docling 2.69.0__py3-none-any.whl

docling/backend/docling_parse_v4_backend.py

@@ -0,0 +1,260 @@
+import logging
+from collections.abc import Iterable
+from io import BytesIO
+from pathlib import Path
+from typing import TYPE_CHECKING, Optional, Union
+
+import pypdfium2 as pdfium
+from docling_core.types.doc import BoundingBox, CoordOrigin
+from docling_core.types.doc.page import SegmentedPdfPage, TextCell
+from docling_parse.pdf_parser import DoclingPdfParser, PdfDocument
+from PIL import Image
+from pypdfium2 import PdfPage
+
+from docling.backend.pdf_backend import PdfDocumentBackend, PdfPageBackend
+from docling.datamodel.backend_options import PdfBackendOptions
+from docling.datamodel.base_models import Size
+from docling.utils.locks import pypdfium2_lock
+
+if TYPE_CHECKING:
+    from docling.datamodel.document import InputDocument
+
+_log = logging.getLogger(__name__)
+
+
+class DoclingParseV4PageBackend(PdfPageBackend):
+    def __init__(
+        self,
+        *,
+        dp_doc: PdfDocument,
+        page_obj: PdfPage,
+        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)
+
+    def _ensure_parsed(self) -> None:
+        if self._dpage is not None:
+            return
+
+        seg_page = self._dp_doc.get_page(
+            self._page_no + 1,
+            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,
+        )
+
+        # In Docling, all TextCell instances are expected with top-left origin.
+        [
+            tc.to_top_left_origin(seg_page.dimension.height)
+            for tc in seg_page.textline_cells
+        ]
+        [tc.to_top_left_origin(seg_page.dimension.height) for tc in seg_page.char_cells]
+        [tc.to_top_left_origin(seg_page.dimension.height) for tc in seg_page.word_cells]
+
+        self._dpage = seg_page
+
+    def is_valid(self) -> bool:
+        return self.valid
+
+    def get_text_in_rect(self, bbox: BoundingBox) -> str:
+        self._ensure_parsed()
+        assert self._dpage is not None
+
+        # Find intersecting cells on the page
+        text_piece = ""
+        page_size = self.get_size()
+
+        scale = (
+            1  # FIX - Replace with param in get_text_in_rect across backends (optional)
+        )
+
+        for i, cell in enumerate(self._dpage.textline_cells):
+            cell_bbox = (
+                cell.rect.to_bounding_box()
+                .to_top_left_origin(page_height=page_size.height)
+                .scaled(scale)
+            )
+
+            overlap_frac = cell_bbox.intersection_over_self(bbox)
+
+            if overlap_frac > 0.5:
+                if len(text_piece) > 0:
+                    text_piece += " "
+                text_piece += cell.text
+
+        return text_piece
+
+    def get_segmented_page(self) -> Optional[SegmentedPdfPage]:
+        self._ensure_parsed()
+        return self._dpage
+
+    def get_text_cells(self) -> Iterable[TextCell]:
+        self._ensure_parsed()
+        assert self._dpage is not None
+
+        return self._dpage.textline_cells
+
+    def get_bitmap_rects(self, scale: float = 1) -> Iterable[BoundingBox]:
+        self._ensure_parsed()
+        assert self._dpage is not None
+
+        AREA_THRESHOLD = 0  # 32 * 32
+
+        images = self._dpage.bitmap_resources
+
+        for img in images:
+            cropbox = img.rect.to_bounding_box().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).model_copy()
+            padbox.r = page_size.width - padbox.r
+            padbox.t = page_size.height - padbox.t
+
+        with pypdfium2_lock:
+            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) -> Size:
+        with pypdfium2_lock:
+            return Size(width=self._ppage.get_width(), height=self._ppage.get_height())
+
+        # TODO: Take width and height from docling-parse.
