Working on a first version of DOCX native backend

Signed-off-by: Maxim Lysak <mly@zurich.ibm.com>

docling/backend/msword_backend.py

@@ -2,19 +2,52 @@ from io import BytesIO
 from pathlib import Path
 from typing import Set, Union
 
+import docx
 from docling_core.types.experimental import (
+    BasePictureData,
+    BaseTableData,
     DescriptionItem,
     DocItemLabel,
     DoclingDocument,
+    GroupLabel,
+    ImageRef,
+    PictureItem,
+    SectionHeaderItem,
+    TableCell,
+    TableItem,
 )
+from lxml import etree
 
 from docling.backend.abstract_backend import DeclarativeDocumentBackend
 from docling.datamodel.base_models import InputFormat
 
 
 class MsWordDocumentBackend(DeclarativeDocumentBackend):
+
     def __init__(self, path_or_stream: Union[BytesIO, Path], document_hash: str):
+        self.XML_KEY = (
+            "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val"
+        )
         super().__init__(path_or_stream, document_hash)
+        # self.initialise(path_or_stream)
+        # Word file:
+        self.path_or_stream = path_or_stream
+        # Initialise the parents for the hierarchy
+        self.max_levels = 10
+        self.level_at_new_list = None
+        self.parents = {}
+        for i in range(-1, self.max_levels):
+            self.parents[i] = None
+
+        self.history = {
+            "names": [None],
+            "levels": [None],
+            "numids": [None],
+            "indents": [None],
+        }
+
+    def warn(self, message):
+        print(f"WARN: {message}")
 
     def is_valid(self) -> bool:
         return True
@@ -28,11 +61,333 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
 
         self.path_or_stream = None
 
+    def update_history(self, name, level, numid, ilevel):
+        self.history["names"].append(name)
+        self.history["levels"].append(level)
+
+        self.history["numids"].append(numid)
+        self.history["indents"].append(ilevel)
+
+    def prev_name(self):
+        return self.history["names"][-1]
+
+    def prev_level(self):
+        return self.history["levels"][-1]
+
+    def prev_numid(self):
+        return self.history["numids"][-1]
+
+    def prev_indent(self):
+        return self.history["indents"][-1]
+
+    def get_level(self) -> int:
+        """Return the first None index."""
+        for k, v in self.parents.items():
+            if k >= 0 and v == None:
+                return k
+        return 0
+
     @classmethod
     def supported_formats(cls) -> Set[InputFormat]:
         return {InputFormat.DOCX}
 
