11992535834

Committed 24 Nov 2024 02:51AM UTC coverage: 75.597%. First build

Build # 11992535834

Build Type

push

github

Committed by

mborsetti

Commit Message

Version 3.27.0b2

Run Details

1703 of 2575 branches covered (66.14%)

Branch coverage included in aggregate %.

29 of 40 new or added lines in 6 files covered. (72.5%)

4505 of 5637 relevant lines covered (79.92%)

6.19 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

66.98

/webchanges/cli.py

#!/usr/bin/env python3

"""Module containing the entry point: the function main()."""

# See config module for the command line arguments.

# The code below is subject to the license contained in the LICENSE file, which is part of the source code.

from __future__ import annotations

import logging
import os
import platform
import shutil
import signal
import subprocess  # noqa: S404 Consider possible security implications associated with the subprocess module.
import sys
import warnings
from pathlib import Path, PurePath

import platformdirs

from webchanges import __copyright__, __docs_url__, __min_python_version__, __project_name__, __version__
from webchanges.config import CommandConfig
from webchanges.util import file_ownership_checks, get_new_version_number, import_module_from_source

# Ignore signal SIGPIPE ("broken pipe") for stdout (see https://github.com/thp/urlwatch/issues/77)
if os.name != 'nt':  # Windows does not have signal.SIGPIPE
    signal.signal(signal.SIGPIPE, signal.SIG_DFL)  # type: ignore[attr-defined]  # not defined in Windows

logger = logging.getLogger(__name__)


def python_version_warning() -> None:
    """Check if we're running on the minimum Python version supported and if so print and issue a pending deprecation
    warning."""
    if sys.version_info[0:2] == __min_python_version__:
        current_minor_version = '.'.join(str(n) for n in sys.version_info[0:2])
        next_minor_version = f'{__min_python_version__[0]}.{__min_python_version__[1] + 1}'
        warning = (
            f'Support for Python {current_minor_version} will be ending three years from the date Python '
            f'{next_minor_version} was released'
        )
        print(f'WARNING: {warning}\n')
        PendingDeprecationWarning(warning)


def migrate_from_legacy(
    legacy_package: str,
    config_file: Path | None = None,
    jobs_file: Path | None = None,
    hooks_file: Path | None = None,
    ssdb_file: Path | None = None,
) -> None:
    """Check for existence of legacy files for configuration, jobs and Python hooks and migrate them (i.e. make a copy
    to new folder and/or name). Original files are not deleted.

    :param legacy_package: The name of the legacy package to migrate (e.g. urlwatch).
    :param config_file: The new Path to the configuration file.
    :param jobs_file: The new Path to the jobs file.
    :param hooks_file: The new Path to the hooks file.
    :param ssdb_file: The new Path to the snapshot database file.
    """
    legacy_project_path = Path.home().joinpath(f'.{legacy_package}')
    leagacy_config_file = legacy_project_path.joinpath(f'{legacy_package}.yaml')
    legacy_urls_file = legacy_project_path.joinpath('urls.yaml')
    legacy_hooks_file = legacy_project_path.joinpath('hooks.py')
    legacy_cache_path = platformdirs.user_cache_path(legacy_package)
    legacy_cache_file = legacy_cache_path.joinpath('cache.db')
    for old_file, new_file in zip(
        (leagacy_config_file, legacy_urls_file, legacy_hooks_file, legacy_cache_file),
        (config_file, jobs_file, hooks_file, ssdb_file),
    ):
        if new_file and old_file.is_file() and not new_file.is_file():
            new_file.parent.mkdir(parents=True, exist_ok=True)
            shutil.copyfile(old_file, new_file)
            logger.warning(f"Copied {legacy_package} '{old_file}' file to {__project_name__} '{new_file}'.")
            logger.warning(f"You can safely delete '{old_file}'.")


def setup_logger(verbose: int | None = None, log_file: Path | None = None) -> None:
    """Set up the logger.

    :param verbose: the verbosity level (1 = INFO, 2 = ERROR).
    """
    log_level = None
    if verbose is not None:
        if verbose >= 3:
            log_level = 'NOTSET'
            # https://playwright.dev/python/docs/debug#verbose-api-logs
            os.environ['DEBUG'] = 'pw:api pytest -s'
        if verbose >= 2:
            log_level = 'DEBUG'
            # https://playwright.dev/python/docs/debug#verbose-api-logs
            os.environ['DEBUG'] = 'pw:api pytest -s'
        elif verbose == 1:
            log_level = 'INFO'

    if not verbose:
        sys.tracebacklimit = 0

    if log_file:
        handlers: tuple[logging.Handler, ...] | None = (logging.FileHandler(log_file),)
    else:
        handlers = None

    logging.basicConfig(
        format='%(asctime)s %(module)s[%(thread)s] %(levelname)s: %(message)s',
        level=log_level,
        handlers=handlers,
    )
    logger.info(f'{__project_name__}: {__version__} {__copyright__}')
    logger.info(
        f'{platform.python_implementation()}: {platform.python_version()} '
        f'{platform.python_build()} {platform.python_compiler()}'
    )
    logger.info(f'System: {platform.platform()}')


def teardown_logger(verbose: int | None = None) -> None:
    """Clean up logging.

