Function _add_child_elements

docling/utils/glm_utils.py:332–361

(container_el, doc, obj, pelem)

def _add_child_elements(container_el, doc, obj, pelem):
    payload = obj.get("payload")
    if payload is not None:
        children = payload.get("children", [])

        for child in children:
            c_label = DocItemLabel(child["label"])
            c_bbox = BoundingBox.model_validate(child["bbox"]).to_bottom_left_origin(
                doc.pages[pelem["page"]].size.height
            )
            c_text = " ".join(
                [
                    cell["text"].replace("\x02", "-").strip()
                    for cell in child["cells"]
                    if len(cell["text"].strip()) > 0
                ]
            )

            c_prov = ProvenanceItem(
                page_no=pelem["page"], charspan=(0, len(c_text)), bbox=c_bbox
            )
            if c_label == DocItemLabel.LIST_ITEM:
                # TODO: Infer if this is a numbered or a bullet list item
                doc.add_list_item(parent=container_el, text=c_text, prov=c_prov)
            elif c_label == DocItemLabel.SECTION_HEADER:
                doc.add_heading(parent=container_el, text=c_text, prov=c_prov)
            else:
                doc.add_text(
                    parent=container_el, label=c_label, text=c_text, prov=c_prov
                )