@stll/folio-core 0.47.7 → 0.49.0

package/dist/docx/paragraphProperties.js

@@ -1,6 +1,6 @@
 import { readAttributeBag } from "./attributeRemainder.js";
 import { parseBorderSpec } from "./borderParser.js";
-import { CAPTURE, dispatchChildren, ownedElsewhere, sequencePositions } from "./containerChildren.js";
+import { CAPTURE, dispatchChildrenWithContext, ownedElsewhere, sequencePositions } from "./containerChildren.js";
 import { readParagraphNumbering } from "./numberingReference.js";
 import { FrameWrapSchema, FrameXAlignSchema, FrameYAlignSchema, LineSpacingRuleSchema, ParagraphAlignmentSchema, TabLeaderSchema, TabStopAlignmentSchema, narrowEnum } from "./parserEnums.js";
 import { FRAME_ATTRIBUTES, INDENTATION_ATTRIBUTES, SPACING_ATTRIBUTES, TAB_STOP_ATTRIBUTES } from "./propertyElementAttributes.js";
@@ -74,6 +74,123 @@ function parseFrameProperties(framePr) {
 	return frame;
 }
 /**
+* The content model declares each property once. A source that states one
+* twice keeps the first as the model's value and the repeat as bytes at the
+* same schema ordinal, rather than letting the second silently win or fall off
+* the end of the walk.
+*/
+const readOnce = (name, read) => (child, { formatting, taken }) => {
+	if (taken.has(name) || !read(child, formatting)) return CAPTURE;
+	taken.add(name);
+};
+const readToggle = (name, field) => readOnce(name, (child, formatting) => {
+	formatting[field] = parseBooleanElement(child);
+	return true;
+});
+/**
+* The `w:pPr` decision map. Built once: every reader takes the record it fills
+* from the walk's context rather than closing over it.
+*/
+const PARAGRAPH_PROPERTY_HANDLERS = {
+	pStyle: readOnce("pStyle", (child, formatting) => {
+		const val = getAttribute(child, "w", "val");
+		if (!val) return false;
+		formatting.styleId = val;
+		return true;
+	}),
+	keepNext: readToggle("keepNext", "keepNext"),
+	keepLines: readToggle("keepLines", "keepLines"),
+	pageBreakBefore: readToggle("pageBreakBefore", "pageBreakBefore"),
+	framePr: readOnce("framePr", (child, formatting) => {
+		const frame = parseFrameProperties(child);
+		if (frame === void 0) return false;
+		formatting.frame = frame;
+		return true;
+	}),
+	widowControl: readToggle("widowControl", "widowControl"),
+	numPr: readOnce("numPr", (child, formatting) => {
+		const stated = readParagraphNumbering(child);
+		if (stated !== void 0) formatting.numPr = stated;
+		const numberingChange = findChild(child, "w", "numberingChange");
+		if (numberingChange) formatting.numberingChangeXml = captureVerbatimXml(numberingChange);
+		const numberingInsertion = findChild(child, "w", "ins");
+		if (numberingInsertion) formatting.numberingInsertionXml = captureVerbatimXml(numberingInsertion);
+		const statedLevel = parseNumericAttribute(findChild(child, "w", "ilvl"), "w", "val");
+		return stated !== void 0 || numberingChange !== null || numberingInsertion !== null || statedLevel === -1;
+	}),
+	suppressLineNumbers: readToggle("suppressLineNumbers", "suppressLineNumbers"),
+	pBdr: readOnce("pBdr", (child, formatting) => {
+		const borders = {};
+		for (const side of PARAGRAPH_BORDER_SIDES) {
+			const border = parseBorderSpec(findChild(child, "w", side));
+			if (border) borders[side] = border;
+		}
+		if (Object.keys(borders).length === 0) return false;
+		formatting.borders = borders;
+		return true;
+	}),
+	shd: readOnce("shd", (child, formatting) => {
+		const shading = parseShading(child);
+		if (shading === void 0) return false;
+		formatting.shading = shading;
