Be aware about imports meta information
Source Code: https://github.com/andresbena19/py-imports
Python 3.7+
py-imports stands on the shoulders of giants:
- ast — Abstract Syntax Trees to traverse python code.
$ pip install py-imports
---> 100%
All it's ready to begin
- Create a file
main.py
with:
import logging
from py_imports.manager import PyImports
from .parse_local_dir import file_imports_from_dir_one
myself = "main.py"
# Let's introspect myself
with PyImports() as manager:
imports_file = manager.get_imports(myself)
# Now you have access to the imports used in each file
imports_file
{
'main.py': <py_imports.base.models.ImportsCollectionFile object at 0x10b889220>
}
# Get details about the absolute, relative and standard imports in the file
collector_object = imports_file.get(myself)
absolute_imports = collector_object.absolute_imports
relative_imports = collector_object.relative_imports
standard_imports = collector_object.imports
Get meta information about absolute imports...absolute_imports
# Absolute imports
# --- from py_imports.manager import PyImports ---
# If we introspect the object, we will get the following
example_abs_import = absolute_imports[0]
example_abs_import.children -> ['PyImports']
example_abs_import.parent -> 'py_imports.manager'
example_abs_import.statement -> 'from py_imports.manager import PyImports'
example_abs_import.level -> 0
example_abs_import.line -> 2
Get meta information about relative imports...relative_imports
# relative imports
# --- from .parse_local_dir import file_imports_from_dir_one ---
# If we introspect the object, we will get the following
example_relative_import = relative_imports[0]
example_abs_import.children -> ['file_imports_from_dir_one']
example_abs_import.children_unused -> ['file_imports_from_dir_one']
example_abs_import.parent -> 'parse_local_dir'
example_abs_import.statement -> 'from .parse_local_dir import file_imports_from_dir_one'
example_abs_import.level -> 1
example_abs_import.line -> 3
Get meta information about standard imports ...standard_imports
# standard imports
# --- import logging ---
# If we introspect the object, we will get the following
example_standard_import = standard_imports[0]
example_standard_import.children -> ['logging']
example_standard_import.children_unused -> ['logging']
example_standard_import.statement -> 'import logging'
example_standard_import.line -> 1
The util allow identifying and group the imports according ...relative imports, absolu...
-
The util allow identifying and group the imports according to the abstract grammar defined with python
... | Import(alias* names) | ImportFrom(identifier? module, alias* names, int? level)
-
-
Relative imports use leading dots. A single leading dot indicates a relative import, starting with the current package. Two or more leading dots indicate a relative import to the parent(s) of the current package, one level per dot after the first.
-
Relative imports must always use
from <> import
;import <>
is always absolute.- pydocs: https://docs.python.org/3/reference/import.html#package-relative-imports
- Metadata will be abstracted in
RelativeImportStatement
objects.
-
from .moduleY import spam from .moduleY import spam as ham from . import moduleY from ..subpackage1 import moduleY from ..subpackage2.moduleZ import eggs from ..moduleA import foo from ...package import bar from ...sys import path
-
Relative imports must always use
-
Absolute import involves full path i.e., from the project’s root folder to the desired module. An absolute import state that the resource
to be imported using its full path from the project’s root folder.-
Absolute imports may use either the
import <>
orfrom <> import <>
syntax, but relative imports may only use the second form.- PEP328: https://www.python.org/dev/peps/pep-0328/
- Metadata will be abstracted in
AbsoluteImportStatement
objects.
-
from moduleY import spam from moduleY import spam as ham # OR import XXX.YYY.ZZZ
-
Absolute imports may use either the
-
Standard imports will be introspected and the data about it will be saved in an object named
ImportStatement
.-
standard imports use the
import <>
syntax.- PEP328: https://www.python.org/dev/peps/pep-0328/
- Metadata will be abstracted in
ImportStatement
objects.
-
import moduleY import moduleX
-
standard imports use the
-
If some child it's not used in an import ...children_unused...
-
If some child it's not used in an import, this will be added in
children_unused
attribute in every concrete implementation that represent an imports.from ..subpackage1 import moduleY, moduleZ def foo() -> moduleZ: pass
In this case the relative import
from ..subpackage1 import moduleY, moduleZ
has a child that is not used in the file.... # After introspect the file relative_imports = imports_file.relative_imports relative_imports[0].children_unused -> ["moduleY"] # But the total of children present in this file relative_imports[0].children -> ["moduleY", "moduleZ"]
it's used
pyflakes
to determine the unused imports, because follow the same philosophy to get the information just using a static analysis.
If the position of the import statement ...in_inner_scope...
-
If some imports are located inside an inner scope, the import object will contain a boolean field named
in_inner_scope
indicating that is located outside his default position (the top of the file or in the global scope), also will be included an attribute namedouter_parent_node
that will contain theAST
node, to allow the user get more information about the data structure node parent that is around the import.def foo(): from pkg import moduleY, moduleZ ...
In this case the absolute import is located inside a function named
foo
.... # After introspect the file absolute_imports = imports_file.absolute_imports absolute_imports[0].in_inner_scope -> True # it's possible to get the ast node parent with absolute_imports[0].outer_parent_node -> ast.AST object
This library does not execute any part of the python target code, this just make a static analysis over the code to describe the meta information about the imports in the file.
This project is licensed under the terms of the MIT license.