+        # return Size(
+        #    width=self._dpage.dimension.width,
+        #    height=self._dpage.dimension.height,
+        # )
+
+    def unload(self):
+        if not self._unloaded and self._dp_doc is not None:
+            self._dp_doc.unload_pages((self._page_no + 1, self._page_no + 2))
+            self._unloaded = True
+
+        self._ppage = None
+        self._dpage = None
+        self._dp_doc = None
+
+
+class DoclingParseV4DocumentBackend(PdfDocumentBackend):
+    def __init__(
+        self,
+        in_doc: "InputDocument",
+        path_or_stream: Union[BytesIO, Path],
+        options: PdfBackendOptions = PdfBackendOptions(),
+    ):
+        super().__init__(in_doc, path_or_stream, options)
+
+        password = (
+            self.options.password.get_secret_value() if self.options.password else None
+        )
+        with pypdfium2_lock:
+            self._pdoc = pdfium.PdfDocument(self.path_or_stream, password=password)
+        self.parser = DoclingPdfParser(loglevel="fatal")
+        self.dp_doc: PdfDocument = self.parser.load(
+            path_or_stream=self.path_or_stream, password=password
+        )
+        success = self.dp_doc is not None
+
+        if not success:
+            raise RuntimeError(
+                f"docling-parse v4 could not load document {self.document_hash}."
+            )
+
+    def page_count(self) -> int:
+        # return len(self._pdoc)  # To be replaced with docling-parse API
+
+        len_1 = len(self._pdoc)
+        len_2 = self.dp_doc.number_of_pages()
+
+        if len_1 != len_2:
+            _log.error(f"Inconsistent number of pages: {len_1}!={len_2}")
+
+        return len_2
+
+    def load_page(
+        self, page_no: int, create_words: bool = True, create_textlines: bool = True
+    ) -> DoclingParseV4PageBackend:
+        with pypdfium2_lock:
+            ppage = self._pdoc[page_no]
+
+        return DoclingParseV4PageBackend(
+            dp_doc=self.dp_doc,
+            page_obj=ppage,
+            page_no=page_no,
+            create_words=create_words,
+            create_textlines=create_textlines,
+        )
+
+    def is_valid(self) -> bool:
+        return self.page_count() > 0
+
+    def unload(self):
+        super().unload()
+        # Unload docling-parse document first
+        if self.dp_doc is not None:
+            self.dp_doc.unload()
+            self.dp_doc = None
+
+        # Then close pypdfium2 document with proper locking
+        if self._pdoc is not None:
+            with pypdfium2_lock:
+                try:
+                    self._pdoc.close()
+                except Exception:
+                    # Ignore cleanup errors
+                    pass
+            self._pdoc = None

docling/backend/docx/drawingml/utils.py

@@ -0,0 +1,131 @@
+import os
+import shutil
+import subprocess
+from pathlib import Path
+from tempfile import mkdtemp
+from typing import Callable, Optional
+
+import pypdfium2
+from docx.document import Document
+from PIL import Image, ImageChops
+
+
+def get_libreoffice_cmd(raise_if_unavailable: bool = False) -> Optional[str]:
+    """Return the libreoffice cmd and optionally test it."""
+
+    libreoffice_cmd = (
+        shutil.which("libreoffice")
+        or shutil.which("soffice")
+        or (
+            "/Applications/LibreOffice.app/Contents/MacOS/soffice"
+            if os.path.isfile("/Applications/LibreOffice.app/Contents/MacOS/soffice")
+            else None
+        )
+    )
+
+    if raise_if_unavailable:
+        if libreoffice_cmd is None:
+            raise RuntimeError("Libreoffice not found")
+
+        # The following test will raise if the libreoffice_cmd cannot be used
+        subprocess.run(
+            [
+                libreoffice_cmd,
+                "-h",
+            ],
+            stdout=subprocess.DEVNULL,
+            stderr=subprocess.DEVNULL,
+            check=True,
+        )
+
+    return libreoffice_cmd
+
+
+def get_docx_to_pdf_converter() -> Optional[Callable]:
+    """
+    Detects the best available DOCX to PDF tool and returns a conversion function.
+    The returned function accepts (input_path, output_path).
+    Returns None if no tool is available.
