ArturG9 commited on
Commit
bccf29a
1 Parent(s): ccb7696

Update utills.py

Browse files
Files changed (1) hide show
  1. utills.py +18 -1
utills.py CHANGED
@@ -33,7 +33,7 @@ def get_session_history(session_id: str):
33
 
34
 
35
  @st.cache_resource
36
- def load_documents(data_path):
37
  try:
38
  document_loader = PyPDFDirectoryLoader(data_path)
39
  return document_loader.load()
@@ -42,6 +42,15 @@ def load_documents(data_path):
42
  return None # or handle the error in an appropriate manner
43
 
44
 
 
 
 
 
 
 
 
 
 
45
  @st.cache_resource
46
  def split_docs(documents, chunk_size, chunk_overlap):
47
  try:
@@ -55,6 +64,14 @@ def split_docs(documents, chunk_size, chunk_overlap):
55
  print(f"Error splitting documents: {e}")
56
  return [] # or handle the error in an appropriate manner
57
 
 
 
 
 
 
 
 
 
58
  @st.cache_resource
59
  def chroma_db(docs, embeddings):
60
  try:
 
33
 
34
 
35
  @st.cache_resource
36
+ def load_pdf_documents(data_path):
37
  try:
38
  document_loader = PyPDFDirectoryLoader(data_path)
39
  return document_loader.load()
 
42
  return None # or handle the error in an appropriate manner
43
 
44
 
45
+ @st.cache_data
46
+ def load_txt_documents(data_path):
47
+ documents = []
48
+ for filename in os.listdir(data_path):
49
+ if filename.endswith('.txt'):
50
+ file_path = os.path.join(data_path, filename)
51
+ documents.extend(TextLoader(file_path).load())
52
+ return documents
53
+
54
  @st.cache_resource
55
  def split_docs(documents, chunk_size, chunk_overlap):
56
  try:
 
64
  print(f"Error splitting documents: {e}")
65
  return [] # or handle the error in an appropriate manner
66
 
67
+ @st.cache_data
68
+ def load_uploaded_documents(uploaded_files):
69
+ documents = []
70
+ for uploaded_file in uploaded_files:
71
+ content = uploaded_file.read().decode("utf-8")
72
+ documents.append({"content": content, "filename": uploaded_file.name})
73
+ return documents
74
+
75
  @st.cache_resource
76
  def chroma_db(docs, embeddings):
77
  try: