def test_repeating(self, datadir, event_loop):
set_config({'unify_booktitle': [
[r'(?P<prefix>.*) remove(?P<suffix>.*)',
r'{prefix}{suffix}', 'priority:50', 'repeat', 'kind:plain'],
]})
unify_me = make_entry(
{'booktitle':
('Proceedings remove of remove some remove conference')})
u = Unifier()
sugg = u.unify_entry(unify_me)
assert (sugg.data['booktitle'][0] ==
((r'Proceedings of some conference'),
Suggestion.KIND_PLAIN))
# Test repeat-unifying suggestion
sugg = Suggestion('test', unify_me)
sugg.add_field('booktitle',
'Proceedings remove of remove some remove conference')
u.unify_suggestion(sugg)
assert(sugg.data['booktitle'][0] ==
('Proceedings of some conference', Suggestion.KIND_PLAIN))
def test_chaining(self, datadir, event_loop):
set_config({'unify_booktitle': [
[r'(?P<prefix>.*)first(?P<suffix>.*)',
r'{prefix}1st{suffix}', 'kind:plain', 'priority:50'],
[r'(?P<prefix>.*) IEEE(?P<suffix>.*)',
r'{prefix}{suffix}', 'kind:regex']
]})
unify_me = make_entry(
{'booktitle':
('Proceedings of the first IEEE conference on whatever')})
u = Unifier()
sugg = u.unify_entry(unify_me)
assert (sugg.data['booktitle'][0] ==
((r'Proceedings of the 1st conference on whatever'),
Suggestion.KIND_RE))
# Test chain-unifying suggestion
sugg = Suggestion('test', unify_me)
sugg.add_field('booktitle',
('Proceedings of the first'
' IEEE conference on whatever'))
u.unify_suggestion(sugg)
assert(sugg.data['booktitle'][0] ==
(r'Proceedings of the 1st conference on whatever',
Suggestion.KIND_RE))
def test_unify_suggestion(self, datadir, event_loop):
set_config({'unify_booktitle': [
[r'\d{4} IEEE (?P<name>[^\(]*) \((?P<short>[^\)]*)\)',
r'Proceedings of the \d*(th|st|nd|rd) {name} \({short}.*\)'],
]})
dummy_entry = make_entry({})
sugg = Suggestion('test', dummy_entry)
sugg.add_field('booktitle',
('2016 IEEE International Parallel and'
' Distributed Processing Symposium (IPDPS)'))
u = Unifier()
u.unify_suggestion(sugg)
assert (sugg.data['booktitle'][0] ==
(r'Proceedings of the \d*(th|st|nd|rd)'
r' International Parallel and Distributed'
r' Processing Symposium \(IPDPS.*\)',
Suggestion.KIND_RE))
def _query_blocking(self, entry, provider):
isbn = entry.data.get('isbn')
# Okay, we're actually going to make a HTTP request
self._ratelimit.get()
if not isbn:
self._ui.finish_subtask('ISBNQuery')
return None
if notisbn(isbn):
self._ui.finish_subtask('ISBNQuery')
return (None, "{} is not a valid ISBN.".format(isbn))
try:
bibtex_data = self._formatter(meta(isbn, service=provider))
except ISBNLibException as e:
self._ui.finish_subtask('ISBNQuery')
return (None, e)
except socket.timeout:
self._ui.finish_subtask('ISBNQuery')
raise RetrievalProblem("Socket timeout during"
" ISBN metadata retrieval")
try:
parsed_data = bibtexparser.loads(bibtex_data)
except:
self._ui.finish_subtask('ISBNQuery')
raise RetrievalProblem("Data from ISBN source could not be parsed")
if len(parsed_data.entries) != 1:
self._ui.finish_subtask('ISBNQuery')
raise RetrievalProblem(
"ISBN search did not return exactly one result.")
