Skip to main content

advanced syntax&semantics extension system for Python

Project description


Build Support codecov

An advanced syntax extension system implemented in pure python.

pip install moshmosh-base


Working with IPython

You should copy to $USER/.ipython/profile_default/startup/.

If this directory does not exist, use command ipython profile create to instantiate.

Some examples about pattern matching, pipelines and quick lambdas:

IPython example 1

Some examples about the scoped operators:

IPython example 2

Working with regular Python files

Import moshmosh in your main module:

Then, in, start coding with a pragma comment # moshmosh?, then you can use moshmosh extension system.

Case Study : Pattern Matching

The matching protocol which stems from Python-ideas mailing list is introduced in, which means you can define your own patterns conveniently. The link is here.

# moshmosh?
# +pattern-matching

class GreaterThan:
    def __init__(self, v):
        self.v = v

    def __match__(self, cnt: int, to_match):
        if isinstance(to_match, int) and cnt is 0 and to_match > self.v:
            return () # matched
        # 'return None' indicates 'unmatched'

with match(114, 514):
    if (GreaterThan(42)() and a, b):
        print(b, a)
# 514 114

Note that the matching clauses should be exhaustive, otherwise, a moshmosh.extensions.pattern_matching.runtime.NotExhaustive might get raised.

The supported Patterns are listed here, which is of course much more powerful than most programming languages.

  • And pattern: pat1 and pat2 and pat3 ...
  • Or pattern: pat1 or pat2 or pat3...
  • Pin pattern: pin(value), this is quite useful. See Elixir Pin Operator
  • Literal pattern: 1, "str", 1+2j, (1, 2)
  • As pattern: a, var
  • Wildcard: _
  • Nested patterns:
    • Tuple: (pat1, pat2, pat3), (pat1, *pat2, pat3)
    • List: [pat1, pat2, pat3], [pat1, *pat2, pat3]
    • Recogniser: Cons(pat1, pat2, pat3), note that, the function Cons.__match__(<n arg>, value_to_match) is exact the protocol.

The pattern matching should be more efficient than those hand-written codes without ugly optimizations.

Besides, Moshmosh's pattern matching is orders of magnitude faster than any other alternatives.

Case Study : Template-Python

This is relatively a simple quasiquote implementation, inspired by MetaOCaml. It does not support manual splices or nested quotations, but the function arguments are automatically spliced.

# moshmosh?
# +template-python

def f(x):
    x + 1
    x = y + 1

from moshmosh.ast_compat import ast
from astpretty import pprint

stmts = f(ast.Name("a"))

# =>
        left=Name(lineno=7, col_offset=4, id='a', ctx=Load()),
        right=Num(lineno=7, col_offset=8, n=1),
    targets=[Name(lineno=8, col_offset=4, id='a', ctx=Store())],
        left=Name(lineno=8, col_offset=8, id='y', ctx=Load()),
        right=Num(lineno=8, col_offset=12, n=1),


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for moshmosh-base, version 0.3
Filename, size File type Python version Upload date Hashes
Filename, size moshmosh-base-0.3.tar.gz (24.4 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page