Reorganize imports from docling-core

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

docling/backend/abstract_backend.py

@@ -3,10 +3,11 @@ from io import BytesIO
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, Iterable, Optional, Union
 
+from docling_core.types.experimental.base import BoundingBox, Size
 from PIL import Image
 
 if TYPE_CHECKING:
-    from docling.datamodel.base_models import BoundingBox, Cell, PageSize
+    from docling.datamodel.base_models import Cell
 
 
 class PdfPageBackend(ABC):
@@ -30,7 +31,7 @@ class PdfPageBackend(ABC):
         pass
 
     @abstractmethod
-    def get_size(self) -> "PageSize":
+    def get_size(self) -> "Size":
         pass
 
     @abstractmethod

docling/backend/docling_parse_backend.py

@@ -5,12 +5,13 @@ from pathlib import Path
 from typing import Iterable, List, Optional, Union
 
 import pypdfium2 as pdfium
+from docling_core.types.experimental.base import BoundingBox, CoordOrigin, Size
 from docling_parse.docling_parse import pdf_parser
 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
+from docling.datamodel.base_models import Cell
 
 _log = logging.getLogger(__name__)
 
@@ -177,8 +178,8 @@ class DoclingParsePageBackend(PdfPageBackend):
 
         return image
 
-    def get_size(self) -> PageSize:
-        return PageSize(width=self._ppage.get_width(), height=self._ppage.get_height())
+    def get_size(self) -> Size:
+        return Size(width=self._ppage.get_width(), height=self._ppage.get_height())
 
     def unload(self):
         self._ppage = None

docling/backend/pypdfium2_backend.py

@@ -6,12 +6,13 @@ from typing import Iterable, List, Optional, Union
 
 import pypdfium2 as pdfium
 import pypdfium2.raw as pdfium_c
+from docling_core.types.experimental.base import BoundingBox, CoordOrigin, Size
 from PIL import Image, ImageDraw
 from pypdfium2 import PdfPage, PdfTextPage
 from pypdfium2._helpers.misc import PdfiumError
 
 from docling.backend.abstract_backend import PdfDocumentBackend, PdfPageBackend
-from docling.datamodel.base_models import BoundingBox, Cell, CoordOrigin, PageSize
+from docling.datamodel.base_models import Cell
 
 _log = logging.getLogger(__name__)
 
@@ -222,8 +223,8 @@ class PyPdfiumPageBackend(PdfPageBackend):
 
         return image
 
-    def get_size(self) -> PageSize:
-        return PageSize(width=self._ppage.get_width(), height=self._ppage.get_height())
+    def get_size(self) -> Size:
+        return Size(width=self._ppage.get_width(), height=self._ppage.get_height())
 
     def unload(self):
         self._ppage = None

docling/datamodel/base_models.py

@@ -4,6 +4,7 @@ from enum import Enum, auto
 from io import BytesIO
 from typing import Annotated, Any, Dict, List, Optional, Tuple, Union
 
+from docling_core.types.experimental.base import BoundingBox, Size
 from PIL.Image import Image
 from pydantic import BaseModel, ConfigDict, Field, model_validator
 from typing_extensions import Self
@@ -24,11 +25,6 @@ class DocInputType(str, Enum):
     STREAM = auto()
 
 
-class CoordOrigin(str, Enum):
-    TOPLEFT = auto()
-    BOTTOMLEFT = auto()
-
-
 class DoclingComponentType(str, Enum):
     PDF_BACKEND = auto()
     MODEL = auto()
@@ -41,115 +37,6 @@ class ErrorItem(BaseModel):
     error_message: str
 
