source code generation library (with overuse with-syntax)
Project description
this package is heavily inspired by srcgen .
(todo: gentle introduction)
features
generating code with with-syntax
string injection after writing string
generating code with with-syntax
from prestring.python import PythonModule
m = PythonModule()
with m.class_("Point", metaclass="InterfaceMeta"):
with m.def_("__init__", "self", "value"):
m.stmt("self.value = value")
with m.def_("__str__", "self"):
m.return_("self.value")
output is.
class Point(object, metaclass=InterfaceMeta)
def __init__(self, value):
self.value = value
def __str__(self):
return self.value
string injection after writing string
from prestring.python import PythonModule
m = PythonModule()
with m.def_("setup", "config"):
import_area = m.submodule()
m.sep()
for k in ["a", "b", "c", "d", "e"]:
import_area.stmt("from .plugins import {k}_plugin", k=k)
m.stmt("config.activate({}_plugin)", k)
print(m)
def setup(config):
from .plugins import(
a_plugin,
b_plugin,
c_plugin,
d_plugin,
e_plugin
)
config.activate(a_plugin)
config.activate(b_plugin)
config.activate(c_plugin)
config.activate(d_plugin)
config.activate(e_plugin)
sub modules
prestring.output
prestring.python.transform, prestring.text.transform
prestring.output
prestring.output can write multiple files.
import sys
from prestring.python import Module
from prestring.output import output, cleanup_all # noqa
dst = sys.argv[1]
with output(root=dst) as fs:
with fs.open("projects/x.txt", "w") as wf:
print("hello x", file=wf)
print("bye x", file=wf)
with fs.open("projects/y.txt", "w") as wf:
print("hello y", file=wf)
print("bye y", file=wf)
with fs.open("projects/z.py", "w", opener=Module) as m:
with m.def_("hello"):
m.stmt("print('hello')")
Above code will generate three files. if creating directory is needed, if will be created automatically.
$ python src/main.py dst
[D] create dst/projects
[F] create dst/projects/x.txt
[F] create dst/projects/y.txt
[F] create dst/projects/z.py
On rerun, no message is displayed. And rerun with VERBOSE=1 var env to see more detailed output.
$ python src/main.py dst
$ VERBOSE=1 python src/main.py dst
[F] no change dst/projects/x.txt
[F] no change dst/projects/y.txt
[F] no change dst/projects/z.py
dry-run
Running with CONSOLE=1 varenv or calling with use_console=True option, doesn’t save files.
$ CONSOLE=1 python src/main.py dst
[F] update dst/projects/x.txt
[F] update dst/projects/y.txt
[F] update dst/projects/z.py
# more verbose output
VERBOSE=1 CONSOLE=1 python src/00/main.py dst/00/create
# dst/00/create/projects/x.txt
----------------------------------------
hello x
bye x
# dst/00/create/projects/y.txt
----------------------------------------
hello y
bye y
# dst/00/create/projects/z.py
----------------------------------------
def hello():
print('hello')
prestring.python.transform, prestring.text.transform
the Transform function means converting raw source code (or text) to prestring’s code. And you can use python -m prestring.python (or running python -m prestring.text) as a CLI command, as follows.
$ cat hello.py
def hello(name, *, message: str = "hello world"):
"""
greeting message
"""
print(f"{name}: {message}")
if __name__ == "__main__":
hello("foo")
$ python -m prestring.python hello.py
from prestring.python import PythonModule
def gen(*, m=None, indent=' '):
m = m or PythonModule(indent=indent)
import textwrap
with m.def_('hello', 'name', '*', 'message: str = "hello world"'):
m.docstring(textwrap.dedent("""
greeting message
""").strip())
m.stmt('print(f"{name}: {message}")')
with m.if_('__name__ == "__main__"'):
m.stmt('hello("foo")')
return m
if __name__ == "__main__":
m = gen(indent=' ')
print(m)
Of course, reversible.
$ python -m prestring.python hello.py > /tmp/gen_hello.py
$ python /tmp/gen_hello.py
def hello(name, *, message: str = "hello world"):
"""
greeting message
"""
print(f"{name}: {message}")
if __name__ == "__main__":
hello("foo")
limitation
prestring.python does not support all python grammers (e.g. async definition is not supported, yet). If you want to prestring’s expression as first step, prestring.text is probably useful.
$ python -m prestring.text hello.py
from prestring.text import Module
def gen(*, m=None, indent=' '):
m = m or Module(indent=indent)
m.stmt('def hello(name, *, message: str = "hello world"):')
with m.scope():
m.stmt('"""')
m.stmt('greeting message')
m.stmt('"""')
m.stmt('print(f"{name}: {message}")')
m.sep()
m.sep()
m.stmt('if __name__ == "__main__":')
with m.scope():
m.stmt('hello("foo")')
return m
if __name__ == "__main__":
m = gen(indent=' ')
print(m)
0.7.0
changes “python -m prestring.python.transform” to “python -m prestring.python”
some logging adjustments and fix bugs (prestring.output)
add prestring.text
0.6.0
support only python3.6+
refactoring prestring.output
0.5.3
rollback, python’s import part output
0.5.1, 0.5.2
some bug fixes
0.5.0
add prestring.python.transform
0.4.7
add prestring.output.SeparatedOutput
LazyKeywords is sorted
0.4.3
clear if empty (e.g. go-module’s import_group)
0.4.2
bug fix LazyArgumentsAndKeywords
0.4
more lazy (python module)
lazy object is cached
0.3.3
goname function
0.3.1
more lazy (go module)
0.3.0
go code output support
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Hashes for prestring-0.7.0-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 78d2f9a3acb4744707e3ba6466b6fffe55ffe289ca71cdfd53f98210b31acf31 |
|
MD5 | 743796ee39f99440d1ed349c91ea2bf0 |
|
BLAKE2b-256 | 61322bb7d880890a62952091715807129a5e6ed283b75ce611e6ff005eacc9c5 |