Welcome to PDF Text Extraction API
This API allows you to upload PDFs and extract text — including optional OCR for images.
Available endpoints:
- POST /extract-text - Extract plain text from PDF pages.
- POST /extract-text-ocr - Extract text including OCR from image-based PDFs.
- POST /extract-text-structured - Extract structured text using pdfminer.
Use a tool like Postman or write your own client to send PDF files to the endpoints.
"""
return HTMLResponse(content=html_content, status_code=200)
@app.post("/extract-text")
async def extract_text(file: UploadFile = File(...)):
try:
contents = await file.read()
doc = fitz.open(stream=contents, filetype="pdf")
extracted_text = ""
for i, page in enumerate(doc):
extracted_text += f"\n\n--- Page {i + 1} ---\n\n" + page.get_text()
return {"filename": file.filename, "text": extracted_text}
except Exception as e:
return {"error": str(e)}
@app.post("/extract-text-ocr")
async def extract_text_ocr(file: UploadFile = File(...)):
try:
contents = await file.read()
doc = fitz.open(stream=contents, filetype="pdf")
full_text = ""
for i in range(len(doc)):
page = doc.load_page(i)
# Normal text
text = page.get_text()
# Render page to an image
pix = page.get_pixmap()
img = Image.open(io.BytesIO(pix.tobytes()))
# OCR text
ocr_text = pytesseract.image_to_string(img)
full_text += f"\n\n--- Page {i + 1} ---\n\n"
full_text += text + "\n"
full_text += "[OCR Text]\n" + ocr_text
return {"filename": file.filename, "text": full_text}
except Exception as e:
return {"error": str(e)}
@app.post("/extract-text-structured")
async def extract_text_structured(file: UploadFile = File(...)):
try:
contents = await file.read()
# Save to temp file to use with extract_pages
import tempfile
with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp_file:
tmp_file.write(contents)
temp_pdf_path = tmp_file.name
structured_text = ""
for i, page_layout in enumerate(extract_pages(temp_pdf_path)):
structured_text += f"\n\n--- Page {i + 1} ---\n\n"
for element in page_layout:
if isinstance(element, LTTextContainer):
structured_text += element.get_text()
return {"filename": file.filename, "text": structured_text}
except Exception as e:
return {"error": str(e)}