines indented below them. Languages are given as a comma separated list and the 3 contexts are argv, stdin, and code. Section headers end with a colon.

Argv for Python, JavaScript:
    foo
Stdin for Python:
    bar
Python:
    import sys
    print(sys.argv[1] + input())  # will be "foobar"
JavaScript:
    console.log(process.argv[2])  // will be "foo"

The keywords Argv and Stdin are used to define the argument vector and standard input for a set of languages. Otherwise the section is assumed to be code.

So the RunMany program above will send "foo" to Python and JavaScript on argv, and "bar" to Python on stdin when it runs each language's code.

Importantly, a .many file always runs from top to bottom just-in-time, that is, the top lines will run normally even if the bottom lines are invalid syntax. For this reason, argv and stdin sections only apply to code sections that come after them.

Those are the essentials but read on for more details and nuance about the syntax of .many files. Notably the Also Section and hardcoded settings.

Also check syntax.many and the other examples for concrete syntax samples.

Syntax Specifics


Comments

% at the very start of a line makes an inline comment.

% this is a comment

There are no block comments.


Section Syntax

A .many file can be split up into sections, each of which has an unindented header line that ends in a colon (:), and a potentially multiline string of content that can appear after the colon and on indented lines below the header. Each indent must be either a single tab or 4 spaces and the indents do not end up as part of the section content.

Any whitespace just after the colon of the section header is ignored, so this is a working Code Section:

Python: import math
    print(math.pi)

As it corresponds to the Python program:

import math
print(math.pi)

Blank lines above or below sections are only for readability and not required.

As detailed below, only a few types of sections exist and some require comma (,) separated language lists in their headers. Language names are stripped of whitespace and matched to corresponding "name" keys in the languages arrays of the settings JSON.

Language names are not case-sensitive (Python is the same as python) but other keywords like Argv, Stdin, for, Also, and Exit are.

Language names cannot contain , or : and to be safe they should not start with Argv, Stdin, Also, Exit, or !.


Code Section

A Code Section starts right out with a comma separated list of languages and its content is the program to be run in those languages.

One language in the list is almost always sufficient unless you are writing polyglots,

JavaScript:
    console.log('This is some code that will be run in JS.')
Python, Python 2:
    print('This is some code that will be run in Python 3 and then Python 2.')

Argv Section

An Argv Section can either start Argv: to apply to all languages, or Argv for Language1, Language2, ...: to apply to the listed languages. Either way overwrites any previous argv set for those languages, but Also Sections can be used to supply a series of argvs.

The Argv Section's content is stripped of newlines and sent as the argument vector to all the subsequent programs in Code Sections it applies to.

Argv:
    argv sent to all languages
Argv for Python, JavaScript:
    argv specifically sent to Python and Javascript

For argv to work the $argv placeholder must be placed properly into the command of the language.


Stdin Section

Almost exactly like an Argv Section but for stdin.

A Stdin Section can either start Stdin: to apply to all languages, or Stdin for Language1, Language2, ...: to apply to the listed languages. Either way overwrites any previous stdin set for those languages, but Also Sections can be used to supply a series of stdins.

The Stdin Section's content is stripped of trailing newlines except one and sent as the standard input stream to all the subsequent programs in Code Sections it applies to.

Stdin:
    stdin sent to all languages
Stdin for Python, JavaScript:
    stdin specifically sent to Python and Javascript

When a program expects stdin but there is no Stdin Section to give it, the stdin can be typed into the console normally.


Also Section

An Also Section starts with Also: (with no language list) and is a way to add a series of argvs or stdins to run, or to avoid repeating a Code Section header. It cannot be the first section in the file because it needs to attach to the Code, Stdin, or Argv Section above it.

When below an Argv Section or Stdin Section, an Also Section adds an additional input to the list of argvs or stdins to run when applicable Code Sections are encountered.

For example, the final Python program here is run 6 times for all the combinations of argvs and stdins that apply to it (1A 1B 2A 2B 3A 3B):

Argv: 1
Also: 2
Also: 3

Stdin: A
Also:  B

Python:
    import sys
    print(sys.argv[1] + input())

This is the real power of the Also Section -- giving multiple argvs and stdins to a program without repeating code. Another example with output.

When below a Code Section, an Also Section is simply shorthand for repeating the Code Section's header.

For example, Also: here behaves exactly the same as Python, Python 2: would:

Python, Python 2:
    print(123)
Also:
    print(456)
Also:
    print(789)

Disabling Sections

Putting ! at the very start of any section header will disable that section and any Also Sections attached to it.