    :param verbose: the verbosity level (1 = INFO, 2 = ERROR).
    """
    if verbose is not None:
        if verbose >= 2:
            # https://playwright.dev/python/docs/debug#verbose-api-logs
            os.environ.pop('DEBUG', None)


def _expand_glob_files(filename: Path, default_path: Path, ext: str | None = None) -> list[Path]:
    """Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.

    :param filename: The filename.
    :param default_path: The default directory.
    :param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.

    :returns: The filename, either original or one with path where found and/or extension.
    """
    search_filenames = [filename]

    # if ext is given, iterate both on raw filename and the filename with ext if different
    if ext and filename.suffix != ext:
        search_filenames.append(filename.with_suffix(ext))
        # also iterate on file pre-pended with 'jobs-'
        search_filenames.append(filename.with_stem(f'jobs-{filename.stem}').with_suffix(ext))

    # try as given
    for file in search_filenames:
        # https://stackoverflow.com/questions/56311703/globbing-absolute-paths-with-pathlib
        file_list = list(Path(file.anchor).glob(str(file.relative_to(file.anchor))))
        if any(f.is_file() for f in file_list):
            return file_list

        # no directory specified (and not in current one): add default one
        if not file.is_absolute() and not Path(file).parent == Path.cwd():
            file_list = list(default_path.glob(str(file)))
            if any(f.is_file() for f in file_list):
                return file_list

    # no matches found
    return [filename]


def locate_glob_files(filenames: list[Path], default_path: Path, ext: str | None = None) -> list[Path]:
    job_files = set()
    for filename in filenames:
        for file in _expand_glob_files(filename, default_path, ext):
            job_files.add(file)
    return list(job_files)


def locate_storage_file(filename: Path, default_path: Path, ext: str | None = None) -> Path:
    """Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.

    :param filename: The filename.
    :param default_path: The default directory.
    :param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.

    :returns: The filename, either original or one with path where found and/or extension.
    """
    search_filenames = [filename]

    # if ext is given, iterate both on raw filename and the filename with ext if different
    if ext and filename.suffix != ext:
        search_filenames.append(filename.with_suffix(ext))

    for file in search_filenames:
        # return if found
        if file.is_file():
            return file

        # no directory specified (and not in current one): add default one
        if file.parent == PurePath('.'):
            new_file = default_path.joinpath(file)
            if new_file.is_file():
                return new_file

    # no matches found
    return filename


def locate_storage_files(filename_list: list[Path], default_path: Path, ext: str | None = None) -> set[Path]:
    """Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.

    :param filename_list: The list of filenames.
    :param default_path: The default directory.
    :param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.

    :returns: The list filenames, either originals or ones with path where found and/or extension.
    """
    filenames = set()
    for filename in filename_list:
        filenames.add(locate_storage_file(filename, default_path, ext))
    return filenames


def first_run(command_config: CommandConfig) -> None:
    """Create configuration and jobs files.

    :param command_config: the CommandConfig containing the command line arguments selected.
    """
    if not command_config.config_file.is_file():
        command_config.config_file.parent.mkdir(parents=True, exist_ok=True)
        from webchanges.storage import YamlConfigStorage

        YamlConfigStorage.write_default_config(command_config.config_file)
        print(f'Created default config file at {command_config.config_file}')
        if not command_config.edit_config:
            print(f'> Edit it with {__project_name__} --edit-config')
    if not any(f.is_file() for f in command_config.jobs_files):
        command_config.jobs_files[0].parent.mkdir(parents=True, exist_ok=True)
        command_config.jobs_files[0].write_text(
            f'# {__project_name__} jobs file. See {__docs_url__}en/stable/jobs.html\n'
        )
        print(f'Created default jobs file at {command_config.jobs_files[0]}')
        if not command_config.edit:
            print(f'> Edit it with {__project_name__} --edit')


def load_hooks(hooks_file: Path) -> None:
    """Load hooks file."""
    if not hooks_file.is_file():
        warnings.warn(
            f'Hooks file not imported because {hooks_file} is not a file',
            ImportWarning,
        )
        return

