apunkt/trustgraph
1
0
Fork 0
mirror of https://github.com/trustgraph-ai/trustgraph.git synced 2026-04-27 17:36:23 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Feature / collections (#96)

Browse source 
* Update schema defs for source -> metadata
* Migrate to use metadata part of schema, also add metadata to triples & vecs
* Add user/collection metadata to query
* Use user/collection in RAG
* Write and query working on triples
This commit is contained in:
cybermaggedon 2024-10-02 18:14:29 +01:00 • committed by GitHub
parent 709221fa10
commit b0f4c58200
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
31 changed files with 459 additions and 251 deletions
Download patch file Download diff file Expand all files Collapse all files

14
trustgraph-flow/trustgraph/decoding/pdf/pdf_decoder.py
View file

@ -8,7 +8,7 @@ import tempfile
import base64
from langchain_community.document_loaders import PyPDFLoader
from ... schema import Document, TextDocument, Source
from ... schema import Document, TextDocument, Metadata
from ... schema import document_ingest_queue, text_ingest_queue
from ... log_level import LogLevel
from ... base import ConsumerProducer
@ -45,7 +45,7 @@ class Processor(ConsumerProducer):
v = msg.value()
print(f"Decoding {v.source.id}...", flush=True)
print(f"Decoding {v.metadata.id}...", flush=True)
with tempfile.NamedTemporaryFile(delete_on_close=False) as fp:
@ -59,12 +59,14 @@ class Processor(ConsumerProducer):
for ix, page in enumerate(pages):
id = v.source.id + "-p" + str(ix)
id = v.metadata.id + "-p" + str(ix)
r = TextDocument(
source=Source(
source=v.source.source,
title=v.source.title,
metadata=Metadata(
source=v.metadata.source,
title=v.metadata.title,
id=id,
user=v.metadata.user,
collection=v.metadata.collection,
),
text=page.page_content.encode("utf-8"),
)