Spaces:
Paused
Paused
Update helper_functions.py
Browse files- helper_functions.py +11 -7
helper_functions.py
CHANGED
@@ -1,16 +1,21 @@
|
|
1 |
-
from langchain_community.document_loaders import PyMuPDFLoader, TextLoader
|
2 |
from langchain_community.vectorstores import Qdrant
|
3 |
import os
|
4 |
|
5 |
-
def process_file(
|
|
|
|
|
|
|
|
|
|
|
6 |
# save the file temporarily
|
7 |
-
temp_file = "./temp/"+
|
8 |
with open(temp_file, "wb") as file:
|
9 |
-
file.write(
|
10 |
-
file_name =
|
11 |
|
12 |
documents = []
|
13 |
-
if
|
14 |
loader = PyMuPDFLoader(temp_file)
|
15 |
docs = loader.load()
|
16 |
documents.extend(docs)
|
@@ -22,7 +27,6 @@ def process_file(uploaded_file):
|
|
22 |
|
23 |
|
24 |
def add_to_qdrant(documents, embeddings, qdrant_client, collection_name):
|
25 |
-
|
26 |
Qdrant.from_documents(
|
27 |
documents,
|
28 |
embeddings,
|
|
|
1 |
+
from langchain_community.document_loaders import PyMuPDFLoader, TextLoader, WebBaseLoader
|
2 |
from langchain_community.vectorstores import Qdrant
|
3 |
import os
|
4 |
|
5 |
+
def process_file(file_or_url):
|
6 |
+
if isinstance(file_or_url, str) and file_or_url.startswith(('http://', 'https://')):
|
7 |
+
# Handle URL
|
8 |
+
loader = WebBaseLoader(file_or_url)
|
9 |
+
docs = loader.load()
|
10 |
+
documents.extend(docs)
|
11 |
# save the file temporarily
|
12 |
+
temp_file = "./temp/"+file_or_url.path
|
13 |
with open(temp_file, "wb") as file:
|
14 |
+
file.write(file_or_url.content)
|
15 |
+
file_name = file_or_url.name
|
16 |
|
17 |
documents = []
|
18 |
+
if file_or_url.path.endswith(".pdf"):
|
19 |
loader = PyMuPDFLoader(temp_file)
|
20 |
docs = loader.load()
|
21 |
documents.extend(docs)
|
|
|
27 |
|
28 |
|
29 |
def add_to_qdrant(documents, embeddings, qdrant_client, collection_name):
|
|
|
30 |
Qdrant.from_documents(
|
31 |
documents,
|
32 |
embeddings,
|