-    def convert(self) -> DoclingDocument:
-        doc = DoclingDocument(description=DescriptionItem(), name="dummy")
-        doc.add_text(text="I am a Word document.", label=DocItemLabel.TEXT)
+    def walk_linear(self, body, docx_obj, doc) -> DoclingDocument:
+        for element in body:
+            tag_name = etree.QName(element).localname
+            # Check for Text
+            if tag_name in ["p"]:
+                self.add_text(element, docx_obj, doc)
+            # Check for Tables
+            elif element.tag.endswith("tbl"):
+                self.add_table(element, docx_obj, doc)
+            # Check for Inline Images (drawings or blip elements)
+            elif element.tag.endswith("drawing") or element.tag.endswith("blip"):
+                self.add_figure(element, docx_obj, doc)
+            else:
+                self.warn(f"ignoring element in DOCX with tag: {tag_name}")
+        return doc
+
+    def get_numId_and_ilvl(self, paragraph):
+        # Access the XML element of the paragraph
+        numPr = paragraph._element.find(
+            ".//w:numPr", namespaces=paragraph._element.nsmap
+        )
+
+        if numPr is not None:
+            # Get the numId element and extract the value
+            numId_elem = numPr.find("w:numId", namespaces=paragraph._element.nsmap)
+            ilvl_elem = numPr.find("w:ilvl", namespaces=paragraph._element.nsmap)
+
+            numId = numId_elem.get(self.XML_KEY) if numId_elem is not None else None
+            ilvl = ilvl_elem.get(self.XML_KEY) if ilvl_elem is not None else None
+
+            return int(numId), int(ilvl)
+
+        return None, None  # If the paragraph is not part of a list
+
+    def get_label_and_level(self, paragraph):
+        if paragraph.style is None:
+            return "Normal", None
+        label = paragraph.style.name
+        if ":" in label:
+            parts = label.split(":")
+
+            if len(parts) == 2:
+                return parts[0], int(parts[1])
+
+        parts = label.split(" ")
+
+        if "Heading" in label and len(parts) == 2:
+            return parts[0], int(parts[1])
+        else:
+            return label, None
+
+    def add_text(self, element, docx_obj, doc):
+        paragraph = docx.text.paragraph.Paragraph(element, docx_obj)
+
+        if paragraph.text is None:
+            # self.warn(f"paragraph has text==None")
+            return
+
+        text = paragraph.text.strip()
+
+        # if len(text)==0: # keep empty paragraphs, they seperate adjacent lists!
+        #     self.warn(f"paragraph has len(text)==0")
+
+        pname, plevel = self.get_label_and_level(paragraph)
+        numid, ilevel = self.get_numId_and_ilvl(paragraph)
+
+        # we detected a list
+        if numid is not None and ilevel is not None:
+            self.add_listitem(
+                element, docx_obj, doc, pname, plevel, numid, ilevel, text
+            )
+            self.update_history(pname, plevel, numid, ilevel)
+
+            return
+
+        elif numid is None and self.prev_numid() is not None:  # Close list
+
+            for key, val in self.parents.items():
+                if key >= self.level_at_new_list:
+                    self.parents[key] = None
+
+            self.level = self.level_at_new_list - 1
+            self.level_at_new_list = None
+
+        if pname in ["Title"]:
+
+            for key, val in self.parents.items():
+                self.parents[key] = None
+
+            self.parents[0] = doc.add_text(
+                parent=None, label=DocItemLabel.TITLE, text=text
+            )
+
+        elif "Heading" in pname:
+            self.add_header(element, docx_obj, doc, pname, plevel, text)
+
+        elif pname in [
+            "Paragraph",
+            "Normal",
+            "Subtitle",
+            "Author",
+            "Default Text",
+            "List Paragraph",
+            "List Bullet",
+            "Quote",
+        ]:
+            level = self.get_level()
+            doc.add_text(
+                label=DocItemLabel.PARAGRAPH, parent=self.parents[level - 1], text=text
+            )
+
+        else:
+            assert False, f"need to add a new paragraph: {pname}"
+
+        self.update_history(pname, plevel, numid, ilevel)
+
+    def add_header(self, element, docx_obj, doc, curr_name, curr_level, text: str):
+
+        level = self.get_level()
+        # print(f"level: {level} => add_header(self, element, docx_obj, doc, {curr_name}, {curr_level}): {text}")
+
+        if isinstance(curr_level, int):
+
+            if curr_level == level:
+
+                self.parents[level] = doc.add_heading(
+                    parent=self.parents[level - 1], text=text
+                )
+
+            elif curr_level > level:
+
+                # add invisible group
+                for i in range(level, curr_level):
+                    self.parents[i] = doc.add_group(
+                        parent=self.parents[i - 1],
+                        label=GroupLabel.SECTION,
+                        name=f"header-{i}",
+                    )
+
+                self.parents[curr_level] = doc.add_heading(
+                    parent=self.parents[curr_level - 1], text=text
+                )
+
+            elif curr_level < level:
+
+                # remove the tail
+                for key, val in self.parents.items():
+                    if key >= curr_level:
+                        self.parents[key] = None
+
+                self.parents[curr_level] = doc.add_heading(
+                    parent=self.parents[curr_level - 1], text=text
+                )
+
+        else:
+            self.parents[self.level] = doc.add_heading(
+                parent=self.parents[self.level - 1], text=text
+            )
+
+    def add_listitem(
+        self, element, docx_obj, doc, pname, plevel, numid, ilevel, text: str