 
-class PageSize(BaseModel):
-    width: float = 0.0
-    height: float = 0.0
-
-
-class BoundingBox(BaseModel):
-    l: float  # left
-    t: float  # top
-    r: float  # right
-    b: float  # bottom
-
-    coord_origin: CoordOrigin = CoordOrigin.TOPLEFT
-
-    @property
-    def width(self):
-        return self.r - self.l
-
-    @property
-    def height(self):
-        return abs(self.t - self.b)
-
-    def scaled(self, scale: float) -> "BoundingBox":
-        out_bbox = copy.deepcopy(self)
-        out_bbox.l *= scale
-        out_bbox.r *= scale
-        out_bbox.t *= scale
-        out_bbox.b *= scale
-
-        return out_bbox
-
-    def normalized(self, page_size: PageSize) -> "BoundingBox":
-        out_bbox = copy.deepcopy(self)
-        out_bbox.l /= page_size.width
-        out_bbox.r /= page_size.width
-        out_bbox.t /= page_size.height
-        out_bbox.b /= page_size.height
-
-        return out_bbox
-
-    def as_tuple(self):
-        if self.coord_origin == CoordOrigin.TOPLEFT:
-            return (self.l, self.t, self.r, self.b)
-        elif self.coord_origin == CoordOrigin.BOTTOMLEFT:
-            return (self.l, self.b, self.r, self.t)
-
-    @classmethod
-    def from_tuple(cls, coord: Tuple[float, ...], origin: CoordOrigin):
-        if origin == CoordOrigin.TOPLEFT:
-            l, t, r, b = coord[0], coord[1], coord[2], coord[3]
-            if r < l:
-                l, r = r, l
-            if b < t:
-                b, t = t, b
-
-            return BoundingBox(l=l, t=t, r=r, b=b, coord_origin=origin)
-        elif origin == CoordOrigin.BOTTOMLEFT:
-            l, b, r, t = coord[0], coord[1], coord[2], coord[3]
-            if r < l:
-                l, r = r, l
-            if b > t:
-                b, t = t, b
-
-            return BoundingBox(l=l, t=t, r=r, b=b, coord_origin=origin)
-
-    def area(self) -> float:
-        return (self.r - self.l) * (self.b - self.t)
-
-    def intersection_area_with(self, other: "BoundingBox") -> float:
-        # Calculate intersection coordinates
-        left = max(self.l, other.l)
-        top = max(self.t, other.t)
-        right = min(self.r, other.r)
-        bottom = min(self.b, other.b)
-
-        # Calculate intersection dimensions
-        width = right - left
-        height = bottom - top
-
-        # If the bounding boxes do not overlap, width or height will be negative
-        if width <= 0 or height <= 0:
-            return 0.0
-
-        return width * height
-
-    def to_bottom_left_origin(self, page_height) -> "BoundingBox":
-        if self.coord_origin == CoordOrigin.BOTTOMLEFT:
-            return self
-        elif self.coord_origin == CoordOrigin.TOPLEFT:
-            return BoundingBox(
-                l=self.l,
-                r=self.r,
-                t=page_height - self.t,
-                b=page_height - self.b,
-                coord_origin=CoordOrigin.BOTTOMLEFT,
-            )
-
-    def to_top_left_origin(self, page_height):
-        if self.coord_origin == CoordOrigin.TOPLEFT:
-            return self
-        elif self.coord_origin == CoordOrigin.BOTTOMLEFT:
-            return BoundingBox(
-                l=self.l,
-                r=self.r,
-                t=page_height - self.t,  # self.b
-                b=page_height - self.b,  # self.t
-                coord_origin=CoordOrigin.TOPLEFT,
-            )
-
-
 class Cell(BaseModel):
     id: int
     text: str
@@ -266,7 +153,7 @@ class Page(BaseModel):
 
     page_no: int
     page_hash: Optional[str] = None
-    size: Optional[PageSize] = None
+    size: Optional[Size] = None
     cells: List[Cell] = []
     predictions: PagePredictions = PagePredictions()
     assembled: Optional[AssembledUnit] = None

docling/datamodel/document.py

@@ -4,13 +4,13 @@ from pathlib import Path, PurePath
 from typing import ClassVar, Dict, Iterable, List, Optional, Tuple, Type, Union
 