    hooks_file_errors = file_ownership_checks(hooks_file)
    if hooks_file_errors:
        warnings.warn(
            f'Hooks file {hooks_file} not imported because '
            f" {' and '.join(hooks_file_errors)}.\n"
            f'(see {__docs_url__}en/stable/hooks.html#important-note-for-hooks-file)',
            ImportWarning,
        )
    else:
        logger.info(f'Importing hooks module from {hooks_file}')
        import_module_from_source('hooks', hooks_file)
        logger.info('Finished importing hooks module')


def handle_unitialized_actions(urlwatch_config: CommandConfig) -> None:
    """Handles CLI actions that do not require all classes etc. to be initialized (and command.py loaded). For speed
    purposes."""

    def _exit(arg: str | int | None) -> None:
        logger.info(f'Exiting with exit code {arg}')
        sys.exit(arg)

    def print_new_version() -> int:
        """Will print alert message if a newer version is found on PyPi."""
        print(f'{__project_name__} {__version__}.', end='')
        new_release = get_new_version_number(timeout=2)
        if new_release:
            print(
                f'\nNew release version {new_release} is available; we recommend updating using e.g. '
                f"'pip install -U {__project_name__}'."
            )
            return 0
        elif new_release == '':
            print(' You are running the latest release.')
            return 0
        else:
            print(' Error contacting PyPI to determine the latest release.')
            return 1

    def playwright_install_chrome() -> int:  # pragma: no cover
        """
        Replicates playwright.___main__.main() function, which is called by the playwright executable, in order to
        install the browser executable.

        :return: Playwright's executable return code.
        """
        try:
            from playwright._impl._driver import compute_driver_executable
        except ImportError:  # pragma: no cover
            raise ImportError('Python package playwright is not installed; cannot install the Chrome browser') from None

        driver_executable = compute_driver_executable()
        env = os.environ.copy()
        env['PW_CLI_TARGET_LANG'] = 'python'
        cmd = [str(driver_executable), 'install', 'chrome']
        logger.info(f"Running playwright CLI: {' '.join(cmd)}")
        completed_process = subprocess.run(cmd, env=env, capture_output=True, text=True)  # noqa: S603
        if completed_process.returncode:
            print(completed_process.stderr)
            return completed_process.returncode
        if completed_process.stdout:
            logger.info(f'Success! Output of Playwright CLI: {completed_process.stdout}')
        return 0

    if urlwatch_config.check_new:
        _exit(print_new_version())

    if urlwatch_config.install_chrome:  # pragma: no cover
        _exit(playwright_install_chrome())


def main() -> None:  # pragma: no cover
    """The entry point run when __name__ == '__main__'.

    Contains all the high-level logic to instantiate all classes that run the program.

    :raises NotImplementedError: If a `--database-engine` is specified that is not supported.
    :raises RuntimeError: If `--database-engine redis` is selected but `--cache` with a redis URI is not provided.
    """
    # Make sure that PendingDeprecationWarning are displayed from all modules (otherwise only those in __main__ are)
    warnings.filterwarnings('default', category=PendingDeprecationWarning)

    # Issue deprecation warning if running on minimum version supported
    python_version_warning()

    # Path where the config, jobs and hooks files are located
    if os.name != 'nt':
        config_path = platformdirs.user_config_path(__project_name__)  # typically ~/.config/{__project_name__}
    else:
        config_path = platformdirs.user_documents_path().joinpath(__project_name__)

    # Path where the snapshot database is located; typically ~/.local/share/{__project_name__} or
    # $XDG_DATA_HOME/{__project_name__} # in linux, ~/Library/Application Support/webchanges in macOS  and
    # or %LOCALAPPDATA%\{__project_name__}\{__project_name__} in Windows
    data_path = platformdirs.user_data_path(__project_name__, __project_name__.capitalize())

    # Default config, jobs, hooks and ssdb (database) files
    default_config_file = config_path.joinpath('config.yaml')
    default_jobs_file = config_path.joinpath('jobs.yaml')
    default_hooks_file = config_path.joinpath('hooks.py')
    default_ssdb_file = data_path.joinpath('snapshots.db')

    # Check for and if found migrate snapshot database file from version <= 3.21, which was called cache.db and located
    # in user_cache_path
    migrate_from_legacy('webchanges', ssdb_file=default_ssdb_file)

    # Check for and if found migrate legacy (urlwatch) files
    migrate_from_legacy('urlwatch', default_config_file, default_jobs_file, default_hooks_file, default_ssdb_file)

