Source code for spack.cmd.license

# Copyright 2013-2021 Lawrence Livermore National Security, LLC and other
# Spack Project Developers. See the top-level COPYRIGHT file for details.
# SPDX-License-Identifier: (Apache-2.0 OR MIT)

from __future__ import print_function

import os
import re
from collections import defaultdict

import llnl.util.filesystem as fs
import llnl.util.tty as tty

import spack.paths
from spack.util.executable import which

description = 'list and check license headers on files in spack'
section = "developer"
level = "long"

#: need the git command to check new files
git = which('git')

#: SPDX license id must appear in the first <license_lines> lines of a file
license_lines = 7

#: Spack's license identifier
apache2_mit_spdx = "(Apache-2.0 OR MIT)"

#: regular expressions for licensed files.
licensed_files = [
    # spack scripts

    # all of spack core

    # rst files in documentation

    # 2 files in external

    # shell scripts in share

    # action workflows

    # all packages

#: licensed files that can have LGPL language in them
#: so far, just this command -- so it can find LGPL things elsewhere
lgpl_exceptions = [

def _all_spack_files(root=spack.paths.prefix):
    """Generates root-relative paths of all files in the spack repository."""
    visited = set()
    for cur_root, folders, files in os.walk(root):
        for filename in files:
            path = os.path.realpath(os.path.join(cur_root, filename))

            if path not in visited:
                yield os.path.relpath(path, root)

def _licensed_files(args):
    for relpath in _all_spack_files(args.root):
        if any(regex.match(relpath) for regex in licensed_files):
            yield relpath

[docs]def list_files(args): """list files in spack that should have license headers""" for relpath in sorted(_licensed_files(args)): print(os.path.join(spack.paths.spack_root, relpath))
# Error codes for license verification. All values are chosen such that # bool(value) evaluates to True OLD_LICENSE, SPDX_MISMATCH, GENERAL_MISMATCH = range(1, 4) strict_date = r'Copyright 2013-2021'
[docs]class LicenseError(object): def __init__(self): self.error_counts = defaultdict(int)
[docs] def add_error(self, error): self.error_counts[error] += 1
[docs] def has_errors(self): return sum(self.error_counts.values()) > 0
[docs] def error_messages(self): total = sum(self.error_counts.values()) missing = self.error_counts[GENERAL_MISMATCH] spdx_mismatch = self.error_counts[SPDX_MISMATCH] old_license = self.error_counts[OLD_LICENSE] return ( '%d improperly licensed files' % (total), 'files with wrong SPDX-License-Identifier: %d' % spdx_mismatch, 'files with old license header: %d' % old_license, 'files not containing expected license: %d' % missing)
def _check_license(lines, path): license_lines = [ r'Copyright 2013-(?:202[01]) Lawrence Livermore National Security, LLC and other', # noqa: E501 r'Spack Project Developers\. See the top-level COPYRIGHT file for details.', # noqa: E501 r'SPDX-License-Identifier: \(Apache-2\.0 OR MIT\)' ] found = [] for line in lines: line = re.sub(r'^[\s#\%\.]*', '', line) line = line.rstrip() for i, license_line in enumerate(license_lines): if re.match(license_line, line): # The first line of the license contains the copyright date. # We allow it to be out of date but print a warning if it is # out of date. if i == 0: if not, line): tty.debug('{0}: copyright date mismatch'.format(path)) found.append(i) if len(found) == len(license_lines) and found == list(sorted(found)): return def old_license(line, path): if'This program is free software', line): print('{0}: has old LGPL license header'.format(path)) return OLD_LICENSE # If the SPDX identifier is present, then there is a mismatch (since it # did not match the above regex) def wrong_spdx_identifier(line, path): m ='SPDX-License-Identifier: ([^\n]*)', line) if m and != apache2_mit_spdx: print('{0}: SPDX license identifier mismatch' '(expecting {1}, found {2})' .format(path, apache2_mit_spdx, return SPDX_MISMATCH checks = [old_license, wrong_spdx_identifier] for line in lines: for check in checks: error = check(line, path) if error: return error print('{0}: the license header at the top of the file does not match the \ expected format'.format(path)) return GENERAL_MISMATCH
[docs]def verify(args): """verify that files in spack have the right license header""" license_errors = LicenseError() for relpath in _licensed_files(args): path = os.path.join(args.root, relpath) with open(path) as f: lines = [line for line in f][:license_lines] error = _check_license(lines, path) if error: license_errors.add_error(error) if license_errors.has_errors(): tty.die(*license_errors.error_messages()) else: tty.msg('No license issues found.')
[docs]def setup_parser(subparser): subparser.add_argument( '--root', action='store', default=spack.paths.prefix, help='scan a different prefix for license issues') sp = subparser.add_subparsers(metavar='SUBCOMMAND', dest='license_command') sp.add_parser('list-files', help=list_files.__doc__) sp.add_parser('verify', help=verify.__doc__) sp.add_parser('update-copyright-year', help=update_copyright_year.__doc__)
[docs]def license(parser, args): if not git: tty.die('spack license requires git in your environment') licensed_files[:] = [re.compile(regex) for regex in licensed_files] commands = { 'list-files': list_files, 'verify': verify, 'update-copyright-year': update_copyright_year, } return commands[args.license_command](args)