docling 1.19.1__py3-none-any.whl → 1.20.0__py3-none-any.whl

docling/backend/docling_parse_v2_backend.py

@@ -0,0 +1,236 @@
+import logging
+import random
+from io import BytesIO
+from pathlib import Path
+from typing import Iterable, List, Optional, Union
+
+import pypdfium2 as pdfium
+from docling_parse.docling_parse import pdf_parser_v2
+from PIL import Image, ImageDraw
+from pypdfium2 import PdfPage
+
+from docling.backend.abstract_backend import PdfDocumentBackend, PdfPageBackend
+from docling.datamodel.base_models import BoundingBox, Cell, CoordOrigin, PageSize
+
+_log = logging.getLogger(__name__)
+
+
+class DoclingParseV2PageBackend(PdfPageBackend):
+    def __init__(
+        self, parser: pdf_parser_v2, document_hash: str, page_no: int, page_obj: PdfPage
+    ):
+        self._ppage = page_obj
+        parsed_page = parser.parse_pdf_from_key_on_page(document_hash, page_no)
+
+        self.valid = "pages" in parsed_page
+        if self.valid:
+            self._dpage = parsed_page["pages"][page_no]
+        else:
+            _log.info(
+                f"An error occured when loading page {page_no} of document {document_hash}."
+            )
+
+    def is_valid(self) -> bool:
+        return self.valid
+
+    def get_text_in_rect(self, bbox: BoundingBox) -> str:
+        if not self.valid:
+            return ""
+        # Find intersecting cells on the page
+        text_piece = ""
+        page_size = self.get_size()
+
+        parser_width = self._dpage["sanitized"]["dimension"]["width"]
+        parser_height = self._dpage["sanitized"]["dimension"]["height"]
+
+        scale = (
+            1  # FIX - Replace with param in get_text_in_rect across backends (optional)
+        )
+
+        cells_data = self._dpage["sanitized"]["cells"]["data"]
+        cells_header = self._dpage["sanitized"]["cells"]["header"]
+
+        for i, cell_data in enumerate(cells_data):
+            x0 = cell_data[cells_header.index("x0")]
+            y0 = cell_data[cells_header.index("y0")]
+            x1 = cell_data[cells_header.index("x1")]
+            y1 = cell_data[cells_header.index("y1")]
+
+            cell_bbox = BoundingBox(
+                l=x0 * scale * page_size.width / parser_width,
+                b=y0 * scale * page_size.height / parser_height,
+                r=x1 * scale * page_size.width / parser_width,
+                t=y1 * scale * page_size.height / parser_height,
+                coord_origin=CoordOrigin.BOTTOMLEFT,
+            ).to_top_left_origin(page_height=page_size.height * scale)
+
+            overlap_frac = cell_bbox.intersection_area_with(bbox) / cell_bbox.area()
+
+            if overlap_frac > 0.5:
+                if len(text_piece) > 0:
+                    text_piece += " "
+                text_piece += cell_data[cells_header.index("text")]
+
+        return text_piece
+
+    def get_text_cells(self) -> Iterable[Cell]:
+        cells: List[Cell] = []
+        cell_counter = 0
+
+        if not self.valid:
+            return cells
+
+        page_size = self.get_size()
+
+        parser_width = self._dpage["sanitized"]["dimension"]["width"]
+        parser_height = self._dpage["sanitized"]["dimension"]["height"]
+
+        cells_data = self._dpage["sanitized"]["cells"]["data"]
+        cells_header = self._dpage["sanitized"]["cells"]["header"]
+
+        for i, cell_data in enumerate(cells_data):
+            x0 = cell_data[cells_header.index("x0")]
+            y0 = cell_data[cells_header.index("y0")]
+            x1 = cell_data[cells_header.index("x1")]
+            y1 = cell_data[cells_header.index("y1")]
+
+            if x1 < x0:
+                x0, x1 = x1, x0
+            if y1 < y0:
+                y0, y1 = y1, y0
+
+            text_piece = cell_data[cells_header.index("text")]
+            cells.append(
+                Cell(
+                    id=cell_counter,
+                    text=text_piece,
+                    bbox=BoundingBox(
+                        # l=x0, b=y0, r=x1, t=y1,
+                        l=x0 * page_size.width / parser_width,
+                        b=y0 * page_size.height / parser_height,
+                        r=x1 * page_size.width / parser_width,
+                        t=y1 * page_size.height / parser_height,