retrieved = Entry(parsed_data.entries[0], self._ui)
s = Suggestion("isbn_{}".format(provider), entry)
for (k, v) in retrieved.data.items():
if k.lower() == 'id':
continue
s.add_field(k, v)
for (first, last) in s.authors:
s.add_author(first, last)
for (first, last) in s.editors:
s.add_editor(first, last)
return (s, None)
def test_re_suggestion(self, datadir):
e = make_entry({
'title':
'This is some title.',
'booktitle':
"Proceedings of the 20th Conference on Something Awesome (CSA'20)"
})
s = Suggestion('test', e)
s.add_field('title', 'This is some title.', kind=Suggestion.KIND_RE)
s.add_field('booktitle',
r'Proceedings of the \d+(th|st|rd|nd) .* \(.*\)',
kind=Suggestion.KIND_RE)
d = Differ(e)
result = d.diff(s)
assert result == []
s = Suggestion('nonmatching_test', e)
s.add_field('booktitle', r'Nope', kind=Suggestion.KIND_RE)
result = d.diff(s)
assert len(result) == 1
def test_list_ignore_order(self, datadir):
e = make_entry({
'title': 'This is some title.',
'issn': '1234-5678, 2345-67890'
})
s = Suggestion('test', e)
s.add_field('title', 'This is some title.')
s.add_field('issn', '2345-67890, 1234-5678')
d = Differ(e)
result = d.diff(s)
assert result == []
async def _execute_query(self, entry, url, retry_number=0):
if not url:
self._ui.finish_subtask('MetaQuery')
return None
# Okay, we're actually going to make a HTTP request
await self._ratelimit.get()
try:
async with aiohttp.ClientSession() as session:
async with session.get(url,
headers=MetaSource.HEADERS) as resp:
status = resp.status
if status == 403:
try:
html = await resp.text()
if self._detect_captcha(html):
self._ui.finish_subtask('MetaQuery')
LOGGER.info(
(f"URL {url} requires a captcha to "
"be solved. Giving up."))
raise RetrievalProblem(
(f"URL {url} requires a "
"captcha to be solved.")
)
except:
pass
if retry_number == self._max_retries:
self._ui.finish_subtask('MetaQuery')
raise RetrievalProblem(
(f"URL {url} still results in 403 "
f"after {self._max_retries} retries."
" Giving up."))
LOGGER.debug((f"Got a 403 while accessing {url}."
f" Backing off. "
f"Retry {retry_number+1}..."))
await self._ratelimit.backoff()
await asyncio.sleep(self._retry_pause)
return await self._execute_query(entry, url,
retry_number+1)
if status != 200:
self._ui.finish_subtask('MetaQuery')
raise RetrievalProblem(
"Accessing URL {} returns status {}"
.format(url, status))
try:
html = await resp.text()
except UnicodeDecodeError:
self._ui.finish_subtask('MetaQuery')
raise RetrievalProblem(
f"Content at URL {url} could not be interpreted")
parser = MetadataHTMLParser(self._ui, str(resp.url))
parser.feed(html)
sugg = Suggestion("meta", entry)
for (k, v) in parser.get_metadata().items():
if isinstance(v, list):
sugg.add_field(k,
[remove_tags(vi) for vi in v])
else:
sugg.add_field(k, remove_tags(v))
for (first, last) in parser.get_authors():
sugg.add_author(first, last)
self._ui.finish_subtask('MetaQuery')
return sugg
except asyncio.TimeoutError:
self._ui.finish_subtask('MetaQuery')
LOGGER.error(f"Timeout trying to retrieve URL {url}")
raise RetrievalProblem(
f"Timeout trying to retrieve URL {url}")
def _query_blocking(self, entry):
doi = entry.get_probable_doi()
if not doi:
self._ui.finish_subtask('CrossrefQuery')
return None
try:
data = crossref_commons.retrieval.get_publication_as_json(doi)
except ValueError as e:
self._ui.finish_subtask('CrossrefQuery')
if str(e) == f"DOI {doi} does not exist":
# This isn't really an error, CrossRef just does not know
# about them
pass
else:
LOGGER.error((f"Error retrieving data for {entry.get_id()}. "
f"{e}"))
return None
except ConnectionError as e:
# TODO retry?
self._ui.finish_subtask('CrossrefQuery')
LOGGER.error(
(f"Connection error retrieving data for {entry.get_id()}. "
f"{e}"))
return None
s = Suggestion("crossref", entry)
# Special handling for type
btype = TYPE_MAPPING.get(data['type'])
if not btype:
LOGGER.warn(
"Type {} not found in crossref source. (Entry {})".format(
data['type'], entry.get_id()))
else:
s.add_field('entrytype', btype)
# Special handling for authors
for author_data in data.get('author', []):
s.add_author(
author_data.get('given', "").strip(),
author_data.get('family', "").strip())
# Special handling for editors
for editor_data in data.get('editor', []):
s.add_editor(
editor_data.get('given', "").strip(),
editor_data.get('family', "").strip())
# Special handling for journal / book title
if btype in ['journal-article', 'book-chapter']:
journal = flexistrip(data.get('container-title'))
if journal:
s.add_field('journal', journal)
# Special handling for URL. Only take it if it's not a DOI-Url
url = flexistrip(data.get('URL'))
if url and (CrossrefSource.DOI_URL_RE.match(url) is None):
s.add_field('url', url)
# All other fields
for field_from, field_to in FIELD_MAPPING.items():
if isinstance(field_to, dict):
if entry.data['entrytype'] in field_to:
field_to = field_to[entry.data['entrytype']]
else:
field_to = field_to.get('default')
if not field_to:
continue
if field_from in data:
s.add_field(field_to, flexistrip(data[field_from]))
self._ui.finish_subtask('CrossrefQuery')
return s
def _query_blocking(self, entry):
doi = entry.get_probable_doi()
if not doi:
self._ui.finish_subtask('DataCiteQuery')
return None
# Okay, we're actually going to make a HTTP request
self._ratelimit.get()
url = "https://api.datacite.org/dois/{}".format(
urllib.parse.quote(doi))
response = requests.get(url)
if response.status_code != 200:
self._ui.finish_subtask('DataCiteQuery')
return None
try:
data = response.json()
except ValueError:
LOGGER.warn("Response did not contain JSON")
self._ui.finish_subtask('DataCiteQuery')
return None
if 'errors' in data:
self._ui.finish_subtask('DataCiteQuery')
return None
attrs = data['data']['attributes']
s = Suggestion('datacite', entry)
# Authors
for i in range(0, len(attrs['creators'])):
adata = attrs['creators'][i]
if 'givenName' in adata and 'familyName' in adata:
s.add_author(adata['givenName'], adata['familyName'])
# Editors
for i in range(0, len(attrs['contributors'])):
adata = attrs['contributors'][i]
if adata.get('contributorType') == 'Editor':
if 'givenName' in adata and 'familyName' in adata:
s.add_editor(adata['givenName'], adata['familyName'])
# Title…s?
# TODO what happens if there are multiple titles?
if path_exists(attrs, ('titles', 0, 'title')):
s.add_field('title', attrs['titles'][0]['title'])
if 'publisher' in attrs:
s.add_field('publisher', attrs['publisher'])
if 'publicationYear' in attrs:
s.add_field('year', attrs['publicationYear'])
if 'url' in attrs:
s.add_field('url', attrs['url'])
ctype = None
if path_exists(attrs, ('container', 'type')):
ctype = attrs['container']['type']
cdata = attrs['container']
if ctype == 'Journal':
if 'title' in cdata:
s.add_field('journal', cdata['title'])
elif ctype == 'Book Series':
if 'title' in cdata:
s.add_field('booktitle', cdata['title'])
if ctype in ('Journal', 'Book Series'):
if 'volume' in cdata:
s.add_field('volume', cdata['volume'])
if 'issue' in cdata:
s.add_field('issue', cdata['issue'])
if cdata.get('identifierType') == 'ISSN':
s.add_field('issn', cdata['identifier'])
if 'firstPage' in cdata and 'lastPage' in cdata:
s.add_field(
'pages', '{}--{}'.format(cdata['firstPage'],
cdata['lastPage']))
if path_exists(attrs, ('type', 'bibtex')):
s.add_field('ENTRYTYPE', attrs['type']['bibtex'])
self._ui.finish_subtask('DataCiteQuery')
return s