    # Parse command line arguments
    command_config = CommandConfig(
        sys.argv[1:],
        config_path,
        default_config_file,
        default_jobs_file,
        default_hooks_file,
        default_ssdb_file,
    )

    # Set up the logger to verbose if needed
    setup_logger(command_config.verbose, command_config.log_file)

    # For speed, run these here
    handle_unitialized_actions(command_config)

    # Only now, after configuring logging, we can load other modules
    from webchanges.command import UrlwatchCommand
    from webchanges.main import Urlwatch
    from webchanges.storage import (
        SsdbDirStorage,
        SsdbRedisStorage,
        SsdbSQLite3Storage,
        SsdbStorage,
        YamlConfigStorage,
        YamlJobsStorage,
    )

    # Locate config, job and hooks files
    command_config.config_file = locate_storage_file(command_config.config_file, command_config.config_path, '.yaml')
    command_config.jobs_files = locate_glob_files(command_config.jobs_files, command_config.config_path, '.yaml')
    command_config.hooks_files = locate_glob_files(command_config.hooks_files, command_config.config_path, '.py')

    # Check for first run
    if command_config.config_file == default_config_file and not Path(command_config.config_file).is_file():
        first_run(command_config)

    # Setup config file API
    config_storage = YamlConfigStorage(command_config.config_file)  # storage.py

    # load config (which for syntax checking requires hooks to be loaded too)
    if command_config.hooks_files:
        for hooks_file in command_config.hooks_files:
            load_hooks(hooks_file)
    config_storage.load()

    # Setup database API
    database_engine = (
        command_config.database_engine or config_storage.config.get('database', {}).get('engine') or 'sqlite3'
    )  # "or 'sqlite3'" is not needed except for a mypy bug; same for the "or 4" below
    max_snapshots = command_config.max_snapshots or config_storage.config.get('database', {}).get('max_snapshots') or 4
    if database_engine == 'sqlite3':
        ssdb_storage: SsdbStorage = SsdbSQLite3Storage(command_config.ssdb_file, max_snapshots)  # storage.py
    elif any(str(command_config.ssdb_file).startswith(prefix) for prefix in {'redis://', 'rediss://'}):
        ssdb_storage = SsdbRedisStorage(command_config.ssdb_file)  # storage.py
    elif database_engine.startswith('redis'):
        ssdb_storage = SsdbRedisStorage(database_engine)
    elif database_engine == 'textfiles':
        ssdb_storage = SsdbDirStorage(command_config.ssdb_file)  # storage.py
    elif database_engine == 'minidb':
        # legacy code imported only if needed (requires minidb, which is not a dependency)
        from webchanges.storage_minidb import SsdbMiniDBStorage

        ssdb_storage = SsdbMiniDBStorage(command_config.ssdb_file)  # storage.py
    else:
        raise NotImplementedError(f'Database engine {database_engine} not implemented')

    # Setup jobs file API
    jobs_storage = YamlJobsStorage(command_config.jobs_files)  # storage.py

    # Setup 'webchanges'
    urlwatcher = Urlwatch(command_config, config_storage, ssdb_storage, jobs_storage)  # main.py
    urlwatch_command = UrlwatchCommand(urlwatcher)  # command.py

    # Run 'webchanges', starting with processing command line arguments
    urlwatch_command.run()

    # Remove Playwright debug mode if there
    teardown_logger(command_config.verbose)


if __name__ == '__main__':
    main()