 from docling_core.types import BaseCell, BaseText
-from docling_core.types import BoundingBox as DsBoundingBox
 from docling_core.types import Document as DsDocument
 from docling_core.types import DocumentDescription as DsDocumentDescription
 from docling_core.types import FileInfoObject as DsFileInfoObject
 from docling_core.types import PageDimensions, PageReference, Prov, Ref
 from docling_core.types import Table as DsSchemaTable
 from docling_core.types import TableCell
+from docling_core.types.doc.base import BoundingBox as DsBoundingBox
 from docling_core.types.doc.base import Figure
 from pydantic import BaseModel
 from typing_extensions import deprecated

docling/models/base_ocr_model.py

@@ -5,11 +5,12 @@ from typing import Iterable, List, Tuple
 
 import numpy
 import numpy as np
+from docling_core.types.experimental.base import BoundingBox, CoordOrigin
 from PIL import Image, ImageDraw
 from rtree import index
 from scipy.ndimage import find_objects, label
 
-from docling.datamodel.base_models import BoundingBox, CoordOrigin, OcrCell, Page
+from docling.datamodel.base_models import OcrCell, Page
 
 _log = logging.getLogger(__name__)
 

docling/models/ds_glm_model.py

@@ -7,9 +7,10 @@ from deepsearch_glm.utils.load_pretrained_models import load_pretrained_nlp_mode
 from docling_core.types import BaseText
 from docling_core.types import Document as DsDocument
 from docling_core.types import Ref
+from docling_core.types.experimental.base import BoundingBox, CoordOrigin
 from PIL import ImageDraw
 
-from docling.datamodel.base_models import BoundingBox, Cluster, CoordOrigin
+from docling.datamodel.base_models import Cluster
 from docling.datamodel.document import ConversionResult
 
 

docling/models/easyocr_model.py

@@ -2,8 +2,9 @@ import logging
 from typing import Iterable
 
 import numpy
+from docling_core.types.experimental.base import BoundingBox, CoordOrigin
 
-from docling.datamodel.base_models import BoundingBox, CoordOrigin, OcrCell, Page
+from docling.datamodel.base_models import OcrCell, Page
 from docling.models.base_ocr_model import BaseOcrModel
 
 _log = logging.getLogger(__name__)

docling/models/layout_model.py

@@ -4,6 +4,7 @@ import random
 import time
 from typing import Iterable, List
 
+from docling_core.types.experimental.base import CoordOrigin
 from docling_ibm_models.layoutmodel.layout_predictor import LayoutPredictor
 from PIL import ImageDraw
 
@@ -11,7 +12,6 @@ from docling.datamodel.base_models import (
     BoundingBox,
     Cell,
     Cluster,
-    CoordOrigin,
     LayoutPrediction,
     Page,
 )

docling/models/table_structure_model.py

@@ -2,11 +2,11 @@ import copy
 from typing import Iterable, List
 
 import numpy
+from docling_core.types.experimental.base import BoundingBox
 from docling_ibm_models.tableformer.data_management.tf_predictor import TFPredictor
 from PIL import ImageDraw
 
 from docling.datamodel.base_models import (
-    BoundingBox,
     Page,
     TableCell,
     TableElement,

docling/utils/export.py

@@ -1,9 +1,17 @@
 import logging
 from typing import Any, Dict, Iterable, List, Tuple, Union
 
-from docling_core.types.doc.base import BaseCell, BaseText, Ref, Table, TableCell
+from docling_core.types.doc.base import (
+    BaseCell,
+    BaseText,
+    BoundingBox,
+    Ref,
+    Table,
+    TableCell,
+)
+from docling_core.types.experimental.base import CoordOrigin
 
-from docling.datamodel.base_models import BoundingBox, CoordOrigin, OcrCell
+from docling.datamodel.base_models import OcrCell
 from docling.datamodel.document import ConversionResult, Page
 
 _log = logging.getLogger(__name__)