apunkt/trustgraph
1
0
Fork 0
mirror of https://github.com/trustgraph-ai/trustgraph.git synced 2026-07-21 19:21:03 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

Fix error, putting flow stuff into API

Browse source
This commit is contained in:
Cyber MacGeddon 2025-05-03 10:00:16 +01:00
parent 031da4842d
commit b212dd3aae
11 changed files with 155 additions and 134 deletions
Download patch file Download diff file Expand all files Collapse all files

81
trustgraph-cli/scripts/tg-load-pdf
View file

@ -27,13 +27,14 @@ class Loader:
def __init__(
self,
url,
flow_id,
user,
collection,
metadata,
pulsar_api_key=None,
):
self.api = Api(url)
self.api = Api(url).flow(flow_id)
self.user = user
self.collection = collection
@ -60,14 +61,15 @@ class Loader:
self.api.load_document(
document=data, id=id, metadata=self.metadata,
# user=self.user,
# collection=self.collection,
user=self.user,
collection=self.collection,
)
print(f"{file}: Loaded successfully.")
except Exception as e:
print(f"{file}: Failed: {str(e)}", flush=True)
raise e
def main():
@ -155,53 +157,46 @@ def main():
args = parser.parse_args()
while True:
try:
try:
document = DigitalDocument(
id,
name=args.name,
description=args.description,
copyright_notice=args.copyright_notice,
copyright_holder=args.copyright_holder,
copyright_year=args.copyright_year,
license=args.license,
url=args.document_url,
keywords=args.keyword,
)
document = DigitalDocument(
id,
name=args.name,
description=args.description,
copyright_notice=args.copyright_notice,
copyright_holder=args.copyright_holder,
copyright_year=args.copyright_year,
license=args.license,
url=args.document_url,
keywords=args.keyword,
if args.publication_organization:
org = Organization(
id=to_uri(PREF_ORG, hash(args.publication_organization)),
name=args.publication_organization,
)
document.publication = PublicationEvent(
id = to_uri(PREF_PUBEV, str(uuid.uuid4())),
organization=org,
description=args.publication_description,
start_date=args.publication_date,
end_date=args.publication_date,
)
if args.publication_organization:
org = Organization(
id=to_uri(PREF_ORG, hash(args.publication_organization)),
name=args.publication_organization,
)
document.publication = PublicationEvent(
id = to_uri(PREF_PUBEV, str(uuid.uuid4())),
organization=org,
description=args.publication_description,
start_date=args.publication_date,
end_date=args.publication_date,
)
p = Loader(
url=args.url,
flow_id = args.flow_id,
user=args.user,
collection=args.collection,
metadata=document,
)
p = Loader(
url=args.url,
user=args.user,
collection=args.collection,
metadata=document,
)
p.load(args.files)
p.load(args.files)
except Exception as e:
print("All done.")
break
except Exception as e:
print("Exception:", e, flush=True)
print("Will retry...", flush=True)
time.sleep(10)
print("Exception:", e, flush=True)
main()