+    """
+
+    # Try LibreOffice
+    libreoffice_cmd = get_libreoffice_cmd()
+
+    if libreoffice_cmd:
+
+        def convert_with_libreoffice(input_path, output_path):
+            subprocess.run(
+                [
+                    libreoffice_cmd,
+                    "--headless",
+                    "--convert-to",
+                    "pdf",
+                    "--outdir",
+                    os.path.dirname(output_path),
+                    input_path,
+                ],
+                stdout=subprocess.DEVNULL,
+                stderr=subprocess.DEVNULL,
+                check=True,
+            )
+
+            expected_output = os.path.join(
+                os.path.dirname(output_path),
+                os.path.splitext(os.path.basename(input_path))[0] + ".pdf",
+            )
+            if expected_output != output_path:
+                os.rename(expected_output, output_path)
+
+        return convert_with_libreoffice
+
+    ## Space for other DOCX to PDF converters if available
+
+    # No tools found
+    return None
+
+
+def crop_whitespace(image: Image.Image, bg_color=None, padding=0) -> Image.Image:
+    if bg_color is None:
+        bg_color = image.getpixel((0, 0))
+
+    bg = Image.new(image.mode, image.size, bg_color)
+    diff = ImageChops.difference(image, bg)
+    bbox = diff.getbbox()
+
+    if bbox:
+        left, upper, right, lower = bbox
+        left = max(0, left - padding)
+        upper = max(0, upper - padding)
+        right = min(image.width, right + padding)
+        lower = min(image.height, lower + padding)
+        return image.crop((left, upper, right, lower))
+    else:
+        return image
+
+
+def get_pil_from_dml_docx(
+    docx: Document, converter: Optional[Callable]
+) -> Optional[Image.Image]:
+    if converter is None:
+        return None
+
+    temp_dir = Path(mkdtemp())
+    temp_docx = Path(temp_dir / "drawing_only.docx")
+    temp_pdf = Path(temp_dir / "drawing_only.pdf")
+
+    # 1) Save docx temporarily
+    docx.save(str(temp_docx))
+
+    # 2) Export to PDF
+    converter(temp_docx, temp_pdf)
+
+    # 3) Load PDF as PNG
+    pdf = pypdfium2.PdfDocument(temp_pdf)
+    page = pdf[0]
+    image = crop_whitespace(page.render(scale=2).to_pil())
+    page.close()
+    pdf.close()
+
+    shutil.rmtree(temp_dir, ignore_errors=True)
+
+    return image

docling/backend/docx/latex/latex_dict.py

@@ -0,0 +1,274 @@
+"""
+Adapted from https://github.com/xiilei/dwml/blob/master/dwml/latex_dict.py
+On 23/01/2025
+"""
+
+CHARS = ("{", "}", "_", "^", "#", "&", "$", "%", "~")
+
+BLANK = ""
+BACKSLASH = "\\"
+ALN = "&"
+
+CHR = {
+    # Unicode : Latex Math Symbols
+    # Top accents
+    "\u0300": "\\grave{{{0}}}",
+    "\u0301": "\\acute{{{0}}}",
+    "\u0302": "\\hat{{{0}}}",
+    "\u0303": "\\tilde{{{0}}}",
+    "\u0304": "\\bar{{{0}}}",
+    "\u0305": "\\overbar{{{0}}}",
+    "\u0306": "\\breve{{{0}}}",
+    "\u0307": "\\dot{{{0}}}",
+    "\u0308": "\\ddot{{{0}}}",
+    "\u0309": "\\ovhook{{{0}}}",
+    "\u030a": "\\ocirc{{{0}}}}",
+    "\u030c": "\\check{{{0}}}}",
+    "\u0310": "\\candra{{{0}}}",
+    "\u0312": "\\oturnedcomma{{{0}}}",