+		return true;
+	}),
+	tabs: readOnce("tabs", (child, formatting) => {
+		const tabs = parseTabStops(child);
+		if (tabs === void 0) return false;
+		formatting.tabs = tabs;
+		return true;
+	}),
+	suppressAutoHyphens: readToggle("suppressAutoHyphens", "suppressAutoHyphens"),
+	kinsoku: readToggle("kinsoku", "kinsoku"),
+	wordWrap: CAPTURE,
+	overflowPunct: readToggle("overflowPunct", "overflowPunctuation"),
+	topLinePunct: CAPTURE,
+	autoSpaceDE: CAPTURE,
+	autoSpaceDN: CAPTURE,
+	bidi: readToggle("bidi", "bidi"),
+	adjustRightInd: CAPTURE,
+	snapToGrid: readToggle("snapToGrid", "snapToGrid"),
+	spacing: readOnce("spacing", (child, formatting) => readParagraphSpacing(child, formatting)),
+	ind: readOnce("ind", (child, formatting) => readParagraphIndentation(child, formatting)),
+	contextualSpacing: readToggle("contextualSpacing", "contextualSpacing"),
+	mirrorIndents: CAPTURE,
+	suppressOverlap: CAPTURE,
+	jc: readOnce("jc", (child, formatting) => {
+		const val = narrowEnum(getAttribute(child, "w", "val"), ParagraphAlignmentSchema);
+		if (!val) return false;
+		formatting.alignment = val;
+		return true;
+	}),
+	textDirection: CAPTURE,
+	textAlignment: CAPTURE,
+	textboxTightWrap: CAPTURE,
+	outlineLvl: readOnce("outlineLvl", (child, formatting) => {
+		const val = parseNumericAttribute(child, "w", "val");
+		const level = val === void 0 ? void 0 : outlineLevelFromStatedValue(val);
+		if (level === void 0) return val === -1;
+		formatting.outlineLevel = level;
+		return true;
+	}),
+	divId: CAPTURE,
+	cnfStyle: CAPTURE,
+	rPr: ownedElsewhere({
+		container: "paragraph-properties",
+		child: "rPr",
+		reader: "paragraphProperties#parseParagraphProperties"
+	}),
+	sectPr: ownedElsewhere({
+		container: "paragraph-properties",
+		child: "sectPr",
+		reader: "sectionParser#parseSectionProperties"
+	}),
+	pPrChange: ownedElsewhere({
+		container: "paragraph-properties",
+		child: "pPrChange",
+		reader: "paragraphParser#parseParagraphPropertyChanges"
+	})
+};
+/**
 * Read `w:pPr` into {@link ParagraphFormatting}, every declared child carrying
 * a decision.
 *
@@ -93,117 +210,14 @@ function parseFrameProperties(framePr) {
 function parseParagraphProperties(pPr, theme) {
 	if (!pPr) return;
 	const formatting = {};
-	const taken = /* @__PURE__ */ new Set();
-	const once = (name, read) => (child) => {
-		if (taken.has(name) || !read(child)) return CAPTURE;
-		taken.add(name);
-	};
-	const toggle = (name, field) => once(name, (child) => {
-		formatting[field] = parseBooleanElement(child);
-		return true;
-	});
-	const preserved = dispatchChildren({
+	const preserved = dispatchChildrenWithContext({
 		element: pPr,
 		container: "paragraph-properties",
 		capturePosition: sequencePositions("paragraph-properties", pPr),
-		handlers: {
-			pStyle: once("pStyle", (child) => {
-				const val = getAttribute(child, "w", "val");
-				if (!val) return false;
-				formatting.styleId = val;
-				return true;
-			}),
-			keepNext: toggle("keepNext", "keepNext"),
-			keepLines: toggle("keepLines", "keepLines"),
-			pageBreakBefore: toggle("pageBreakBefore", "pageBreakBefore"),
-			framePr: once("framePr", (child) => {
-				const frame = parseFrameProperties(child);
-				if (frame === void 0) return false;
-				formatting.frame = frame;
-				return true;
-			}),
-			widowControl: toggle("widowControl", "widowControl"),
-			numPr: once("numPr", (child) => {
-				const stated = readParagraphNumbering(child);
-				if (stated !== void 0) formatting.numPr = stated;
