feat: updating default parameters to get better performance with docling-parse (#2208)

* updated the code

Signed-off-by: Peter Staar <[email protected]>

* updated the parameters

Signed-off-by: Peter Staar <[email protected]>

---------

Signed-off-by: Peter Staar <[email protected]>

Author: PeterStaar-IBM

docling/backend/docling_parse_v4_backend.py

@@ -30,13 +30,21 @@ def __init__(
         page_no: int,
         create_words: bool = True,
         create_textlines: bool = True,
+        keep_chars: bool = False,
+        keep_lines: bool = False,
+        keep_images: bool = True,
     ):
         self._ppage = page_obj
         self._dp_doc = dp_doc
         self._page_no = page_no
+
         self._create_words = create_words
         self._create_textlines = create_textlines
 
+        self._keep_chars = keep_chars
+        self._keep_lines = keep_lines
+        self._keep_images = keep_images
+
         self._dpage: Optional[SegmentedPdfPage] = None
         self._unloaded = False
         self.valid = (self._ppage is not None) and (self._dp_doc is not None)
@@ -47,9 +55,9 @@ def _ensure_parsed(self) -> None:
 
         seg_page = self._dp_doc.get_page(
             self._page_no + 1,
-            keep_chars=True,
-            keep_lines=True,
-            keep_bitmaps=True,
+            keep_chars=self._keep_chars,
+            keep_lines=self._keep_lines,
+            keep_bitmaps=self._keep_images,
             create_words=self._create_words,
             create_textlines=self._create_textlines,
             enforce_same_font=True,