def ingest(document_id, file_path=None):
"""Process a given document by extracting its contents.
This may include creating or updating child documents."""
document = Document.by_id(document_id)
if document is None:
log.error("Could not find document: %s", document_id)
return
# Work path will be used by storagelayer to cache a local
# copy of data from an S3-based archive, and by ingestors
# to perform processing and generate intermediary files.
work_path = mkdtemp(prefix="aleph.ingest.")
if file_path is None:
file_path = archive.load_file(document.content_hash,
file_name=document.safe_file_name,
temp_path=work_path)
try:
manager = get_manager()
result = DocumentResult(manager, document, file_path=file_path)
get_manager().ingest(file_path, result=result, work_path=work_path)
log.debug('Ingested [%s:%s]: %s', document.id, document.schema,
document.name)
db.session.commit()
process_document(document)
except Exception:
db.session.rollback()
document = Document.by_id(document_id)
log.exception("Ingest failed [%s]: %s", document.id, document.name)
finally:
# Removing the temp_path given to storagelayer makes it redundant
# to also call cleanup on the archive.
remove_directory(work_path)
def handle_child(self, parent, file_path, title=None, mime_type=None,
id=None, file_name=None):
file_path = decode_path(file_path)
assert id is not None, (parent, file_path)
doc = Document.by_keys(parent_id=parent.document.id,
collection=parent.document.collection,
foreign_id=id)
doc.title = title or doc.meta.get('title')
doc.file_name = file_name or doc.meta.get('file_name')
doc.mime_type = mime_type or doc.meta.get('mime_type')
from aleph.logic.documents.ingest import ingest_document
ingest_document(doc, file_path, role_id=parent.document.uploader_id)
return DocumentResult(self, doc, file_path=file_path)
def ingest(document_id, file_path=None, refresh=False):
"""Process a given document by extracting its contents.
This may include creating or updating child documents."""
document = Document.by_id(document_id)
if document is None:
log.error("Could not find document: %s", document_id)
return
# Work path will be used by storagelayer to cache a local
# copy of data from an S3-based archive, and by ingestors
# to perform processing and generate intermediary files.
work_path = mkdtemp(prefix="aleph.ingest.")
if file_path is None:
file_path = archive.load_file(document.content_hash,
file_name=document.safe_file_name,
temp_path=work_path)
try:
manager = get_manager()
result = DocumentResult(manager, document, file_path=file_path)
get_manager().ingest(file_path, result=result, work_path=work_path)
document.status = Document.STATUS_SUCCESS
log.debug('Ingested [%s:%s]: %s', document.id, document.schema,
document.name)
if document.collection.casefile and not refresh:
params = {'collection': document.collection, 'document': document}
publish(Events.INGEST_DOCUMENT,
actor_id=document.uploader_id,
params=params)
db.session.commit()
except Exception:
db.session.rollback()
document = Document.by_id(document_id)
log.exception("Ingest failed [%s]: %s", document.id, document.name)
document.status = Document.STATUS_FAIL
db.session.commit()
finally:
# Removing the temp_path given to storagelayer makes it redundant
# to also call cleanup on the archive.
remove_directory(work_path)
extract_document_tags(document)
# delete_entity(document.id, exclude=document.schema)
index_document(document)
refresh_entity(document)