def check_4_4_1():
# Check that there is only one Vcs field.
vcs_fields = []
with open('debian/control') as f:
source = next(Deb822.iter_paragraphs(f))
for name in source:
if name.lower() == 'vcs-browser':
continue
if name.lower().startswith('vcs-'):
vcs_fields.append(name)
if len(vcs_fields) > 1:
raise UpgradeCheckFailure(
"5.6.26", "package has more than one Vcs-<type> field")
# Check that Files entries don't refer to directories.
# They must be wildcards *in* the directories.
try:
with open('debian/copyright', 'r') as f:
copyright = Copyright(f, strict=False)
for para in copyright.all_files_paragraphs():
for glob in para.files:
if os.path.isdir(glob):
raise UpgradeCheckFailure(
"copyright-format",
"Wildcards are required to match the contents of "
"directories")
except FileNotFoundError:
pass
except NotMachineReadableError:
pass
def do_heuristics(fp):
c = Copyright()
num_licenses = 0
for l in fp.readlines():
if l.startswith("License:"):
num_licenses += 1
_, v = l.split(":", 1)
data = {"License": v.strip()}
lic_para = LicenseParagraph(data)
c.add_license_paragraph(lic_para)
if num_licenses > 0:
return c
else:
return None
def do_heuristics(fp):
c = Copyright()
num_licenses = 0
for l in fp.readlines():
if l.startswith("License:"):
num_licenses += 1
_, v = l.split(":", 1)
data = {"License": v.strip()}
lic_para = LicenseParagraph(data)
c.add_license_paragraph(lic_para)
if num_licenses > 0:
return c
return None
def _copyright_from_dep5(path: PathLike, copyright: Copyright) -> SpdxInfo:
"""Find the reuse information of *path* in the dep5 Copyright object."""
result = copyright.find_files_paragraph(Path(path).as_posix())
if result is None:
return SpdxInfo(set(), set())
return SpdxInfo(
set(map(_LICENSING.parse, [result.license.synopsis])),
set(map(str.strip, result.copyright.splitlines())),
)
def get_files_excluded(tree, subpath='', top_level=False):
if top_level:
path = os.path.join(subpath, 'copyright')
else:
path = os.path.join(subpath, 'debian', 'copyright')
with tree.get_file(path) as f:
try:
copyright = Copyright(f, strict=False)
except NotMachineReadableError:
return []
try:
return copyright.header["Files-Excluded"].split()
except KeyError:
return []
def _copyright(self) -> Optional[Copyright]:
if self._copyright_val == 0:
copyright_path = self.root / ".reuse/dep5"
try:
with copyright_path.open() as fp:
self._copyright_val = Copyright(fp)
except OSError:
_LOGGER.debug("no .reuse/dep5 file, or could not read it")
except DebianError:
_LOGGER.exception(_(".reuse/dep5 has syntax errors"))
# This check is a bit redundant, but otherwise I'd have to repeat
# this line under each exception.
if not self._copyright_val:
self._copyright_val = None
return self._copyright_val
groups = copyright_.get_group_dict(options)
for fileinfo in filetree:
if fileinfo.group:
continue
if isinstance(fileinfo, DirInfo):
continue
file_key = fileinfo.get_group_key(options)
group = groups.setdefault(file_key, Group(file_key))
group.add_file(fileinfo)
fileinfo.group = group
licenses = set()
c = Copyright()
# Print files paragraphs
for _, group in sorted(groups.items(), key=lambda i: i[1].sort_key(options)):
if not group.copyright_block_valid():
continue
licenses.update(group.licenses.keys())
if options.glob:
files = group.files.get_patterns()
else:
files = group.files.sorted_members()
if group.copyrights:
holders = '\n '.join(group.copyrights.sorted_members())
def copyright():
"""Create a dep5 Copyright object."""