+                        coord_origin=CoordOrigin.BOTTOMLEFT,
+                    ).to_top_left_origin(page_size.height),
+                )
+            )
+            cell_counter += 1
+
+        def draw_clusters_and_cells():
+            image = (
+                self.get_page_image()
+            )  # make new image to avoid drawing on the saved ones
+            draw = ImageDraw.Draw(image)
+            for c in cells:
+                x0, y0, x1, y1 = c.bbox.as_tuple()
+                cell_color = (
+                    random.randint(30, 140),
+                    random.randint(30, 140),
+                    random.randint(30, 140),
+                )
+                draw.rectangle([(x0, y0), (x1, y1)], outline=cell_color)
+            image.show()
+
+        # draw_clusters_and_cells()
+
+        return cells
+
+    def get_bitmap_rects(self, scale: float = 1) -> Iterable[BoundingBox]:
+        AREA_THRESHOLD = 32 * 32
+
+        images = self._dpage["sanitized"]["images"]["data"]
+        images_header = self._dpage["sanitized"]["images"]["header"]
+
+        for row in images:
+            x0 = row[images_header.index("x0")]
+            y0 = row[images_header.index("y0")]
+            x1 = row[images_header.index("x1")]
+            y1 = row[images_header.index("y1")]
+
+            cropbox = BoundingBox.from_tuple(
+                (x0, y0, x1, y1), origin=CoordOrigin.BOTTOMLEFT
+            ).to_top_left_origin(self.get_size().height)
+
+            if cropbox.area() > AREA_THRESHOLD:
+                cropbox = cropbox.scaled(scale=scale)
+
+                yield cropbox
+
+    def get_page_image(
+        self, scale: float = 1, cropbox: Optional[BoundingBox] = None
+    ) -> Image.Image:
+
+        page_size = self.get_size()
+
+        if not cropbox:
+            cropbox = BoundingBox(
+                l=0,
+                r=page_size.width,
+                t=0,
+                b=page_size.height,
+                coord_origin=CoordOrigin.TOPLEFT,
+            )
+            padbox = BoundingBox(
+                l=0, r=0, t=0, b=0, coord_origin=CoordOrigin.BOTTOMLEFT
+            )
+        else:
+            padbox = cropbox.to_bottom_left_origin(page_size.height)
+            padbox.r = page_size.width - padbox.r
+            padbox.t = page_size.height - padbox.t
+
+        image = (
+            self._ppage.render(
+                scale=scale * 1.5,
+                rotation=0,  # no additional rotation
+                crop=padbox.as_tuple(),
+            )
+            .to_pil()
+            .resize(size=(round(cropbox.width * scale), round(cropbox.height * scale)))
+        )  # We resize the image from 1.5x the given scale to make it sharper.
+
+        return image
+
+    def get_size(self) -> PageSize:
+        return PageSize(width=self._ppage.get_width(), height=self._ppage.get_height())
+
+    def unload(self):
+        self._ppage = None
+        self._dpage = None
+
+
+class DoclingParseV2DocumentBackend(PdfDocumentBackend):
+    def __init__(self, path_or_stream: Union[BytesIO, Path], document_hash: str):
+        super().__init__(path_or_stream, document_hash)
+
+        self._pdoc = pdfium.PdfDocument(path_or_stream)
+        self.parser = pdf_parser_v2("fatal")
+
+        success = False
+        if isinstance(path_or_stream, BytesIO):
+            success = self.parser.load_document_from_bytesio(
+                document_hash, path_or_stream
+            )
+        elif isinstance(path_or_stream, Path):
+            success = self.parser.load_document(document_hash, str(path_or_stream))
+
+        if not success:
+            raise RuntimeError(
+                f"docling-parse could not load document {document_hash}."
+            )
+
+    def page_count(self) -> int:
+        return len(self._pdoc)  # To be replaced with docling-parse API
+
+    def load_page(self, page_no: int) -> DoclingParseV2PageBackend:
+        return DoclingParseV2PageBackend(
+            self.parser, self.document_hash, page_no, self._pdoc[page_no]
+        )
+
+    def is_valid(self) -> bool:
+        return self.page_count() > 0
+
+    def unload(self):
+        super().unload()
+        self.parser.unload_document(self.document_hash)
+        self._pdoc.close()
+        self._pdoc = None

