Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
|
@@ -2,9 +2,11 @@ import streamlit as st
|
|
| 2 |
from PyPDF2 import PdfReader
|
| 3 |
from docx import Document
|
| 4 |
from io import BytesIO
|
|
|
|
|
|
|
| 5 |
|
| 6 |
def pdf_to_word(pdf_file, password=None):
|
| 7 |
-
"""Convert a PDF file to a Word file with optional decryption."""
|
| 8 |
reader = PdfReader(pdf_file)
|
| 9 |
|
| 10 |
# Decrypt the PDF if it's encrypted
|
|
@@ -18,12 +20,22 @@ def pdf_to_word(pdf_file, password=None):
|
|
| 18 |
raise ValueError("The PDF is encrypted. Please provide a password.")
|
| 19 |
|
| 20 |
document = Document()
|
|
|
|
|
|
|
|
|
|
| 21 |
for page in reader.pages:
|
| 22 |
-
if page.extract_text(): #
|
| 23 |
text = page.extract_text()
|
| 24 |
document.add_paragraph(text)
|
| 25 |
else:
|
| 26 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
word_file = BytesIO()
|
| 29 |
document.save(word_file)
|
|
@@ -35,24 +47,4 @@ st.set_page_config(page_title="PDF to Word Converter", page_icon="🖋", layout=
|
|
| 35 |
|
| 36 |
# App header
|
| 37 |
st.title("PDF to Word Converter")
|
| 38 |
-
st.write("Upload a PDF file,
|
| 39 |
-
|
| 40 |
-
# File uploader
|
| 41 |
-
uploaded_file = st.file_uploader("Choose a PDF file", type="pdf")
|
| 42 |
-
password = st.text_input("Enter password (if the PDF is encrypted):", type="password")
|
| 43 |
-
|
| 44 |
-
if uploaded_file is not None:
|
| 45 |
-
with st.spinner("Converting PDF to Word..."):
|
| 46 |
-
try:
|
| 47 |
-
word_file = pdf_to_word(uploaded_file, password)
|
| 48 |
-
st.success("Conversion successful!")
|
| 49 |
-
st.download_button(
|
| 50 |
-
label="Download Word file",
|
| 51 |
-
data=word_file,
|
| 52 |
-
file_name="converted.docx",
|
| 53 |
-
mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document"
|
| 54 |
-
)
|
| 55 |
-
except ValueError as ve:
|
| 56 |
-
st.error(str(ve))
|
| 57 |
-
except Exception as e:
|
| 58 |
-
st.error(f"An error occurred: {str(e)}")
|
|
|
|
| 2 |
from PyPDF2 import PdfReader
|
| 3 |
from docx import Document
|
| 4 |
from io import BytesIO
|
| 5 |
+
from pdf2image import convert_from_bytes
|
| 6 |
+
import pytesseract
|
| 7 |
|
| 8 |
def pdf_to_word(pdf_file, password=None):
|
| 9 |
+
"""Convert a PDF file to a Word file with optional decryption and OCR."""
|
| 10 |
reader = PdfReader(pdf_file)
|
| 11 |
|
| 12 |
# Decrypt the PDF if it's encrypted
|
|
|
|
| 20 |
raise ValueError("The PDF is encrypted. Please provide a password.")
|
| 21 |
|
| 22 |
document = Document()
|
| 23 |
+
|
| 24 |
+
# Extract text from each page
|
| 25 |
+
pdf_bytes = pdf_file.read()
|
| 26 |
for page in reader.pages:
|
| 27 |
+
if page.extract_text(): # Use PyPDF2 for text extraction
|
| 28 |
text = page.extract_text()
|
| 29 |
document.add_paragraph(text)
|
| 30 |
else:
|
| 31 |
+
# Use OCR for non-extractable pages
|
| 32 |
+
images = convert_from_bytes(pdf_bytes)
|
| 33 |
+
for image in images:
|
| 34 |
+
ocr_text = pytesseract.image_to_string(image)
|
| 35 |
+
if ocr_text.strip():
|
| 36 |
+
document.add_paragraph(ocr_text)
|
| 37 |
+
else:
|
| 38 |
+
document.add_paragraph("[This page contains non-extractable content or images]")
|
| 39 |
|
| 40 |
word_file = BytesIO()
|
| 41 |
document.save(word_file)
|
|
|
|
| 47 |
|
| 48 |
# App header
|
| 49 |
st.title("PDF to Word Converter")
|
| 50 |
+
st.write("Upload a PDF file,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|