def parse_article_xml(self, document):
"""
Given article XML, parse
it and return an object representation
"""
try:
soup = parser.parse_document(document)
self.doi = parser.doi(soup)
if self.doi:
self.doi_id = self.get_doi_id(self.doi)
self.doi_url = self.get_doi_url(self.doi)
self.lens_url = self.get_lens_url(self.doi)
self.tweet_url = self.get_tweet_url(self.doi)
self.pub_date = parser.pub_date(soup)
self.pub_date_timestamp = parser.pub_date_timestamp(soup)
self.article_title = parser.title(soup)
self.article_type = parser.article_type(soup)
self.authors = parser.authors(soup)
self.authors_string = self.get_authors_string(self.authors)
self.related_articles = parser.related_article(soup)
self.is_poa = parser.is_poa(soup)
#self.subject_area = self.parse_subject_area(soup)
self.display_channel = parser.display_channel(soup)
return True
except:
return False
def repackage_archive_zip_to_pmc_zip(self, doi_id):
"repackage the zip file in the TMP_DIR to a PMC zip format"
# unzip contents
zip_input_dir = os.path.join(self.get_tmp_dir(), self.TMP_DIR)
zip_extracted_dir = os.path.join(self.get_tmp_dir(), self.JUNK_DIR)
zip_renamed_files_dir = os.path.join(self.get_tmp_dir(), self.RENAME_DIR)
pmc_zip_output_dir = os.path.join(self.get_tmp_dir(), self.INPUT_DIR)
archive_zip_name = glob.glob(zip_input_dir + "/*.zip")[0]
with zipfile.ZipFile(archive_zip_name, 'r') as myzip:
myzip.extractall(zip_extracted_dir)
# rename the files and profile the files
file_name_map = article_processing.rename_files_remove_version_number(
files_dir = zip_extracted_dir,
output_dir = zip_renamed_files_dir
)
if self.logger:
self.logger.info("FTPArticle running %s workflow for article %s, file_name_map"
% (self.workflow, self.doi_id))
self.logger.info(file_name_map)
# convert the XML
article_xml_file = glob.glob(zip_renamed_files_dir + "/*.xml")[0]
article_processing.convert_xml(xml_file=article_xml_file,
file_name_map=file_name_map)
# rezip the files into PMC zip format
soup = parser.parse_document(article_xml_file)
volume = parser.volume(soup)
pmc_zip_file_name = article_processing.new_pmc_zip_filename(self.journal, volume, doi_id)
with zipfile.ZipFile(os.path.join(pmc_zip_output_dir, pmc_zip_file_name), 'w',
zipfile.ZIP_DEFLATED, allowZip64=True) as new_zipfile:
dirfiles = article_processing.file_list(zip_renamed_files_dir)
for df in dirfiles:
filename = df.split(os.sep)[-1]
new_zipfile.write(df, filename)
return True
def to_soup(doc):
if isinstance(doc, basestring):
if os.path.exists(doc):
return parseJATS.parse_document(doc)
return parseJATS.parse_xml(doc)
# assume it's a file-like object and attempt to .read() it's contents
return parseJATS.parse_xml(doc.read())
def setUp(self):
kitchen_sink_xml = sample_xml("elife-kitchen-sink.xml")
# all of these methods are equivalent:
# self.soup = bss(open(kitchen_sink_xml, 'r').read())
# self.soup = parser.parse_xml(open(kitchen_sink_xml, 'r'))
self.soup = parser.parse_document(kitchen_sink_xml)
def parse_article_xml(self, document):
"""
Given article XML, parse
it and return an object representation
"""
try:
soup = parser.parse_document(document)
self.doi = parser.doi(soup)
if self.doi:
self.doi_id = self.get_doi_id(self.doi)
self.doi_url = self.get_doi_url(self.doi)
self.lens_url = self.get_lens_url(self.doi)
self.tweet_url = self.get_tweet_url(self.doi)
self.pub_date = parser.pub_date(soup)
self.pub_date_timestamp = parser.pub_date_timestamp(soup)
self.article_title = parser.title(soup)
self.article_type = parser.article_type(soup)
self.authors = parser.authors(soup)
self.authors_string = self.authors_string(self.authors)
self.related_articles = parser.related_article(soup)
self.is_poa = parser.is_poa(soup)
#self.subject_area = self.parse_subject_area(soup)
self.display_channel = parser.display_channel(soup)
return True
except:
return False
def transform_xml(xml_asset_path, identifier):
"modify the XML"
# remove history tags from XML for certain article types
root = parse.parse_article_xml(xml_asset_path)
soup = parser.parse_document(xml_asset_path)
root = transform_xml_history_tags(root, soup, identifier)
write_xml_file(root, xml_asset_path, identifier)
def to_soup(doc):
if isinstance(doc, basestring):
if os.path.exists(doc):
return parseJATS.parse_document(doc)
return parseJATS.parse_xml(doc)
# assume it's a file-like object and attempt to .read() it's contents
return parseJATS.parse_xml(doc.read())
def test_quickly(self):
struct = [
(parser.doi, u"10.7554/eLife.00013"),
(parser.journal_id, u"eLife"),
(parser.journal_title, u"eLife"),
(parser.journal_issn, u"2050-084X"),
(parser.publisher, u"eLife Sciences Publications, Ltd"),
]
for func, expected in struct:
soup = parser.parse_document(self.kitchen_sink_xml)
got = func(soup)
try:
self.assertEqual(got, expected)
except AssertionError:
print("failed on", func, "expected", expected, "got", got)
raise
soup = parser.parse_document(self.kitchen_sink_xml)
self.assertEqual(parser.journal_issn(soup, pub_format="electronic"),
u"2050-084X")
def article_wrapper(path,version=None):
soup = parser.parse_document(path)
# return a wrapper around the parser module that injects the soup when a function is called
return ParserWrapper(soup,path,version)