-				const numberingChange = findChild(child, "w", "numberingChange");
-				if (numberingChange) formatting.numberingChangeXml = captureVerbatimXml(numberingChange);
-				const numberingInsertion = findChild(child, "w", "ins");
-				if (numberingInsertion) formatting.numberingInsertionXml = captureVerbatimXml(numberingInsertion);
-				const statedLevel = parseNumericAttribute(findChild(child, "w", "ilvl"), "w", "val");
-				return stated !== void 0 || numberingChange !== null || numberingInsertion !== null || statedLevel === -1;
-			}),
-			suppressLineNumbers: toggle("suppressLineNumbers", "suppressLineNumbers"),
-			pBdr: once("pBdr", (child) => {
-				const borders = {};
-				for (const side of PARAGRAPH_BORDER_SIDES) {
-					const border = parseBorderSpec(findChild(child, "w", side));
-					if (border) borders[side] = border;
-				}
-				if (Object.keys(borders).length === 0) return false;
-				formatting.borders = borders;
-				return true;
-			}),
-			shd: once("shd", (child) => {
-				const shading = parseShading(child);
-				if (shading === void 0) return false;
-				formatting.shading = shading;
-				return true;
-			}),
-			tabs: once("tabs", (child) => {
-				const tabs = parseTabStops(child);
-				if (tabs === void 0) return false;
-				formatting.tabs = tabs;
-				return true;
-			}),
-			suppressAutoHyphens: toggle("suppressAutoHyphens", "suppressAutoHyphens"),
-			kinsoku: toggle("kinsoku", "kinsoku"),
-			wordWrap: CAPTURE,
-			overflowPunct: toggle("overflowPunct", "overflowPunctuation"),
-			topLinePunct: CAPTURE,
-			autoSpaceDE: CAPTURE,
-			autoSpaceDN: CAPTURE,
-			bidi: toggle("bidi", "bidi"),
-			adjustRightInd: CAPTURE,
-			snapToGrid: toggle("snapToGrid", "snapToGrid"),
-			spacing: once("spacing", (child) => readParagraphSpacing(child, formatting)),
-			ind: once("ind", (child) => readParagraphIndentation(child, formatting)),
-			contextualSpacing: toggle("contextualSpacing", "contextualSpacing"),
-			mirrorIndents: CAPTURE,
-			suppressOverlap: CAPTURE,
-			jc: once("jc", (child) => {
-				const val = narrowEnum(getAttribute(child, "w", "val"), ParagraphAlignmentSchema);
-				if (!val) return false;
-				formatting.alignment = val;
-				return true;
-			}),
-			textDirection: CAPTURE,
-			textAlignment: CAPTURE,
-			textboxTightWrap: CAPTURE,
-			outlineLvl: once("outlineLvl", (child) => {
-				const val = parseNumericAttribute(child, "w", "val");
-				const level = val === void 0 ? void 0 : outlineLevelFromStatedValue(val);
-				if (level === void 0) return val === -1;
-				formatting.outlineLevel = level;
-				return true;
-			}),
-			divId: CAPTURE,
-			cnfStyle: CAPTURE,
-			rPr: ownedElsewhere({
-				container: "paragraph-properties",
-				child: "rPr",
-				reader: "paragraphProperties#parseParagraphProperties"
-			}),
-			sectPr: ownedElsewhere({
-				container: "paragraph-properties",
-				child: "sectPr",
-				reader: "sectionParser#parseSectionProperties"
-			}),
-			pPrChange: ownedElsewhere({
-				container: "paragraph-properties",
-				child: "pPrChange",
-				reader: "paragraphParser#parseParagraphPropertyChanges"
-			})
+		handlers: PARAGRAPH_PROPERTY_HANDLERS,
+		context: {
+			formatting,
+			taken: /* @__PURE__ */ new Set()
 		}
 	});
 	const rPr = findChild(pPr, "w", "rPr");

package/dist/docx/paragraphPropertySource.d.ts

@@ -111,6 +111,11 @@ declare const recreateProseNodeWithParagraphPropertySource: (source: Node, optio
  * can only name a paragraph in the package it came from.
