def import_aliases(pipeline, fh):
""" Import aliases from a CSV source. This will not create
new entities, but re-name existing entities or merge two
entities if one's name is given as an alias for the other. """
importer = CSVImporter(fh)
canonical_column, alias_column = None, None
for k, v in pipeline.config.get('mapping', {}).items():
if v.get('attribute') == 'alias':
alias_column = k
elif v.get('attribute') == 'canonical':
canonical_column = k
for i, row in enumerate(importer):
source_url = _row_source_url(pipeline, row)
entities.apply_alias(pipeline.project, pipeline.author,
row.get(canonical_column),
row.get(alias_column),
source_url=source_url)
if i % 100 == 0:
percentage = int((float(i) / max(1, len(importer))) * 100)
pipeline.percent_complete = percentage
db.session.commit()
def import_aliases(pipeline, fh):
""" Import aliases from a CSV source. This will not create
new entities, but re-name existing entities or merge two
entities if one's name is given as an alias for the other. """
importer = CSVImporter(fh)
canonical_column, alias_column = None, None
for k, v in pipeline.config.get('mapping', {}).items():
if v.get('attribute') == 'alias':
alias_column = k
elif v.get('attribute') == 'canonical':
canonical_column = k
for i, row in enumerate(importer):
source_url = _row_source_url(pipeline, row)
entities.apply_alias(pipeline.project, pipeline.author,
row.get(canonical_column),
row.get(alias_column),
source_url=source_url)
if i % 100 == 0:
percentage = int((float(i) / max(1, len(importer))) * 100)
pipeline.percent_complete = percentage
db.session.commit()
def import_aliases(project, author, path):
""" Import aliases from a CSV file. This will not create new entities, but
re-name existing entities or merge two entities if one's name is given as
an alias for the other. """
with open(path, 'r') as fh:
reader = DictReader(fh)
for i, row in enumerate(reader):
data = {}
for k, v in row.items():
k = k.lower().strip()
data[k] = v
assert 'canonical' in data, 'No "canonical" column!'
assert 'alias' in data, 'No "alias" column!'
entities.apply_alias(project, author, data.get('canonical'),
data.get('alias'))
if i % 1000 == 0:
db.session.commit()
db.session.commit()
def import_aliases(project, author, path):
""" Import aliases from a CSV file. This will not create new entities, but
re-name existing entities or merge two entities if one's name is given as
an alias for the other. """
with open(path, 'r') as fh:
reader = DictReader(fh)
for i, row in enumerate(reader):
data = {}
for k, v in row.items():
k = k.lower().strip()
data[k] = v
assert 'canonical' in data, 'No "canonical" column!'
assert 'alias' in data, 'No "alias" column!'
entities.apply_alias(project, author,
data.get('canonical'),
data.get('alias'))
if i % 1000 == 0:
db.session.commit()
db.session.commit()
