- Notifications
You must be signed in to change notification settings - Fork14
a restricted python to javascript / c# / go / ruby compiler
License
pseudo-lang/pseudo-python
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
A restricted Python to idiomatic JavaScript / Ruby / Go / C# translator
Pseudo is a framework for high level code generation: it is used by this compiler to translate a subset of Python to all Pseudo-supported languages
If you are using Python3.5 and you experience problems with an already installed version of pseudo-python, please upgrade it to0.2.34
(pip3 install pseudo-python --upgrade
)
Sorry: I lied, manipulating the git history in a way that it appears most of the work is after 1 march: I was finishing my work in a company in february and I wanted to hide the fact i've worked on pseudo in that time: I actually started more active work somewhere around 20-th february if I remember correctly, somewhere around the second half of february. I am not sure if i have that original git history anymore: sorry to Clevertech(the company)/any observers, this was a really ugly thing to do.
Pseudo-Python compiles topseudo ast
.
Pseudo defines a language-independent AST model and an unified standard library.It can map its own standard library to target language libraries and concepts automatically and it tries to generate readable and idiomatic code.
Pseudo-Python translates a subset of Python to Pseudo AST and then it receives the JS/Ruby/C#/Go backends for free. (+ at least 4-5 backends in the future)
Pseudo was inspired by the need to generate algorithms/code in different languages or portint tools/libraries to a new environment
That's why it can be mapped to a well defined subset of a language
It is meant as a framework consuming ast from parser generators / compilers / various tools and generating snippets / codebases in different target languages
basic types and collections and standard library methods for them
integer, float, string, boolean
lists
dicts
sets
tuples/structs(fixed length heterogeneous lists)
fixed size arrays
regular expressions
functions with normal parameters (no default/keyword/vararg parameters)
classes
- single inheritance
- polymorphism
- no dynamic instance variables
- basically a constructor + a collection of instance methods, no fancy metaprogramming etc supported
exception-based error handling with support for custom exceptions(target languages support return-based error handling too)
io: print/input, file read/write, system and subprocess commands
iteration (for-in-range / for-each / iterating over several collections / while)
conditionals (if / else if / else)
standard math/logical operations
pip install pseudo-python
pseudo-python<filename.py> rubypseudo-python<filename.py> csharp
etc for all the supported pseudo targets (javascript, c#, go, ruby and python)
Each example contains a detailed README and working translations to Python, JS, Ruby, Go and C#, generated by Pseudo
a football results processing command line tool
a verbal expressions-like library ported to all the target languages
A lot of work has been put into making pseudo-python error messages as clear and helpful as possible: they show the offending snippet of code andoften they offer suggestions, list possible fixes or right/wrong ways to write something
Beware, pseudo and especially pseudo-python are still in early stage, so if there is anything weird going on, don't hesitate to submit an issue
pseudo-python checks if your program is using a valid pseudo-translatable subset of Python, type checks it according to pseudo type rules and then generates a pseudo ast and passes it to pseudo for code generation.
The rules are relatively simple: currently pseudo-python infers everythingfrom the usage of functions/classes, so has sufficient information when the program is calling/initializing allof its functions/classes (except for no-arg functions)
Often you don't really need to do that forall of them, you just need to do it in a way that can create call graphs covering all of them (e.g. often you'll havea
callingb
callingx
and you only need to have ana
invocation in your source)
You can also use type annotations. We are trying to respect existing Python3 type annotation conventions and currently pseudo-python recognizesint
,float
,str
,bool
,List[<type>]
,Dict[<key-type>, <value-type>]
,Tuple[<type>..]
,Set[<type>]
andCallable[[<type>..], <type>]
Beware, you can't just annotate one param, if you provide type annotations for a function/method, pseudo-python expects type hints for all params and a return type
Variables can't change their types, the equivalents for builtin types are
list :List[@element_type]# genericdict:Dictionary[@key_type @value_type]# genericset:Set[@element_type]# generictuple:Array[@element_type]# for homogeneous tuplesTuple[@element0_type, @element1_type..]# for heterogeneous tuplesint:Intfloat:Floatint/float:Numberstr:Stringbool:Boolean
There are several limitations which will probably be fixed in v0.3
If you initialize a variable/do first call to a function with a collection literal, it should have at least one element(that limitation will be until v0.3)
All attributes used in a class should be initialized in its__init__
Other pseudo-tips:
Homogeneous tuples are converted to
pseudo
fixed length arrays and heterogeneous topseudo
tuples.Pseudo analyzes the tuples usage in the code and sometimes it translates them to classes/structs with meaningful names if the target language isC#
C++
orGo
Attributes that aren't called from other classes are translated as
private
, the other ones aspublic
. The rule for methods is different:_name
ones are only translated asprivate
. That can be added asconfig option in the futureMultiple returns values are supported, but they are converted to
array
/tuple
Single inheritance is supported,
pseudo-python
supports polymorphismbut methods in children should accept the same types as their equivalents in the hierarchy (except__init__
)
The easiest way to play with the type system is to just try several programs:pseudo-python
errors should be enough to guide you, if not,you can always open an issue
The implementation goal is to make the definitions of new supported languages really clear and simple.
If you dive in, you'll find outa lot of the code/api transformations are defined using a declarative dsl with rare ocassionsof edge case handling helpers.
That has a lot of advantages:
Less bugs: the core transformation code is really generalized, it's reused as a dsl and its results are well tested
Easy to comprehend: it almost looks like a config file
Easy to add support for other languages: I was planning to support just python and c# in the initial version but it is so easy to add support for a language similar to the current supported ones, that Iadded support for 4 more.
Easy to test: there is a simple test dsl too which helps alllanguage tests to share input exampleslike that
However language translation is related to a lot of details anda lot of little gotchas, tuning and refining some of them took days. Pseudo uses different abstractions to streamline the process and to reuse logic across languages.
PSEUDOAST:NORMALCODEPSEUDOSTANDARDLIBRARYINVOCATIONS || || || || ||APITRANSLATOR || || || || || \/ ||IDIOMATICTARGETLANGUAGE ||STANDARDLIBRARYINVOCATIONS || || \/ \/STANDARDORLANGUAGE-SPECIFICMIDDLEWARESe.g.namecamel_case/snake_casemiddlewareconvert-tuples-to-classesmiddlewareconvert-exception-basederrorshandlingtoreturn-basederrorhandlingmiddlewareetc || || || ||TARGETLANGUAGECODEGENERATORdefinedwithadslawarethathandlesformattingautomatically || || || \/OUTPUT
They might seem comparable at a first glance, but they have completely different goals.
Pseudo wants to generate readable code, ideally something that looks like a human wrote it/ported it
Pseudo doesn't use a target language runtime, it uses the target language standard library for everything (except for JS, but even there is useslodash
which is pretty popular and standard)
Pseudo's goal is to help with automated translation for caseslike algorithm generation, parser generation, refactoring, porting codebases etc. The fact that you can write compilers targetting Pseudo and receiver translation to many languages for free is just a happy accident
Copyright © 2015 2016Alexander Ivanov
Distributed under the MIT License.
About
a restricted python to javascript / c# / go / ruby compiler
Topics
Resources
License
Uh oh!
There was an error while loading.Please reload this page.
Stars
Watchers
Forks
Packages0
Uh oh!
There was an error while loading.Please reload this page.
Contributors4
Uh oh!
There was an error while loading.Please reload this page.