packagelister.packagelister

 1import importlib.metadata
 2import sys
 3from pathlib import Path
 4
 5from pathcrawler import crawl
 6from printbuddies import ProgBar
 7
 8
 9def scan(project_dir: Path | str = None, include_builtins: bool = False) -> dict:
10    """Recursively scans a directory for python files to determine
11    what packages are in use, as well as the version number
12    if applicable.
13
14    Returns a dictionary where the keys are package
15    names and the values are the version number of the package if there is one
16    (None if there isn't) and a list of the files that import the package.
17
18    :param project_dir: Can be an absolute or relative path
19    to a directory or a single file (.py).
20    If it is relative, it will be assumed to be relative to
21    the current working directory.
22    If an argument isn't given, the current working directory
23    will be scanned.
24    If the path doesn't exist, an empty dictionary is returned."""
25    if not project_dir:
26        project_dir = Path.cwd()
27    elif type(project_dir) is str:
28        project_dir = Path(project_dir)
29    if not project_dir.is_absolute():
30        project_dir = project_dir.absolute()
31
32    # Return empty dict if project_dir doesn't exist
33    if not project_dir.exists():
34        return {}
35    # You can scan a non python file one at a time if you reeeally want to.
36    if project_dir.is_file():
37        files = [project_dir]
38    else:
39        files = [file for file in crawl(project_dir) if file.suffix == ".py"]
40
41    bar = ProgBar(len(files) - 1, width_ratio=0.33)
42    # If scanning one file, the progress bar will show 0% complete if bar.counter == 0
43    if len(files) == 1:
44        bar.counter = 1
45    packages = {}
46    standard_lib = list(sys.stdlib_module_names) if not include_builtins else []
47    for file in files:
48        bar.display(suffix=f"Scanning {file.name}")
49        contents = [
50            line.split()[1]
51            for line in file.read_text(encoding="utf-8").splitlines()
52            if line.startswith(("from", "import"))
53        ]
54        for package in contents:
55            if package.startswith("."):
56                package = package[1:]
57            if "." in package:
58                package = package[: package.find(".")]
59            if "," in package:
60                package = package[:-1]
61            if file.with_stem(package) not in files and package not in standard_lib:
62                if package in packages and str(file) not in packages[package]["files"]:
63                    packages[package]["files"].append(str(file))
64                else:
65                    try:
66                        package_version = importlib.metadata.version(package)
67                    except Exception as e:
68                        package_version = None
69                    packages[package] = {
70                        "files": [str(file)],
71                        "version": package_version,
72                    }
73    return packages
def scan( project_dir: pathlib.Path | str = None, include_builtins: bool = False) -> dict:
10def scan(project_dir: Path | str = None, include_builtins: bool = False) -> dict:
11    """Recursively scans a directory for python files to determine
12    what packages are in use, as well as the version number
13    if applicable.
14
15    Returns a dictionary where the keys are package
16    names and the values are the version number of the package if there is one
17    (None if there isn't) and a list of the files that import the package.
18
19    :param project_dir: Can be an absolute or relative path
20    to a directory or a single file (.py).
21    If it is relative, it will be assumed to be relative to
22    the current working directory.
23    If an argument isn't given, the current working directory
24    will be scanned.
25    If the path doesn't exist, an empty dictionary is returned."""
26    if not project_dir:
27        project_dir = Path.cwd()
28    elif type(project_dir) is str:
29        project_dir = Path(project_dir)
30    if not project_dir.is_absolute():
31        project_dir = project_dir.absolute()
32
33    # Return empty dict if project_dir doesn't exist
34    if not project_dir.exists():
35        return {}
36    # You can scan a non python file one at a time if you reeeally want to.
37    if project_dir.is_file():
38        files = [project_dir]
39    else:
40        files = [file for file in crawl(project_dir) if file.suffix == ".py"]
41
42    bar = ProgBar(len(files) - 1, width_ratio=0.33)
43    # If scanning one file, the progress bar will show 0% complete if bar.counter == 0
44    if len(files) == 1:
45        bar.counter = 1
46    packages = {}
47    standard_lib = list(sys.stdlib_module_names) if not include_builtins else []
48    for file in files:
49        bar.display(suffix=f"Scanning {file.name}")
50        contents = [
51            line.split()[1]
52            for line in file.read_text(encoding="utf-8").splitlines()
53            if line.startswith(("from", "import"))
54        ]
55        for package in contents:
56            if package.startswith("."):
57                package = package[1:]
58            if "." in package:
59                package = package[: package.find(".")]
60            if "," in package:
61                package = package[:-1]
62            if file.with_stem(package) not in files and package not in standard_lib:
63                if package in packages and str(file) not in packages[package]["files"]:
64                    packages[package]["files"].append(str(file))
65                else:
66                    try:
67                        package_version = importlib.metadata.version(package)
68                    except Exception as e:
69                        package_version = None
70                    packages[package] = {
71                        "files": [str(file)],
72                        "version": package_version,
73                    }
74    return packages

Recursively scans a directory for python files to determine what packages are in use, as well as the version number if applicable.

Returns a dictionary where the keys are package names and the values are the version number of the package if there is one (None if there isn't) and a list of the files that import the package.

Parameters
  • project_dir: Can be an absolute or relative path to a directory or a single file (.py). If it is relative, it will be assumed to be relative to the current working directory. If an argument isn't given, the current working directory will be scanned. If the path doesn't exist, an empty dictionary is returned.