+    "\u0315": "\\ocommatopright{{{0}}}",
+    "\u031a": "\\droang{{{0}}}",
+    "\u0338": "\\not{{{0}}}",
+    "\u20d0": "\\leftharpoonaccent{{{0}}}",
+    "\u20d1": "\\rightharpoonaccent{{{0}}}",
+    "\u20d2": "\\vertoverlay{{{0}}}",
+    "\u20d6": "\\overleftarrow{{{0}}}",
+    "\u20d7": "\\vec{{{0}}}",
+    "\u20db": "\\dddot{{{0}}}",
+    "\u20dc": "\\ddddot{{{0}}}",
+    "\u20e1": "\\overleftrightarrow{{{0}}}",
+    "\u20e7": "\\annuity{{{0}}}",
+    "\u20e9": "\\widebridgeabove{{{0}}}",
+    "\u20f0": "\\asteraccent{{{0}}}",
+    # Bottom accents
+    "\u0330": "\\wideutilde{{{0}}}",
+    "\u0331": "\\underbar{{{0}}}",
+    "\u20e8": "\\threeunderdot{{{0}}}",
+    "\u20ec": "\\underrightharpoondown{{{0}}}",
+    "\u20ed": "\\underleftharpoondown{{{0}}}",
+    "\u20ee": "\\underledtarrow{{{0}}}",
+    "\u20ef": "\\underrightarrow{{{0}}}",
+    # Over | group
+    "\u23b4": "\\overbracket{{{0}}}",
+    "\u23dc": "\\overparen{{{0}}}",
+    "\u23de": "\\overbrace{{{0}}}",
+    # Under| group
+    "\u23b5": "\\underbracket{{{0}}}",
+    "\u23dd": "\\underparen{{{0}}}",
+    "\u23df": "\\underbrace{{{0}}}",
+}
+
+CHR_BO = {
+    # Big operators,
+    "\u2140": "\\Bbbsum",
+    "\u220f": "\\prod",
+    "\u2210": "\\coprod",
+    "\u2211": "\\sum",
+    "\u222b": "\\int",
+    "\u222c": "\\iint",
+    "\u222d": "\\iiint",
+    "\u222e": "\\oint",
+    "\u222f": "\\oiint",
+    "\u2230": "\\oiiint",
+    "\u22c0": "\\bigwedge",
+    "\u22c1": "\\bigvee",
+    "\u22c2": "\\bigcap",
+    "\u22c3": "\\bigcup",
+    "\u2a00": "\\bigodot",
+    "\u2a01": "\\bigoplus",
+    "\u2a02": "\\bigotimes",
+}
+
+T = {
+    # Greek letters
+    "\U0001d6fc": "\\alpha ",
+    "\U0001d6fd": "\\beta ",
+    "\U0001d6fe": "\\gamma ",
+    "\U0001d6ff": "\\theta ",
+    "\U0001d700": "\\epsilon ",
+    "\U0001d701": "\\zeta ",
+    "\U0001d702": "\\eta ",
+    "\U0001d703": "\\theta ",
+    "\U0001d704": "\\iota ",
+    "\U0001d705": "\\kappa ",
+    "\U0001d706": "\\lambda ",
+    "\U0001d707": "\\m ",
+    "\U0001d708": "\\n ",
+    "\U0001d709": "\\xi ",
+    "\U0001d70a": "\\omicron ",
+    "\U0001d70b": "\\pi ",
+    "\U0001d70c": "\\rho ",
+    "\U0001d70d": "\\varsigma ",
+    "\U0001d70e": "\\sigma ",
+    "\U0001d70f": "\\ta ",
+    "\U0001d710": "\\upsilon ",
+    "\U0001d711": "\\phi ",
+    "\U0001d712": "\\chi ",
+    "\U0001d713": "\\psi ",
+    "\U0001d714": "\\omega ",
+    "\U0001d715": "\\partial ",
+    "\U0001d716": "\\varepsilon ",
+    "\U0001d717": "\\vartheta ",
+    "\U0001d718": "\\varkappa ",
+    "\U0001d719": "\\varphi ",
+    "\U0001d71a": "\\varrho ",
+    "\U0001d71b": "\\varpi ",
+    # Relation symbols
+    "\u2190": "\\leftarrow ",
+    "\u2191": "\\uparrow ",
+    "\u2192": "\\rightarrow ",
+    "\u2193": "\\downright ",
+    "\u2194": "\\leftrightarrow ",
+    "\u2195": "\\updownarrow ",
+    "\u2196": "\\nwarrow ",
+    "\u2197": "\\nearrow ",
+    "\u2198": "\\searrow ",
+    "\u2199": "\\swarrow ",
+    "\u22ee": "\\vdots ",
+    "\u22ef": "\\cdots ",
+    "\u22f0": "\\adots ",
+    "\u22f1": "\\ddots ",
+    "\u2260": "\\ne ",
+    "\u2264": "\\leq ",
