Update app.py
Browse files
app.py
CHANGED
|
@@ -29,31 +29,28 @@ def get_pdf_text(pdf_docs):
|
|
| 29 |
# μλ ν
μ€νΈ μΆμΆ ν¨μλ₯Ό μμ±
|
| 30 |
|
| 31 |
def get_text_file(docs):
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
|
|
|
| 37 |
|
| 38 |
def get_csv_file(docs):
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
else:
|
| 46 |
-
st.warning("Unsupported file type for get_csv_file")
|
| 47 |
|
| 48 |
def get_json_file(docs):
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
else:
|
| 56 |
-
st.warning("Unsupported file type for get_json_file")
|
| 57 |
|
| 58 |
|
| 59 |
# λ¬Έμλ€μ μ²λ¦¬νμ¬ ν
μ€νΈ μ²ν¬λ‘ λλλ ν¨μμ
λλ€.
|
|
@@ -155,32 +152,33 @@ def main():
|
|
| 155 |
docs = st.file_uploader(
|
| 156 |
"Upload your documents here and click on 'Process'", accept_multiple_files=True)
|
| 157 |
if st.button("Process"):
|
| 158 |
-
|
| 159 |
-
|
| 160 |
-
|
| 161 |
-
|
| 162 |
-
|
| 163 |
-
|
| 164 |
-
|
| 165 |
-
|
| 166 |
-
|
| 167 |
-
|
| 168 |
-
|
| 169 |
-
|
| 170 |
-
|
| 171 |
-
|
| 172 |
-
|
| 173 |
-
|
| 174 |
-
|
| 175 |
-
|
| 176 |
-
|
| 177 |
-
|
| 178 |
-
|
| 179 |
-
|
| 180 |
-
|
| 181 |
-
|
| 182 |
-
|
| 183 |
-
|
|
|
|
| 184 |
|
| 185 |
|
| 186 |
if __name__ == '__main__':
|
|
|
|
| 29 |
# μλ ν
μ€νΈ μΆμΆ ν¨μλ₯Ό μμ±
|
| 30 |
|
| 31 |
def get_text_file(docs):
|
| 32 |
+
text_list = []
|
| 33 |
+
for file in docs:
|
| 34 |
+
if file.type == 'text/plain':
|
| 35 |
+
# file is .txt
|
| 36 |
+
text_list.append(file.getvalue().decode('utf-8'))
|
| 37 |
+
return text_list
|
| 38 |
|
| 39 |
def get_csv_file(docs):
|
| 40 |
+
csv_list = []
|
| 41 |
+
for file in docs:
|
| 42 |
+
if file.type == 'text/csv':
|
| 43 |
+
# file is .csv
|
| 44 |
+
csv_list.extend(csv.reader(file.getvalue().decode('utf-8').splitlines()))
|
| 45 |
+
return csv_list
|
|
|
|
|
|
|
| 46 |
|
| 47 |
def get_json_file(docs):
|
| 48 |
+
json_list = []
|
| 49 |
+
for file in docs:
|
| 50 |
+
if file.type == 'application/json':
|
| 51 |
+
# file is .json
|
| 52 |
+
json_list.extend(json.load(file))
|
| 53 |
+
return json_list
|
|
|
|
|
|
|
| 54 |
|
| 55 |
|
| 56 |
# λ¬Έμλ€μ μ²λ¦¬νμ¬ ν
μ€νΈ μ²ν¬λ‘ λλλ ν¨μμ
λλ€.
|
|
|
|
| 152 |
docs = st.file_uploader(
|
| 153 |
"Upload your documents here and click on 'Process'", accept_multiple_files=True)
|
| 154 |
if st.button("Process"):
|
| 155 |
+
with st.spinner("Processing"):
|
| 156 |
+
# get pdf text
|
| 157 |
+
doc_list = []
|
| 158 |
+
|
| 159 |
+
for file in docs:
|
| 160 |
+
print('file - type : ', file.type)
|
| 161 |
+
if file.type == 'text/plain':
|
| 162 |
+
# file is .txt
|
| 163 |
+
doc_list.extend(get_text_file([file]))
|
| 164 |
+
elif file.type in ['application/octet-stream', 'application/pdf']:
|
| 165 |
+
# file is .pdf
|
| 166 |
+
doc_list.extend(get_pdf_text(file))
|
| 167 |
+
elif file.type == 'text/csv':
|
| 168 |
+
# file is .csv
|
| 169 |
+
doc_list.extend(get_csv_file([file]))
|
| 170 |
+
elif file.type == 'application/json':
|
| 171 |
+
# file is .json
|
| 172 |
+
doc_list.extend(get_json_file([file]))
|
| 173 |
+
|
| 174 |
+
# get the text chunks
|
| 175 |
+
text_chunks = get_text_chunks(doc_list)
|
| 176 |
+
|
| 177 |
+
# create vector store
|
| 178 |
+
vectorstore = get_vectorstore(text_chunks)
|
| 179 |
+
|
| 180 |
+
# create conversation chain
|
| 181 |
+
st.session_state.conversation = get_conversation_chain(vectorstore)
|
| 182 |
|
| 183 |
|
| 184 |
if __name__ == '__main__':
|