1	#!/usr/bin/env python3
2
3	"""Module containing the entry point: the function main()."""	2✔
4
5	# See config module for the command line arguments.
6
7	# The code below is subject to the license contained in the LICENSE file, which is part of the source code.
8
9	from __future__ import annotations	8✔
10
11	import logging	8✔
12	import os	8✔
13	import platform	8✔
14	import shutil	8✔
15	import signal	8✔
16	import subprocess # noqa: S404 Consider possible security implications associated with the subprocess module.	8✔
17	import sys	8✔
18	import warnings	8✔
19	from pathlib import Path, PurePath	8✔
20
21	import platformdirs	8✔
22
23	from webchanges import __copyright__, __docs_url__, __min_python_version__, __project_name__, __version__	8✔
24	from webchanges.config import CommandConfig	8✔
25	from webchanges.util import file_ownership_checks, get_new_version_number, import_module_from_source	8✔
26
27	# Ignore signal SIGPIPE ("broken pipe") for stdout (see https://github.com/thp/urlwatch/issues/77)
28	if os.name != 'nt': # Windows does not have signal.SIGPIPE	8!
29	signal.signal(signal.SIGPIPE, signal.SIG_DFL) # type: ignore[attr-defined] # not defined in Windows	8✔
30
31	logger = logging.getLogger(__name__)	8✔
32
33
34	def python_version_warning() -> None:	8✔
35	"""Check if we're running on the minimum Python version supported and if so print and issue a pending deprecation
36	warning."""
37	if sys.version_info[0:2] == __min_python_version__:	8✔
38	current_minor_version = '.'.join(str(n) for n in sys.version_info[0:2])	2!
39	next_minor_version = f'{__min_python_version__[0]}.{__min_python_version__[1] + 1}'	2✔
40	warning = (	2✔
41	f'Support for Python {current_minor_version} will be ending three years from the date Python '
42	f'{next_minor_version} was released'
43	)
44	print(f'WARNING: {warning}\n')	2✔
45	PendingDeprecationWarning(warning)	2✔
46
47
48	def migrate_from_legacy(	8✔
49	legacy_package: str,
50	config_file: Path \| None = None,
51	jobs_file: Path \| None = None,
52	hooks_file: Path \| None = None,
53	ssdb_file: Path \| None = None,
54	) -> None:
55	"""Check for existence of legacy files for configuration, jobs and Python hooks and migrate them (i.e. make a copy
56	to new folder and/or name). Original files are not deleted.
57
58	:param legacy_package: The name of the legacy package to migrate (e.g. urlwatch).
59	:param config_file: The new Path to the configuration file.
60	:param jobs_file: The new Path to the jobs file.
61	:param hooks_file: The new Path to the hooks file.
62	:param ssdb_file: The new Path to the snapshot database file.
63	"""
64	legacy_project_path = Path.home().joinpath(f'.{legacy_package}')	8✔
65	leagacy_config_file = legacy_project_path.joinpath(f'{legacy_package}.yaml')	8✔
66	legacy_urls_file = legacy_project_path.joinpath('urls.yaml')	8✔
67	legacy_hooks_file = legacy_project_path.joinpath('hooks.py')	8✔
68	legacy_cache_path = platformdirs.user_cache_path(legacy_package)	8✔
69	legacy_cache_file = legacy_cache_path.joinpath('cache.db')	8✔
70	for old_file, new_file in zip(	8✔
71	(leagacy_config_file, legacy_urls_file, legacy_hooks_file, legacy_cache_file),
72	(config_file, jobs_file, hooks_file, ssdb_file),
73	):
74	if new_file and old_file.is_file() and not new_file.is_file():	8!
75	new_file.parent.mkdir(parents=True, exist_ok=True)	×
76	shutil.copyfile(old_file, new_file)	×
77	logger.warning(f"Copied {legacy_package} '{old_file}' file to {__project_name__} '{new_file}'.")	×
78	logger.warning(f"You can safely delete '{old_file}'.")	×
79
80
81	def setup_logger(verbose: int \| None = None, log_file: Path \| None = None) -> None:	8✔
82	"""Set up the logger.
83
84	:param verbose: the verbosity level (1 = INFO, 2 = ERROR).
85	"""
86	log_level = None	8✔
87	if verbose is not None:	8!
88	if verbose >= 3:	×
89	log_level = 'NOTSET'	×
90	# https://playwright.dev/python/docs/debug#verbose-api-logs
91	os.environ['DEBUG'] = 'pw:api pytest -s'	×
92	if verbose >= 2:	×
93	log_level = 'DEBUG'	×
94	# https://playwright.dev/python/docs/debug#verbose-api-logs
95	os.environ['DEBUG'] = 'pw:api pytest -s'	×
96	elif verbose == 1:	×
97	log_level = 'INFO'	×
98
99	if not verbose:	8!
100	sys.tracebacklimit = 0	8✔
101
102	if log_file:	8!
NEW 103	handlers: tuple[logging.Handler, ...] \| None = (logging.FileHandler(log_file),)	×
104	else:
105	handlers = None	8✔
106
107	logging.basicConfig(	8✔
108	format='%(asctime)s %(module)s[%(thread)s] %(levelname)s: %(message)s',
109	level=log_level,
110	handlers=handlers,
111	)
112	logger.info(f'{__project_name__}: {__version__} {__copyright__}')	8✔
113	logger.info(	8✔
114	f'{platform.python_implementation()}: {platform.python_version()} '
115	f'{platform.python_build()} {platform.python_compiler()}'
116	)
117	logger.info(f'System: {platform.platform()}')	8✔
118
119
120	def teardown_logger(verbose: int \| None = None) -> None:	8✔
121	"""Clean up logging.
122
123	:param verbose: the verbosity level (1 = INFO, 2 = ERROR).
124	"""
125	if verbose is not None:	×
126	if verbose >= 2:	×
127	# https://playwright.dev/python/docs/debug#verbose-api-logs
128	os.environ.pop('DEBUG', None)	×
129
130
131	def _expand_glob_files(filename: Path, default_path: Path, ext: str \| None = None) -> list[Path]:	8✔
132	"""Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.
133
134	:param filename: The filename.
135	:param default_path: The default directory.
136	:param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.
137
138	:returns: The filename, either original or one with path where found and/or extension.
139	"""
140	search_filenames = [filename]	8✔
141
142	# if ext is given, iterate both on raw filename and the filename with ext if different
143	if ext and filename.suffix != ext:	8!
144	search_filenames.append(filename.with_suffix(ext))	8✔
145	# also iterate on file pre-pended with 'jobs-'
146	search_filenames.append(filename.with_stem(f'jobs-{filename.stem}').with_suffix(ext))	8✔
147
148	# try as given
149	for file in search_filenames:	8✔
150	# https://stackoverflow.com/questions/56311703/globbing-absolute-paths-with-pathlib
151	file_list = list(Path(file.anchor).glob(str(file.relative_to(file.anchor))))	8✔
152	if any(f.is_file() for f in file_list):	8!
153	return file_list	×
154
155	# no directory specified (and not in current one): add default one
156	if not file.is_absolute() and not Path(file).parent == Path.cwd():	8!
157	file_list = list(default_path.glob(str(file)))	8✔
158	if any(f.is_file() for f in file_list):	8!
159	return file_list	×
160
161	# no matches found
162	return [filename]	8✔
163
164
165	def locate_glob_files(filenames: list[Path], default_path: Path, ext: str \| None = None) -> list[Path]:	8✔
166	job_files = set()	8✔
167	for filename in filenames:	8✔
168	for file in _expand_glob_files(filename, default_path, ext):	8✔
169	job_files.add(file)	8✔
170	return list(job_files)	8✔
171
172
173	def locate_storage_file(filename: Path, default_path: Path, ext: str \| None = None) -> Path:	8✔
174	"""Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.
175
176	:param filename: The filename.
177	:param default_path: The default directory.
178	:param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.
179
180	:returns: The filename, either original or one with path where found and/or extension.
181	"""
182	search_filenames = [filename]	8✔
183
184	# if ext is given, iterate both on raw filename and the filename with ext if different
185	if ext and filename.suffix != ext:	8!
186	search_filenames.append(filename.with_suffix(ext))	8✔
187
188	for file in search_filenames:	8✔
189	# return if found
190	if file.is_file():	8!
191	return file	×
192
193	# no directory specified (and not in current one): add default one
194	if file.parent == PurePath('.'):	8!
195	new_file = default_path.joinpath(file)	8✔
196	if new_file.is_file():	8!
197	return new_file	×
198
199	# no matches found
200	return filename	8✔
201
202
203	def locate_storage_files(filename_list: list[Path], default_path: Path, ext: str \| None = None) -> set[Path]:	8✔
204	"""Searches for file both as specified and in the default directory, then retries with 'ext' extension if defined.
205
206	:param filename_list: The list of filenames.
207	:param default_path: The default directory.
208	:param ext: The extension, e.g. '.yaml', to add for searching if first scan fails.
209
210	:returns: The list filenames, either originals or ones with path where found and/or extension.
211	"""
212	filenames = set()	×
213	for filename in filename_list:	×
214	filenames.add(locate_storage_file(filename, default_path, ext))	×
215	return filenames	×
216
217
218	def first_run(command_config: CommandConfig) -> None:	8✔
219	"""Create configuration and jobs files.
220
221	:param command_config: the CommandConfig containing the command line arguments selected.
222	"""
223	if not command_config.config_file.is_file():	8!
224	command_config.config_file.parent.mkdir(parents=True, exist_ok=True)	8✔
225	from webchanges.storage import YamlConfigStorage	8✔
226
227	YamlConfigStorage.write_default_config(command_config.config_file)	8✔
228	print(f'Created default config file at {command_config.config_file}')	8✔