!Python:
    # this is disabled
Also:
    # this is effectively disabled too
!Also:
    # this is disabled in two ways

Hardcoded Settings

A settings JSON may be placed, indented, before the first section in a .many file. It is only used if a custom setting JSON is not otherwise provided as an argument, and only for the .many file it is in. As with section content, the indents may be either single tabs or 4 spaces.

    {
        "show_time": true,
        "show_command": true
    }
Python:
    print('The time and command will now be shown.')

Exit Command

Exit. at the very start of a line by itself will stop RunMany as if the file ended there.

Exit.
% nothing from here on will be run

Settings JSON

The settings JSON defines what languages RunMany can run and how it will run them. It also defines how the RunMany output will be formatted.

As mentioned, default_settings.json holds the default values for all settings which are automatically used if not otherwise present in a provided or hardcoded JSON.

Most settings are simple flags or values that can be set in the base settings JSON object. See List of Settings below.

The setting to add a custom language is the "languages" key which maps to an array of JSON objects we'll call language objects. Each language object must have a "name" string to identify it and a "command" string to run it (see command format). However, objects in "languages" with a matching "name" in the "default_languages" array will automatically inherit its other values, such as "command" and "ext". Most settings that can be set in the base settings JSON object are also inherited by the language objects and can be overridden.

For example, a settings JSON of

{
    "languages": [{ "name": "Rust", "timeout": 5.0 }],
    "show_code": true
}

will make Rust programs have a 5 second time limit rather than the default of 10, and "command" does not need to be present because Rust is already in the built-in "default_languages" array. The "show_code": true in the base object makes it so all languages in the RunMany output will show their code.

You should not have to set "default_languages" in your custom settings JSON (though technically you can). Only set "languages".

List of Settings

All settings described and whether or not they they can be overridden in a language object in the "languages" array:

JSON Key Type Default Overridable Description
"timeout" float 10.0 yes The time limit of each program in seconds.
"stderr" string "nzec" yes "always" or true to always combine program stderr streams with stdout. "never" or false to always hide program stderr streams. "nzec" or null to only show stderr streams when programs have non-zero exit codes.
"ext" string "" yes The file extension of a language including the dot. Best to always define in the language object.
"spacing" int 1 yes The number of blank lines to add after each run. Note that trailing newlines are not stripped from stdouts.
"show_time" bool false yes Whether the execution time of each program is shown.
"show_command" bool false yes Whether the command used to run each program is shown. Useful for debugging command setup for new languages.
"show_code" bool false yes Whether the source code of the program is shown.
"show_argv" bool true yes Whether the argv for the program is shown (when present and non-empty).
"show_stdin" bool true yes Whether the stdin for the program is shown (when present and not all empty lines).
"show_output" bool true yes Whether the output for the program is shown. This includes the stdout, and, depending on the "stderr" setting, the stderr.
"show_errors" bool true no Whether RunMany errors like !!!| RunMany Error: ... |!!! are sent to stderr or silenced.
"show_runs" bool true no Whether the list of runs is shown. This is usually the bulk of the output.
"show_stats" bool true no Whether the success and failure counts are shown after everything has run.
"show_equal" bool true no Whether the matching stdouts are compared and grouped after everything has run.

Command Format

The "command" key of a language object in the "languages" array defines the terminal command that is run to execute the language.

Placeholders like $file and $dir are used in a command to refer to the temporary file RunMany creates for the code of each program it runs, or the directory that file is stored in:

Placeholder Portion of .../dir/file.ext
$rawdir .../dir
$dir ".../dir"
$rawfile .../dir/file.ext
$file ".../dir/file.ext"
$rawbranch .../dir/file
$branch ".../dir/file"
$name file.ext
$stem file
$ext .ext
$sep / (OS specific)
$argv n/a - the argv is inserted here

Note that some placeholders are "quoted" and some are not. Some operating systems like Windows may have spaces in the path to temporary files so correct quoting is important.

If $ is not present anywhere in the command string, $file $argv is appended to it. For example, the command python is implicitly python $file $argv.

Check the "default_languages" array in default_settings.json for more examples of commands.

About

I was driven to make RunMany by my desire to learn more programming languages combined with my annoyance that whenever I tried I would invariably have to make a whole new project for that language, or even switch IDEs.

I plan to use it to practice solving code challenges in multiple languages from code challenge websites.

Check out some of my other Python packages.