+    ):
+
+        level = self.get_level()
+        # print(f"level: {level} => add_listitem(self, element, docx_obj, doc, {pname}, {plevel}, {numid}, {ilevel}): {text}")
+
+        if self.prev_numid() is None:  # Open new list
+
+            self.level_at_new_list = level
+
+            self.parents[level] = doc.add_group(
+                label=GroupLabel.LIST, name="list", parent=self.parents[level - 1]
+            )
+
+            doc.add_text(
+                label=DocItemLabel.LIST_ITEM, parent=self.parents[level], text=text
+            )
+
+        elif (
+            self.prev_numid() == numid and self.prev_indent() < ilevel
+        ):  # Open indented list
+
+            for i in range(
+                self.level_at_new_list + self.prev_indent() + 1,
+                self.level_at_new_list + ilevel + 1,
+            ):
+                self.parents[i] = doc.add_group(
+                    label=GroupLabel.LIST, name="list", parent=self.parents[i - 1]
+                )
+
+            doc.add_text(
+                label=DocItemLabel.LIST_ITEM,
+                parent=self.parents[self.level_at_new_list + ilevel],
+                text=text,
+            )
+
+        elif self.prev_numid() == numid and ilevel < self.prev_indent():  # Close list
+
+            for k, v in self.parents.items():
+                if k > self.level_at_new_list + ilevel:
+                    self.parents[k] = None
+
+            doc.add_text(
+                label=DocItemLabel.LIST_ITEM,
+                parent=self.parents[self.level_at_new_list + ilevel],
+                text=text,
+            )
+
+        elif self.prev_numid() == numid or self.prev_indent() == ilevel:
+
+            doc.add_text(
+                label=DocItemLabel.LIST_ITEM, parent=self.parents[level - 1], text=text
+            )
+
+    def add_table(self, element, docx_obj, doc):
+
+        # Function to check if a cell has a colspan (gridSpan)
+        def get_colspan(cell):
+            grid_span = cell._element.xpath("@w:gridSpan")
+            if grid_span:
+                return int(grid_span[0])  # Return the number of columns spanned
+            return 1  # Default is 1 (no colspan)
+
+        # Function to check if a cell has a rowspan (vMerge)
+        def get_rowspan(cell):
+            v_merge = cell._element.xpath("@w:vMerge")
+            if v_merge:
+                return v_merge[
+                    0
+                ]  # 'restart' indicates the beginning of a rowspan, others are continuation
+            return 1
+
+        table = docx.table.Table(element, docx_obj)
+
+        num_rows = len(table.rows)
+        num_cols = 0
+        for row in table.rows:
+            # Calculate the max number of columns
+            num_cols = max(num_cols, sum(get_colspan(cell) for cell in row.cells))
+
+        self.warn(f"table: [{num_rows}x{num_cols}]")
+
+        # Initialize the table grid
+        table_grid = [[None for _ in range(num_cols)] for _ in range(num_rows)]
+
+        data = BaseTableData(num_rows=num_rows, num_cols=num_cols, table_cells=[])
+
+        for row_idx, row in enumerate(table.rows):
+            col_idx = 0
+            for c, cell in enumerate(row.cells):
+                row_span = get_rowspan(cell)
+                col_span = get_colspan(cell)
+
+                # Find the next available column in the grid
+                while table_grid[row_idx][col_idx] is not None:
+                    col_idx += 1
+
+                print(f"{row_idx}, {col_idx}, {row_span}, {col_span}")
+
+                # Fill the grid with the cell value, considering rowspan and colspan
+                for i in range(row_span if row_span == "restart" else 1):
+                    for j in range(col_span):
+                        table_grid[row_idx + i][col_idx + j] = ""
+
+                cell = TableCell(
+                    text=cell.text,
+                    row_span=row_span,
+                    col_span=col_span,
+                    start_row_offset_idx=row_idx,
+                    end_row_offset_idx=row_idx + row_span,
+                    start_col_offset_idx=col_idx,
+                    end_col_offset_idx=col_idx + col_span,
+                    col_header=False,  # col_header,
+                    row_header=False,  # ((not col_header) and html_cell.name=='th')
+                )
+
+                # print(row_idx, "\t", col_idx, "\t", row_span, "\t", col_span, "\t", text)
+                data.table_cells.append(cell)
+
+        level = self.get_level()
+        doc.add_table(data=data, parent=self.parents[level - 1])
+
+    def add_figure(self, element, docx_obj, doc):
+        doc.add_picture(
+            data=BasePictureData(), parent=self.parents[self.level], caption=None
+        )
+
+    def convert(self) -> DoclingDocument:
+        # Parses the DOCX into a structured document model.
+        doc = DoclingDocument(description=DescriptionItem(), name="dummy")
+        docx_obj = None
+        try:
+            docx_obj = docx.Document(self.path_or_stream)
+        except Exception:
+            return doc
+
+        # self.initialise()
+        doc = self.walk_linear(docx_obj.element.body, docx_obj, doc)
+
         return doc

examples/run_with_formats.py

@@ -13,10 +13,11 @@ from docling.pipeline.simple_model_pipeline import SimpleModelPipeline
 from docling.pipeline.standard_pdf_model_pipeline import StandardPdfModelPipeline
 
 input_paths = [
-    Path("tests/data/wiki_duck.html"),
+    # Path("tests/data/wiki_duck.html"),
     Path("tests/data/word_sample.docx"),
-    Path("tests/data/powerpoint_sample.pptx"),
-    Path("tests/data/2206.01062.pdf"),
+    Path("tests/data/lorem_ipsum.docx"),
+    # Path("tests/data/powerpoint_sample.pptx"),
+    # Path("tests/data/2206.01062.pdf"),
 ]
 input = DocumentConversionInput.from_paths(input_paths)