def crawl(self, directory=None, source=None, meta={}):
source = source or directory
source = Source.create({
'foreign_id': 'directory:%s' % slugify(source),
'label': source
})
db.session.commit()
source_id = source.id
if os.path.isfile(directory):
self.crawl_file(source_id, directory, meta)
directory = directory or os.getcwd()
directory = directory.encode('utf-8')
for (dirname, dirs, files) in os.walk(directory):
dirparts = [
d for d in dirname.split(os.path.sep) if d in SKIP_DIRECTORIES
]
if len(dirparts):
continue
log.info("Descending: %r", dirname)
for file_name in files:
dirname = string_value(dirname)
file_name = string_value(file_name)
if file_name in SKIP_FILES:
continue
file_path = os.path.join(dirname, file_name)
self.crawl_file(source_id, file_path, meta)
def source(self):
if not hasattr(self, '_source'):
self._source = Source.create({
'foreign_id': self.SOURCE_ID,
'label': self.SOURCE_LABEL or self.SOURCE_ID
})
db.session.commit()
return self._source
def crawl_source(self, engine, foreign_id, data):
source = Source.create({
'foreign_id': foreign_id,
'label': data.get('label')
})
db.session.commit()
meta_base = data.get('meta', {})
for name, query in data.get('queries', {}).items():
self.crawl_query(engine, source, meta_base, name, query)
def crawl_source(self, engine, foreign_id, data):
source = Source.create({
'foreign_id': foreign_id,
'label': data.get('label')
})
db.session.commit()
meta_base = data.get('meta', {})
for name, query in data.get('queries', {}).items():
self.crawl_query(engine, source, meta_base, name, query)
def source(self):
if not hasattr(self, '_source'):
self._source = Source.create({
'foreign_id': self.SOURCE_ID,
'label': self.SOURCE_LABEL or self.SOURCE_ID
})
db.session.commit()
db.session.add(self._source)
return self._source
def crawl(self):
for base_url in SITES:
print 'Working on base_url: {}'.format(base_url)
self.attributes = SITES[base_url]
self.label = self.attributes['label']
Source.create({'label': self.label, 'foreign_id': 'blacklight'})
db.session.commit()
self.failed_articles = 0
page_count = self.get_page_count(base_url)
print "Pages: {}".format(page_count)
page_number = 1
while (page_number <= page_count):
if self.failed_articles >= FAILED_LIMIT:
log.warning('Failure limit reach: {}'.format(FAILED_LIMIT))
break
self.crawl_page(base_url, page_number, page_count)
page_number += 1
def crawl(self):
for base_url in SITES:
print 'Working on base_url: {}'.format(base_url)
self.attributes = SITES[base_url]
self.label = self.attributes['label']
Source.create({
'label': self.label,
'foreign_id': 'blacklight'
})
db.session.commit()
self.failed_articles = 0
page_count = self.get_page_count(base_url)
print "Pages: {}".format(page_count)
page_number = 1
while (page_number <= page_count):
if self.failed_articles >= FAILED_LIMIT:
log.warning('Failure limit reach: {}'.format(FAILED_LIMIT))
break
self.crawl_page(base_url, page_number, page_count)
page_number += 1
def crawl(self, directory=None, source=None):
source = source or directory
source = Source.create({
'foreign_id': 'directory:%s' % slugify(source),
'label': source
})
if os.path.isfile(directory):
meta = self.metadata()
meta.file_name = directory
self.emit_file(source, meta, directory)
directory = directory or os.getcwd()
directory = directory.encode('utf-8')
for (dirname, dirs, files) in os.walk(directory):
log.info("Descending: %r", dirname)
for file_name in files:
if file_name in SKIP_FILES:
continue
file_path = os.path.join(dirname, file_name)
if not os.path.isfile(file_path):
continue
try:
meta = self.metadata()
if isinstance(file_name, six.text_type):
meta.source_path = file_path
else:
enc = chardet.detect(file_name)
enc = enc.get('encoding')
try:
meta.source_path = file_path.decode(enc)
except:
meta.source_path = file_path.decode('ascii', 'ignore')
self.emit_file(source, meta, file_path)
except Exception as ex:
log.exception(ex)
process.exception(process.INDEX, component=self.name,
source_location=directory,
source_id=source.id, exception=ex)
def crawl_item(self, item, source):
source_data = item.meta.get('source', {})
source_fk = source_data.pop('foreign_id', source)
if source_fk is None:
raise ValueError("No foreign_id for source given: %r" % item)
if source_fk not in self.sources:
label = source_data.get('label', source_fk)
self.sources[source_fk] = Source.create({
'foreign_id': source_fk,
'label': label
})
if source_data.get('public'):
Permission.grant_foreign(self.sources[source_fk],
Role.SYSTEM_GUEST, True, False)
db.session.commit()
log.info('Import: %r', item.identifier)
meta = self.normalize_metadata(item)
ingest_file(self.sources[source_fk].id,
meta,
item.data_path,
move=False)
def crawl(self, directory=None, source=None):
source = source or directory
source = Source.create({
'foreign_id': 'directory:%s' % slugify(source),
'label': source
})
if os.path.isfile(directory):
meta = self.metadata()
meta.file_name = directory
self.emit_file(source, meta, directory)
directory = directory or os.getcwd()
directory = directory.encode('utf-8')
for (dirname, dirs, files) in os.walk(directory):
log.info("Descending: %r", dirname)
for file_name in files:
if file_name in SKIP_FILES:
continue
file_path = os.path.join(dirname, file_name)
if not os.path.isfile(file_path):
continue
try:
meta = self.metadata()
if isinstance(file_name, six.text_type):
meta.source_path = file_path
else:
enc = chardet.detect(file_name)
enc = enc.get('encoding')
try:
meta.source_path = file_path.decode(enc)
except:
meta.source_path = file_path.decode('ascii', 'ignore')
self.emit_file(source, meta, file_path)
except Exception as ex:
log.exception(ex)
def create():
authz.require(authz.logged_in())
src = Source.create(request_data(), current_user)
db.session.commit()
return view(src.slug)
def create_source(self, **data):
if 'foreign_id' not in data:
data['foreign_id'] = self.name
return Source.create(data)
def create_source(self, **data):
if 'foreign_id' not in data:
data['foreign_id'] = self.name
return Source.create(data)
def create():
authz.require(authz.is_admin())
src = Source.create(request_data(), current_user)
db.session.commit()
return view(src.slug)