with (RESOURCES_DIRECTORY / "fake_repository/.reuse/dep5").open() as fp:
return Copyright(fp)
def add_copyright_file(self, pkg_name, copyright):
# remove illegal characters from copyright
copyright, _ = remove_re.subn('', copyright)
xmlpkg = self.pkglist.append('pkglicense')
xmlpkg.et.attrib['name'] = pkg_name
txtnode = xmlpkg.append('text')
txtnode.et.text = copyright
bytesio = io.StringIO(unicode(txtnode.et.text))
try:
c = Copyright(bytesio)
files = []
for cc in c.all_files_paragraphs():
files.append((cc.files, cc.license.synopsis, cc.copyright))
xmlpkg.append('machinereadable')
xmllic = xmlpkg.append('debian_licenses')
seen = []
for f in files:
if f[1] in seen:
continue
seen.append(f[1])
ll = xmllic.append('license')
ll.et.text = f[1]
detailed = xmlpkg.append('detailed')
for f in files:
ff = detailed.append('files')
for g in f[0]:
gg = ff.append('glob')
gg.et.text = g
ll = ff.append('license')
ll.et.text = f[1]
cc = ff.append('copyright')
cc.et.text = f[2]
return
except Exception as e:
pass
bytesio.seek(0)
#textio = io.TextIOWrapper (bytesio, encoding='iso-8859-1')
c = do_heuristics(bytesio)
if not c is None:
lics = get_heuristics_license_list(c)
xmlpkg.append('heuristics')
xmllic = xmlpkg.append('debian_licenses')
for i in lics:
l = xmllic.append('license')
l.et.text = i
return
# Heuristics did not find anything either
# just return
return
def add_copyright_file(self, pkg_name, copyright_text):
# pylint: disable=too-many-locals
# remove illegal characters from copyright_text
copyright_text, _ = remove_re.subn('', copyright_text)
xmlpkg = self.pkglist.append('pkglicense')
xmlpkg.et.attrib['name'] = pkg_name
txtnode = xmlpkg.append('text')
txtnode.et.text = copyright_text
bytesio = io.StringIO(unicode(txtnode.et.text))
try:
c = Copyright(bytesio)
files = []
for cc in c.all_files_paragraphs():
files.append((cc.files, cc.license.synopsis, cc.copyright))
xmlpkg.append('machinereadable')
xmllic = xmlpkg.append('debian_licenses')
seen = []
for f in files:
if f[1] in seen:
continue
seen.append(f[1])
ll = xmllic.append('license')
ll.et.text = f[1]
detailed = xmlpkg.append('detailed')
for f in files:
ff = detailed.append('files')
for g in f[0]:
gg = ff.append('glob')
gg.et.text = g
ll = ff.append('license')
ll.et.text = f[1]
cc = ff.append('copyright')
cc.et.text = f[2]
return
except Exception:
pass
bytesio.seek(0)
c = do_heuristics(bytesio)
if c is not None:
lics = get_heuristics_license_list(c)
xmlpkg.append('heuristics')
xmllic = xmlpkg.append('debian_licenses')
for i in lics:
ltag = xmllic.append('license')
ltag.et.text = i
return
# Heuristics did not find anything either
# just return
return
def add_copyright_file(self, pkg_name, copyright_text):
# pylint: disable=too-many-locals
# remove illegal characters from copyright_text
copyright_text, _ = remove_re.subn('', copyright_text)
xmlpkg = self.pkglist.append('pkglicense')
xmlpkg.et.attrib['name'] = pkg_name
txtnode = xmlpkg.append('text')
txtnode.et.text = copyright_text
# in Python2 'txtnode.et.text' is a binary string whereas in Python3 it
# is a unicode string. So make sure that 'txtnode.et.text' ends up as a
# unicode string in both Python2 and Python3.
bytesio = io.StringIO(
txtnode.et.text.encode(encoding='utf-8',
errors='replace').decode(encoding='utf-8',
errors='replace'))
try:
c = Copyright(bytesio, strict=True)
files = []
# Note! Getters of cc can throw nasty exceptions!
for cc in c.all_files_paragraphs():
files.append((cc.files, cc.license.synopsis, cc.copyright))
except (NotMachineReadableError, MachineReadableFormatError) as E:
logging.warning("Error in copyright of package '%s': %s", pkg_name,
E)
except Warning as W:
logging.warning("Warning in copyrigh of package '%s' : %s",
pkg_name, W)
else:
xmlpkg.append('machinereadable')
xmllic = xmlpkg.append('debian_licenses')
seen = []
for f in files:
if f[1] in seen:
continue
seen.append(f[1])
ll = xmllic.append('license')
ll.et.text = f[1]
detailed = xmlpkg.append('detailed')
for f in files:
ff = detailed.append('files')
for g in f[0]:
gg = ff.append('glob')
gg.et.text = g
ll = ff.append('license')
ll.et.text = f[1]
cc = ff.append('copyright')
cc.et.text = f[2]
return
bytesio.seek(0)
c = do_heuristics(bytesio)
if c is not None:
lics = get_heuristics_license_list(c)
xmlpkg.append('heuristics')
xmllic = xmlpkg.append('debian_licenses')
for i in lics:
ltag = xmllic.append('license')
ltag.et.text = i
return
# Heuristics did not find anything either
# just return
return