{docling-1.19.1.dist-info → docling-1.20.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 1.19.1
+Version: 1.20.0
 Summary: Docling PDF conversion package
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -24,7 +24,7 @@ Requires-Dist: certifi (>=2024.7.4)
 Requires-Dist: deepsearch-glm (>=0.22.0,<0.23.0)
 Requires-Dist: docling-core (>=1.7.1,<2.0.0)
 Requires-Dist: docling-ibm-models (>=2.0.0,<3.0.0)
-Requires-Dist: docling-parse (>=1.4.1,<2.0.0)
+Requires-Dist: docling-parse (>=1.6.0,<2.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)
 Requires-Dist: filetype (>=1.2.0,<2.0.0)
 Requires-Dist: huggingface_hub (>=0.23,<1)

{docling-1.19.1.dist-info → docling-1.20.0.dist-info}/RECORD

@@ -2,6 +2,7 @@ docling/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/backend/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/backend/abstract_backend.py,sha256=clJtGxLedpLriEhpx7oyxjmlwMLPorkv-1tdfZm9GdA,1546
 docling/backend/docling_parse_backend.py,sha256=RUWWZbx2cUotZeeTkc-Lbg2k8MVFXFxaDjM4sPfaFZE,7475
+docling/backend/docling_parse_v2_backend.py,sha256=Fieflqr-SSiNLMJi8qsPWw_7GX4bZQzyvDwneyW4Kr0,8207
 docling/backend/pypdfium2_backend.py,sha256=bIIImVM73wmcVcKMqjl4JF8CD-Qj2W5rZbI4G7clU4s,8877
 docling/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/cli/main.py,sha256=Q_5HTL2O20bMlF-U8Ac8ev4iqgLPyrwlHEyLAq6rezg,7913
@@ -27,8 +28,8 @@ docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/utils/export.py,sha256=bKLdbeUcR-rQsGPV1IqJkCHKMCv7X2QOHyxmjNuH3HE,4655
 docling/utils/layout_utils.py,sha256=FOFbL0hKzUoWXdZaeUvEtFqKv0IkPifIr4sdGW4suKs,31804
 docling/utils/utils.py,sha256=llhXSbIDNZ1MHOwBEfLHBAoJIAYI7QlPIonlI1jLUJ0,1208
-docling-1.19.1.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-1.19.1.dist-info/METADATA,sha256=hCQeq3JVB16CfTwtjjwnX5u9bWYjD0CsSbn9h1tZZTM,16800
-docling-1.19.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling-1.19.1.dist-info/entry_points.txt,sha256=VOSzV77znM52dz5ysaDuJ0ijl1cnfrh1ZPg8od5OcTs,48
-docling-1.19.1.dist-info/RECORD,,
+docling-1.20.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
+docling-1.20.0.dist-info/METADATA,sha256=OMQd4Jk1nJnmJi7Q7tQ5wA0Y6z5sQg2njNvIXXISM4c,16800
+docling-1.20.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling-1.20.0.dist-info/entry_points.txt,sha256=VOSzV77znM52dz5ysaDuJ0ijl1cnfrh1ZPg8od5OcTs,48
+docling-1.20.0.dist-info/RECORD,,