Update app.py
Browse files
app.py
CHANGED
|
@@ -64,8 +64,15 @@ mcq_chain = LLMChain(llm=llm, prompt=prompt)
|
|
| 64 |
# Extract text from PDF or Word
|
| 65 |
def extract_text(file):
|
| 66 |
if file.name.endswith(".pdf"):
|
| 67 |
-
|
| 68 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
elif file.name.endswith(".docx"):
|
| 70 |
doc = docx.Document(file)
|
| 71 |
return "\n".join([para.text for para in doc.paragraphs])
|
|
|
|
| 64 |
# Extract text from PDF or Word
|
| 65 |
def extract_text(file):
|
| 66 |
if file.name.endswith(".pdf"):
|
| 67 |
+
# Read the entire file content into memory
|
| 68 |
+
file_bytes = file.read()
|
| 69 |
+
# Open the PDF from the byte stream
|
| 70 |
+
doc = fitz.open(stream=file_bytes, filetype="pdf")
|
| 71 |
+
# Extract text from all pages
|
| 72 |
+
text = ""
|
| 73 |
+
for page in doc:
|
| 74 |
+
text += page.get_text()
|
| 75 |
+
return text
|
| 76 |
elif file.name.endswith(".docx"):
|
| 77 |
doc = docx.Document(file)
|
| 78 |
return "\n".join([para.text for para in doc.paragraphs])
|