229	if not command_config.edit_config:	8!
230	print(f'> Edit it with {__project_name__} --edit-config')	8✔
231	if not any(f.is_file() for f in command_config.jobs_files):	8!
232	command_config.jobs_files[0].parent.mkdir(parents=True, exist_ok=True)	8✔
233	command_config.jobs_files[0].write_text(	8✔
234	f'# {__project_name__} jobs file. See {__docs_url__}en/stable/jobs.html\n'
235	)
236	print(f'Created default jobs file at {command_config.jobs_files[0]}')	8✔
237	if not command_config.edit:	8!
238	print(f'> Edit it with {__project_name__} --edit')	8✔
239
240
241	def load_hooks(hooks_file: Path) -> None:	8✔
242	"""Load hooks file."""
243	if not hooks_file.is_file():	4!
244	warnings.warn(	4✔
245	f'Hooks file not imported because {hooks_file} is not a file',
246	ImportWarning,
247	)
248	return	4✔
249
250	hooks_file_errors = file_ownership_checks(hooks_file)	×
251	if hooks_file_errors:	×
252	warnings.warn(	×
253	f'Hooks file {hooks_file} not imported because '
254	f" {' and '.join(hooks_file_errors)}.\n"
255	f'(see {__docs_url__}en/stable/hooks.html#important-note-for-hooks-file)',
256	ImportWarning,
257	)
258	else:
259	logger.info(f'Importing hooks module from {hooks_file}')	×
260	import_module_from_source('hooks', hooks_file)	×
261	logger.info('Finished importing hooks module')	×
262
263
264	def handle_unitialized_actions(urlwatch_config: CommandConfig) -> None:	8✔
265	"""Handles CLI actions that do not require all classes etc. to be initialized (and command.py loaded). For speed
266	purposes."""
267
268	def _exit(arg: str \| int \| None) -> None:	2✔
269	logger.info(f'Exiting with exit code {arg}')	2✔
270	sys.exit(arg)	2✔
271
272	def print_new_version() -> int:	2✔
273	"""Will print alert message if a newer version is found on PyPi."""
274	print(f'{__project_name__} {__version__}.', end='')	2✔
275	new_release = get_new_version_number(timeout=2)	2✔
276	if new_release:	2!
277	print(	×
278	f'\nNew release version {new_release} is available; we recommend updating using e.g. '
279	f"'pip install -U {__project_name__}'."
280	)
281	return 0	×
282	elif new_release == '':	2!
283	print(' You are running the latest release.')	2✔
284	return 0	2✔
285	else:
286	print(' Error contacting PyPI to determine the latest release.')	×
287	return 1	×
288
289	def playwright_install_chrome() -> int: # pragma: no cover
290	"""
291	Replicates playwright.___main__.main() function, which is called by the playwright executable, in order to
292	install the browser executable.
293
294	:return: Playwright's executable return code.
295	"""
296	try:
297	from playwright._impl._driver import compute_driver_executable
298	except ImportError: # pragma: no cover
299	raise ImportError('Python package playwright is not installed; cannot install the Chrome browser') from None
300
301	driver_executable = compute_driver_executable()
302	env = os.environ.copy()
303	env['PW_CLI_TARGET_LANG'] = 'python'
304	cmd = [str(driver_executable), 'install', 'chrome']
305	logger.info(f"Running playwright CLI: {' '.join(cmd)}")
306	completed_process = subprocess.run(cmd, env=env, capture_output=True, text=True) # noqa: S603
307	if completed_process.returncode:
308	print(completed_process.stderr)
309	return completed_process.returncode
310	if completed_process.stdout:
311	logger.info(f'Success! Output of Playwright CLI: {completed_process.stdout}')
312	return 0
313
314	if urlwatch_config.check_new:	2✔
315	_exit(print_new_version())	2✔
316
317	if urlwatch_config.install_chrome: # pragma: no cover
318	_exit(playwright_install_chrome())
319
320
321	def main() -> None: # pragma: no cover
322	"""The entry point run when __name__ == '__main__'.
323
324	Contains all the high-level logic to instantiate all classes that run the program.
325
326	:raises NotImplementedError: If a `--database-engine` is specified that is not supported.
327	:raises RuntimeError: If `--database-engine redis` is selected but `--cache` with a redis URI is not provided.
328	"""
329	# Make sure that PendingDeprecationWarning are displayed from all modules (otherwise only those in __main__ are)
330	warnings.filterwarnings('default', category=PendingDeprecationWarning)
331
332	# Issue deprecation warning if running on minimum version supported
333	python_version_warning()
334
335	# Path where the config, jobs and hooks files are located
336	if os.name != 'nt':
337	config_path = platformdirs.user_config_path(__project_name__) # typically ~/.config/{__project_name__}
338	else:
339	config_path = platformdirs.user_documents_path().joinpath(__project_name__)
340