+    "\u2265": "\\geq ",
+    "\u2266": "\\leqq ",
+    "\u2267": "\\geqq ",
+    "\u2268": "\\lneqq ",
+    "\u2269": "\\gneqq ",
+    "\u226a": "\\ll ",
+    "\u226b": "\\gg ",
+    "\u2208": "\\in ",
+    "\u2209": "\\notin ",
+    "\u220b": "\\ni ",
+    "\u220c": "\\nni ",
+    # Ordinary symbols
+    "\u221e": "\\infty ",
+    # Binary relations
+    "\u00b1": "\\pm ",
+    "\u2213": "\\mp ",
+    # Italic, Latin, uppercase
+    "\U0001d434": "A",
+    "\U0001d435": "B",
+    "\U0001d436": "C",
+    "\U0001d437": "D",
+    "\U0001d438": "E",
+    "\U0001d439": "F",
+    "\U0001d43a": "G",
+    "\U0001d43b": "H",
+    "\U0001d43c": "I",
+    "\U0001d43d": "J",
+    "\U0001d43e": "K",
+    "\U0001d43f": "L",
+    "\U0001d440": "M",
+    "\U0001d441": "N",
+    "\U0001d442": "O",
+    "\U0001d443": "P",
+    "\U0001d444": "Q",
+    "\U0001d445": "R",
+    "\U0001d446": "S",
+    "\U0001d447": "T",
+    "\U0001d448": "U",
+    "\U0001d449": "V",
+    "\U0001d44a": "W",
+    "\U0001d44b": "X",
+    "\U0001d44c": "Y",
+    "\U0001d44d": "Z",
+    # Italic, Latin, lowercase
+    "\U0001d44e": "a",
+    "\U0001d44f": "b",
+    "\U0001d450": "c",
+    "\U0001d451": "d",
+    "\U0001d452": "e",
+    "\U0001d453": "f",
+    "\U0001d454": "g",
+    "\U0001d456": "i",
+    "\U0001d457": "j",
+    "\U0001d458": "k",
+    "\U0001d459": "l",
+    "\U0001d45a": "m",
+    "\U0001d45b": "n",
+    "\U0001d45c": "o",
+    "\U0001d45d": "p",
+    "\U0001d45e": "q",
+    "\U0001d45f": "r",
+    "\U0001d460": "s",
+    "\U0001d461": "t",
+    "\U0001d462": "u",
+    "\U0001d463": "v",
+    "\U0001d464": "w",
+    "\U0001d465": "x",
+    "\U0001d466": "y",
+    "\U0001d467": "z",
+}
+
+FUNC = {
+    "sin": "\\sin({fe})",
+    "cos": "\\cos({fe})",
+    "tan": "\\tan({fe})",
+    "arcsin": "\\arcsin({fe})",
+    "arccos": "\\arccos({fe})",
+    "arctan": "\\arctan({fe})",
+    "arccot": "\\arccot({fe})",
+    "sinh": "\\sinh({fe})",
+    "cosh": "\\cosh({fe})",
+    "tanh": "\\tanh({fe})",
+    "coth": "\\coth({fe})",
+    "sec": "\\sec({fe})",
+    "csc": "\\csc({fe})",
+    "mod": "\\mod {fe}",
+    "max": "\\max({fe})",
+    "min": "\\min({fe})",
+}
+
+FUNC_PLACE = "{fe}"
+
+BRK = "\\\\"
+
+CHR_DEFAULT = {
+    "ACC_VAL": "\\hat{{{0}}}",
+}
+
+POS = {
+    "top": "\\overline{{{0}}}",  # not sure
+    "bot": "\\underline{{{0}}}",
+}
+
+POS_DEFAULT = {
+    "BAR_VAL": "\\overline{{{0}}}",
+}
+
+SUB = "_{{{0}}}"
+
+SUP = "^{{{0}}}"
+
+F = {
+    "bar": "\\frac{{{num}}}{{{den}}}",
+    "skw": r"^{{{num}}}/_{{{den}}}",
+    "noBar": "\\genfrac{{}}{{}}{{0pt}}{{}}{{{num}}}{{{den}}}",
+    "lin": "{{{num}}}/{{{den}}}",
+}
+F_DEFAULT = "\\frac{{{num}}}{{{den}}}"
+
+D = "\\left{left}{text}\\right{right}"
+
+D_DEFAULT = {
+    "left": "(",
+    "right": ")",
+    "null": ".",
+}
+
+RAD = "\\sqrt[{deg}]{{{text}}}"
+RAD_DEFAULT = "\\sqrt{{{text}}}"
+ARR = "{text}"
+
+LIM_FUNC = {
+    "lim": "\\lim_{{{lim}}}",
+    "max": "\\max_{{{lim}}}",
+    "min": "\\min_{{{lim}}}",
+}
+
+LIM_TO = ("\\rightarrow", "\\to")
+
+LIM_UPP = "\\overset{{{lim}}}{{{text}}}"
+
+M = "\\begin{{matrix}}{text}\\end{{matrix}}"