def test_sub_article(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.sub_article(soup)), expected_len)
def test_article_body(self, filename, expected_type):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(type(raw_parser.article_body(soup)), expected_type)
def test_abstract(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.abstract(soup)), expected_len)
def test_decision_letter(self, filename, expected_type):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(type(raw_parser.decision_letter(soup)), expected_type)
def test_author_response(self, filename, expected_type):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(type(raw_parser.author_response(soup)), expected_type)
def parse_jats_file(jats_file_name):
"parse the jats file into a BeautifulSoup object"
return parser.parse_document(jats_file_name)
def test_funding_group(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.funding_group(soup)), expected_len)
def test_fn_group(self, filename, content_type, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.fn_group(soup, content_type)),
expected_len)
def test_pub_id_uri(self, filename, ext_link_type, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.ext_link(soup, ext_link_type)),
expected_len)
def test_disp_formula(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.disp_formula(soup)), expected_len)
def test_table_wrap_foot(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.table_wrap_foot(soup)), expected_len)
def article_soup(self, xml_file):
return parser.parse_document(xml_file)
def article_soup(self, xml_file):
return parser.parse_document(xml_file)
def test_principal_award_recipient(self, filename, expected_len):
soup = parser.parse_document(sample_xml(filename))
self.assertEqual(len(raw_parser.principal_award_recipient(soup)),
expected_len)
def setUp(self):
self.kitchen_sink_xml = sample_xml("elife-kitchen-sink.xml")
self.xml = sample_xml("elife00013.xml")
self.soup = parser.parse_document(self.kitchen_sink_xml)
def build_article_from_xml(article_xml_filename, detail="brief"):
"""
Parse JATS XML with elifetools parser, and populate an
eLifePOA article object
Basic data crossref needs: article_id, doi, title, contributors with names set
detail="brief" is normally enough,
detail="full" will populate all the contributor affiliations that are linked by xref tags
"""
error_count = 0
soup = parser.parse_document(article_xml_filename)
# Get DOI
doi = parser.doi(soup)
# Create the article object
article = eLifePOA(doi, title=None)
# Related articles
article.related_articles = build_related_articles(parser.related_article(soup))
# Get publisher_id and set object manuscript value
publisher_id = parser.publisher_id(soup)
article.manuscript = publisher_id
# Set the articleType
article_type = parser.article_type(soup)
if article_type:
article.articleType = article_type
# title
article.title = parser.full_title(soup)
#print article.title
# abstract
article.abstract = clean_abstract(parser.full_abstract(soup))
# digest
article.digest = clean_abstract(parser.full_digest(soup))
# elocation-id
article.elocation_id = parser.elocation_id(soup)
# contributors
all_contributors = parser.contributors(soup, detail)
author_contributors = filter(lambda con: con.get('type')
in ['author', 'on-behalf-of'], all_contributors)
contrib_type = "author"
contributors = build_contributors(author_contributors, contrib_type)
contrib_type = "author non-byline"
authors = parser.authors_non_byline(soup, detail)
contributors_non_byline = build_contributors(authors, contrib_type)
article.contributors = contributors + contributors_non_byline
# license href
license = eLifeLicense()
license.href = parser.license_url(soup)
article.license = license
# article_category
article.article_categories = parser.category(soup)
# keywords
article.author_keywords = parser.keywords(soup)
# research organisms
article.research_organisms = parser.research_organism(soup)
# funding awards
article.funding_awards = build_funding(parser.full_award_groups(soup))
# references or citations
article.ref_list = build_ref_list(parser.refs(soup))
# components with component DOI
article.component_list = build_components(parser.components(soup))
# History dates
date_types = ["received", "accepted"]
for date_type in date_types:
history_date = parser.history_date(soup, date_type)
if history_date:
date_instance = eLifeDate(date_type, history_date)
article.add_date(date_instance)
# Pub date
pub_date = parser.pub_date(soup)
if pub_date:
date_instance = eLifeDate("pub", pub_date)
article.add_date(date_instance)
# Set the volume if present
volume = parser.volume(soup)
if volume:
article.volume = volume
article.is_poa = parser.is_poa(soup)
return article, error_count
def test_quickly2(self):
soup = parser.parse_document(self.xml)
self.assertEqual(raw_parser.article_type(soup), "research-article")
def test_mixed_citations(self):
self.soup = parser.parse_document(sample_xml("elife-kitchen-sink.xml"))
self.assertEqual(1, len(raw_parser.mixed_citations(self.soup)))
def article_wrapper(path, version=None):
soup = parser.parse_document(path)
# return a wrapper around the parser module that injects the soup when a function is called
return ParserWrapper(soup, path, version)