[EMNLP 2021] LM-Critic: Language Models for Unsupervised Grammatical Error Correction

LM-Critic: Language Models for Unsupervised Grammatical Error Correction This repo provides the source code & data of our paper: LM-Critic: Language M

Michihiro Yasunaga 98 Nov 24, 2022
Pre-training with Extracted Gap-sentences for Abstractive SUmmarization Sequence-to-sequence models

PEGASUS library Pre-training with Extracted Gap-sentences for Abstractive SUmmarization Sequence-to-sequence models, or PEGASUS, uses self-supervised

Google Research 1.4k Dec 22, 2022
Abhijith Neil Abraham 2 Nov 05, 2021
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition

CRNN paper:An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition 1. create your ow

Tsukinousag1 3 Apr 02, 2022
Python3 to Crystal Translation using Python AST Walker

py2cr.py A code translator using AST from Python to Crystal. This is basically a NodeVisitor with Crystal output. See AST documentation (https://docs.

66 Jul 25, 2022
Athena is an open-source implementation of end-to-end speech processing engine.

Athena is an open-source implementation of end-to-end speech processing engine. Our vision is to empower both industrial application and academic research on end-to-end models for speech processing.

Ke Technologies 34 Sep 08, 2022
VoiceFixer VoiceFixer is a framework for general speech restoration.

VoiceFixer VoiceFixer is a framework for general speech restoration. We aim at the restoration of severly degraded speech and historical speech. Paper

Leo 174 Jan 06, 2023
Auto_code_complete is a auto word-completetion program which allows you to customize it on your needs

auto_code_complete is a auto word-completetion program which allows you to customize it on your needs. the model for this program is one of the deep-learning NLP(Natural Language Process) model struc

RUO 2 Feb 22, 2022
A Practitioner's Guide to Natural Language Processing

Learn how to process, classify, cluster, summarize, understand syntax, semantics and sentiment of text data with the power of Python! This repository contains code and datasets used in my book, Text

Dipanjan (DJ) Sarkar 1.5k Jan 03, 2023
Espresso: A Fast End-to-End Neural Speech Recognition Toolkit

Espresso Espresso is an open-source, modular, extensible end-to-end neural automatic speech recognition (ASR) toolkit based on the deep learning libra

Yiming Wang 919 Jan 03, 2023
Natural language processing summarizer using 3 state of the art Transformer models: BERT, GPT2, and T5

NLP-Summarizer Natural language processing summarizer using 3 state of the art Transformer models: BERT, GPT2, and T5 This project aimed to provide in

Samuel Sharkey 1 Feb 07, 2022
Script to generate VAD dataset used in Asteroid recipe

About the dataset LibriVAD is an open source dataset for voice activity detection in noisy environments. It is derived from LibriSpeech signals (clean

11 Sep 15, 2022
Lyrics generation with GPT2-based Transformer

HuggingArtists - Train a model to generate lyrics Create AI-Artist in just 5 minutes! 🚀 Run the demo notebook to train 🚀 Run the GUI demo to test Di

Aleksey Korshuk 65 Dec 19, 2022
This code extends the neural style transfer image processing technique to video by generating smooth transitions between several reference style images

Neural Style Transfer Transition Video Processing By Brycen Westgarth and Tristan Jogminas Description This code extends the neural style transfer ima

Brycen Westgarth 110 Jan 07, 2023
This repo is to provide a list of literature regarding Deep Learning on Graphs for NLP

This repo is to provide a list of literature regarding Deep Learning on Graphs for NLP

Graph4AI 230 Nov 22, 2022
Toward a Visual Concept Vocabulary for GAN Latent Space, ICCV 2021

Toward a Visual Concept Vocabulary for GAN Latent Space Code and data from the ICCV 2021 paper Sarah Schwettmann, Evan Hernandez, David Bau, Samuel Kl

Sarah Schwettmann 13 Dec 23, 2022
BERT-based Financial Question Answering System

BERT-based Financial Question Answering System In this example, we use Jina, PyTorch, and Hugging Face transformers to build a production-ready BERT-b

Bithiah Yuan 61 Sep 18, 2022
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.

Tensor2Tensor Tensor2Tensor, or T2T for short, is a library of deep learning models and datasets designed to make deep learning more accessible and ac

12.9k Jan 07, 2023
A Python wrapper for simple offline real-time dictation (speech-to-text) and speaker-recognition using Vosk.

Simple-Vosk A Python wrapper for simple offline real-time dictation (speech-to-text) and speaker-recognition using Vosk. Check out the official Vosk G

2 Jun 19, 2022
SimpleChinese2 集成了许多基本的中文NLP功能,使基于 Python 的中文文字处理和信息提取变得简单方便。

SimpleChinese2 SimpleChinese2 集成了许多基本的中文NLP功能,使基于 Python 的中文文字处理和信息提取变得简单方便。 声明 本项目是为方便个人工作所创建的,仅有部分代码原创。

Ming 30 Dec 02, 2022