Update app.py
Browse files
app.py
CHANGED
|
@@ -5,6 +5,7 @@ import torch
|
|
| 5 |
import pickle
|
| 6 |
import re
|
| 7 |
import PyPDF2
|
|
|
|
| 8 |
|
| 9 |
def checker(content):
|
| 10 |
#if isinstance(content, bytes):#falls es ein Byte-String ist
|
|
@@ -46,11 +47,16 @@ def read_file(file):
|
|
| 46 |
while counter < len(reader.pages):
|
| 47 |
content += reader.pages[counter].extract_text()
|
| 48 |
counter += 1
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
else:
|
| 50 |
with open(file.name, 'r', encoding='utf-8') as f:
|
| 51 |
content = f.read()
|
| 52 |
#with open(file.name, 'rb') as f:
|
| 53 |
# content = pickle.load(f)
|
|
|
|
| 54 |
return content
|
| 55 |
|
| 56 |
def is_valid_file(file_path: str, file_types: list[str]) -> bool:
|
|
|
|
| 5 |
import pickle
|
| 6 |
import re
|
| 7 |
import PyPDF2
|
| 8 |
+
import docx
|
| 9 |
|
| 10 |
def checker(content):
|
| 11 |
#if isinstance(content, bytes):#falls es ein Byte-String ist
|
|
|
|
| 47 |
while counter < len(reader.pages):
|
| 48 |
content += reader.pages[counter].extract_text()
|
| 49 |
counter += 1
|
| 50 |
+
if ".docx" in file.name:
|
| 51 |
+
doc = docx.Document(filename)
|
| 52 |
+
for para in doc.paragraphs:
|
| 53 |
+
content += para.text
|
| 54 |
else:
|
| 55 |
with open(file.name, 'r', encoding='utf-8') as f:
|
| 56 |
content = f.read()
|
| 57 |
#with open(file.name, 'rb') as f:
|
| 58 |
# content = pickle.load(f)
|
| 59 |
+
print(content)
|
| 60 |
return content
|
| 61 |
|
| 62 |
def is_valid_file(file_path: str, file_types: list[str]) -> bool:
|