Python functions to get top-level importables names
Improve implementation
Good parts taken from https://gist.github.com/pradyunsg/22ca089b48ca55d75ca843a5946b2691.
Juan Luis Cano Rodríguez
(Dec 4, 2023, 9:28 PM +0100)
75d28430
8272d96b
+79
-22
+79
-22
src/pygetpackages/__init__.py
+79
-22
src/pygetpackages/__init__.py
···
1
1
"""Get the top-level packages of a Python project."""
2
+
# The good parts come from https://gist.github.com/pradyunsg/22ca089b48ca55d75ca843a5946b2691
3
+
# Licensed under the MIT license.
4
+
#
5
+
# Copyright (c) 2022 Pradyun Gedam <mail@pradyunsg.me>
6
+
#
7
+
# Permission is hereby granted, free of charge, to any person obtaining a copy
8
+
# of this software and associated documentation files (the “Software”), to deal
9
+
# in the Software without restriction, including without limitation the rights
10
+
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
11
+
# copies of the Software, and to permit persons to whom the Software is
12
+
# furnished to do so, subject to the following conditions:
13
+
#
14
+
# The above copyright notice and this permission notice shall be included in
15
+
# all copies or substantial portions of the Software.
16
+
#
17
+
# THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
18
+
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
19
+
# FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL THE
20
+
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
21
+
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
22
+
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
23
+
# SOFTWARE.
2
24
3
-
import csv
4
25
import pathlib
5
26
import tempfile
6
27
import tomllib
28
+
import typing as t
7
29
import zipfile
8
30
9
31
from build import ProjectBuilder
10
32
from build.env import DefaultIsolatedEnv
33
+
from installer.sources import WheelFile, WheelSource
34
+
from installer.utils import parse_metadata_file
11
35
from pyproject_hooks import quiet_subprocess_runner
12
36
from validate_pyproject import api, errors
13
37
14
38
15
-
def _simple_build_wheel(source_dir, out_dir, *, build_config_settings=None):
39
+
def _simple_build_wheel(
40
+
source_dir, out_dir, *, build_config_settings=None
41
+
) -> WheelFile:
16
42
"""Silently build wheel using build package."""
17
43
with DefaultIsolatedEnv() as env:
18
44
builder = ProjectBuilder.from_isolated_env(
···
24
50
return path_built_wheel
25
51
26
52
53
+
def _find_importable_components_from_wheel_content_listing(
54
+
filepaths: t.Iterable[str], *, dist_info_dir: str, data_dir: str
55
+
) -> t.Iterable[tuple[str, ...]]:
56
+
purelib_str = f"{data_dir}/purelib/"
57
+
platlib_str = f"{data_dir}/platlib/"
58
+
for path in filepaths:
59
+
if path.startswith(dist_info_dir):
60
+
# Nothing in dist-info is importable.
61
+
continue
62
+
63
+
if path.startswith((platlib_str, purelib_str)):
64
+
# Remove the prefix from purelib and platlib files.
65
+
name = path[len(platlib_str) :]
66
+
elif path.startswith(data_dir):
67
+
# Nothing else in data is importable.
68
+
continue
69
+
else:
70
+
# Top level files end up in an importable location.
71
+
name = path
72
+
73
+
if name.endswith(".py"):
74
+
yield tuple(name[: -len(".py")].split("/"))
75
+
76
+
77
+
def _determine_major_import_names(
78
+
importable_components: t.Iterable[tuple[str, ...]]
79
+
) -> set[str]:
80
+
return {components[0] for components in importable_components}
81
+
82
+
83
+
def _get_importable_components_from_wheel(wheel: WheelSource) -> t.Iterable[str]:
84
+
metadata = parse_metadata_file(wheel.read_dist_info("WHEEL"))
85
+
if not (metadata["Wheel-Version"] and metadata["Wheel-Version"].startswith("1.")):
86
+
raise NotImplementedError("Only supports wheel 1.x")
87
+
88
+
filepaths: t.Iterable[str] = (
89
+
record_elements[0] for record_elements, _, _ in wheel.get_contents()
90
+
)
91
+
importable_components = _find_importable_components_from_wheel_content_listing(
92
+
filepaths, dist_info_dir=wheel.dist_info_dir, data_dir=wheel.data_dir
93
+
)
94
+
95
+
return importable_components
96
+
97
+
27
98
def get_packages_from_wheel(wheel_path: str | pathlib.Path):
28
99
"""Get the top-level packages of a Python project from a wheel file."""
29
100
with zipfile.ZipFile(wheel_path, "r") as zf:
30
-
record_fnames = [
31
-
fname for fname in zf.namelist() if fname.endswith(".dist-info/RECORD")
32
-
]
33
-
record_fname = min(record_fnames, key=len)
34
-
with zf.open(record_fname) as fh:
35
-
record_content = fh.read().decode()
101
+
wheel_file = WheelFile(zf)
102
+
importable_components = _get_importable_components_from_wheel(wheel_file)
103
+
top_packages = _determine_major_import_names(importable_components)
36
104
37
-
record_files = [
38
-
pathlib.Path(row[0]) for row in csv.reader(record_content.splitlines())
39
-
]
40
-
41
-
# A directory is a top-level package if it contains a top-level __init__.py file
42
-
top_packages = {
43
-
record_file.parts[0]
44
-
for record_file in record_files
45
-
if len(record_file.parts) > 1 and record_file.parts[1] == "__init__.py"
46
-
}
47
-
48
-
return list(top_packages)
105
+
return top_packages
49
106
50
107
51
108
def get_packages(source_dir, *, build_config_settings=None):
52
-
"""Get the top-level packages of a Python project."""
109
+
"""Get the top-level packages of a Python source tree."""
53
110
pyproject_toml_path = pathlib.Path(source_dir) / "pyproject.toml"
54
111
if not pyproject_toml_path.is_file():
55
112
raise ValueError(
···
76
133
)
77
134
top_packages = get_packages_from_wheel(wheel_path)
78
135
79
-
return list(top_packages)
136
+
return top_packages