341	# Path where the snapshot database is located; typically ~/.local/share/{__project_name__} or
342	# $XDG_DATA_HOME/{__project_name__} # in linux, ~/Library/Application Support/webchanges in macOS and
343	# or %LOCALAPPDATA%\{__project_name__}\{__project_name__} in Windows
344	data_path = platformdirs.user_data_path(__project_name__, __project_name__.capitalize())
345
346	# Default config, jobs, hooks and ssdb (database) files
347	default_config_file = config_path.joinpath('config.yaml')
348	default_jobs_file = config_path.joinpath('jobs.yaml')
349	default_hooks_file = config_path.joinpath('hooks.py')
350	default_ssdb_file = data_path.joinpath('snapshots.db')
351
352	# Check for and if found migrate snapshot database file from version <= 3.21, which was called cache.db and located
353	# in user_cache_path
354	migrate_from_legacy('webchanges', ssdb_file=default_ssdb_file)
355
356	# Check for and if found migrate legacy (urlwatch) files
357	migrate_from_legacy('urlwatch', default_config_file, default_jobs_file, default_hooks_file, default_ssdb_file)
358
359	# Parse command line arguments
360	command_config = CommandConfig(
361	sys.argv[1:],
362	config_path,
363	default_config_file,
364	default_jobs_file,
365	default_hooks_file,
366	default_ssdb_file,
367	)
368
369	# Set up the logger to verbose if needed
370	setup_logger(command_config.verbose, command_config.log_file)
371
372	# For speed, run these here
373	handle_unitialized_actions(command_config)
374
375	# Only now, after configuring logging, we can load other modules
376	from webchanges.command import UrlwatchCommand
377	from webchanges.main import Urlwatch
378	from webchanges.storage import (
379	SsdbDirStorage,
380	SsdbRedisStorage,
381	SsdbSQLite3Storage,
382	SsdbStorage,
383	YamlConfigStorage,
384	YamlJobsStorage,
385	)
386
387	# Locate config, job and hooks files
388	command_config.config_file = locate_storage_file(command_config.config_file, command_config.config_path, '.yaml')
389	command_config.jobs_files = locate_glob_files(command_config.jobs_files, command_config.config_path, '.yaml')
390	command_config.hooks_files = locate_glob_files(command_config.hooks_files, command_config.config_path, '.py')
391
392	# Check for first run
393	if command_config.config_file == default_config_file and not Path(command_config.config_file).is_file():
394	first_run(command_config)
395
396	# Setup config file API
397	config_storage = YamlConfigStorage(command_config.config_file) # storage.py
398
399	# load config (which for syntax checking requires hooks to be loaded too)
400	if command_config.hooks_files:
401	for hooks_file in command_config.hooks_files:
402	load_hooks(hooks_file)
403	config_storage.load()
404
405	# Setup database API
406	database_engine = (
407	command_config.database_engine or config_storage.config.get('database', {}).get('engine') or 'sqlite3'
408	) # "or 'sqlite3'" is not needed except for a mypy bug; same for the "or 4" below
409	max_snapshots = command_config.max_snapshots or config_storage.config.get('database', {}).get('max_snapshots') or 4
410	if database_engine == 'sqlite3':
411	ssdb_storage: SsdbStorage = SsdbSQLite3Storage(command_config.ssdb_file, max_snapshots) # storage.py
412	elif any(str(command_config.ssdb_file).startswith(prefix) for prefix in {'redis://', 'rediss://'}):
413	ssdb_storage = SsdbRedisStorage(command_config.ssdb_file) # storage.py
414	elif database_engine.startswith('redis'):
415	ssdb_storage = SsdbRedisStorage(database_engine)
416	elif database_engine == 'textfiles':
417	ssdb_storage = SsdbDirStorage(command_config.ssdb_file) # storage.py
418	elif database_engine == 'minidb':
419	# legacy code imported only if needed (requires minidb, which is not a dependency)
420	from webchanges.storage_minidb import SsdbMiniDBStorage
421
422	ssdb_storage = SsdbMiniDBStorage(command_config.ssdb_file) # storage.py
423	else:
424	raise NotImplementedError(f'Database engine {database_engine} not implemented')
425
426	# Setup jobs file API
427	jobs_storage = YamlJobsStorage(command_config.jobs_files) # storage.py
428
429	# Setup 'webchanges'
430	urlwatcher = Urlwatch(command_config, config_storage, ssdb_storage, jobs_storage) # main.py
431	urlwatch_command = UrlwatchCommand(urlwatcher) # command.py
432
433	# Run 'webchanges', starting with processing command line arguments
434	urlwatch_command.run()
435
436	# Remove Playwright debug mode if there
437	teardown_logger(command_config.verbose)
438
439
440	if __name__ == '__main__':
441	main()

mborsetti / webchanges / 11992535834

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous