adjusted pytesseract workflow

app.py

@@ -920,7 +920,7 @@ async def ocr_and_parse_hocr(file_content: bytes) -> list[dict]:
 # --- END: New hOCR Functions ---
 
 
-async def translate_paddle_data_concurrently(
+async def translate_tesseract_data_concurrently(
     paddle_data: list[dict], target_language: str
 ) -> list[dict]:
     """
@@ -982,7 +982,7 @@ def inject_dropdown_script(html_content):
     return f"{html_content}{script}"
 
 
-async def generate_html_from_paddle_data(translated_data: list[dict]) -> str:
+async def generate_html_from_tesseract_data(translated_data: list[dict]) -> str:
     """
     Receives translated OCR data (text with coordinates) and uses Gemini
     to generate a layout-aware HTML document.
@@ -1062,13 +1062,13 @@ async def translate_document_mvp(
         print(f"***** Step 1 Done: Extracted {len(ocr_data)} words ******")
 
         # === MVP STEP 2: Translate each text block concurrently ===
-        translated_data = await translate_paddle_data_concurrently(
+        translated_data = await translate_tesseract_data_concurrently(
             ocr_data, target_language
         )
         print("***** Step 2 Done: Translated data ******")
 
         # === MVP STEP 3: Generate final, layout-aware HTML from Gemini ===
-        final_html = await generate_html_from_paddle_data(translated_data)
+        final_html = await generate_html_from_tesseract_data(translated_data)
         print("***** Step 3 Done: Generated HTML ******")
         return HTMLResponse(content=final_html)