  */
 declare const recreateProseNodeWithDetachedParagraphPropertySource: (source: Node, options?: RecreateProseNodeOptions) => Node;
+/**
+ * A paragraph's attributes without its durable source token, for a copy of
+ * them that is carried by another node and must not claim the paragraph.
+ */
+declare const proseParagraphAttrsWithoutPropertySource: (attrs: Node["attrs"]) => Node["attrs"];
 type SetProseParagraphMarkupOptions = {
   attrs: Node["attrs"];
   ownership: "preserve" | "transfer-allocated-id";
@@ -129,4 +134,4 @@ declare const linkParagraphPropertySourceCandidate: (target: document_d_exports.
 declare const getParagraphPropertySourceCandidate: (paragraph: document_d_exports.Paragraph) => document_d_exports.Paragraph | undefined;
 declare const getParagraphPropertySourceTransferId: (paragraph: document_d_exports.Paragraph) => string | undefined;
 //#endregion
-export { DecodedTableCellParagraphSourcePayload, PARAGRAPH_PROPERTY_SOURCE_VALIDATION_CODES, PROSE_PARAGRAPH_SOURCE_CONTRACT_ATTR, PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR, ParagraphPropertySourceValidationCode, ParagraphPropertySourceValidationError, TABLE_CELL_PARAGRAPH_SOURCE_BINDING_ATTR, TABLE_CELL_PARAGRAPH_SOURCE_PAYLOAD_ERROR_CLASSIFICATIONS, TableCellParagraphPropertySourceBinding, TableCellParagraphSourcePayloadErrorClassification, assignDocumentParagraphPropertySourceContract, assignParagraphPropertySource, cloneDocumentWithParagraphPropertySources, cloneParagraphWithPropertySource, cloneParagraphWithoutPropertySource, cloneTableCellsWithParagraphPropertyCaptures, copyDocumentParagraphPropertySourceContract, copyDocumentParagraphPropertySources, copyParagraphPropertyCapture, copyParagraphPropertySource, createProseParagraphWithPropertySource, decodeTableCellParagraphSourcePayload, getDocumentParagraphPropertySourceContract, getExplicitParagraphPropertySourceTransfers, getParagraphPropertySource, getParagraphPropertySourceCandidate, getParagraphPropertySourceToken, getParagraphPropertySourceTransferId, getProseDocumentParagraphPropertySourceContract, getProseParagraphPropertySourceToken, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };
+export { DecodedTableCellParagraphSourcePayload, PARAGRAPH_PROPERTY_SOURCE_VALIDATION_CODES, PROSE_PARAGRAPH_SOURCE_CONTRACT_ATTR, PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR, ParagraphPropertySourceValidationCode, ParagraphPropertySourceValidationError, TABLE_CELL_PARAGRAPH_SOURCE_BINDING_ATTR, TABLE_CELL_PARAGRAPH_SOURCE_PAYLOAD_ERROR_CLASSIFICATIONS, TableCellParagraphPropertySourceBinding, TableCellParagraphSourcePayloadErrorClassification, assignDocumentParagraphPropertySourceContract, assignParagraphPropertySource, cloneDocumentWithParagraphPropertySources, cloneParagraphWithPropertySource, cloneParagraphWithoutPropertySource, cloneTableCellsWithParagraphPropertyCaptures, copyDocumentParagraphPropertySourceContract, copyDocumentParagraphPropertySources, copyParagraphPropertyCapture, copyParagraphPropertySource, createProseParagraphWithPropertySource, decodeTableCellParagraphSourcePayload, getDocumentParagraphPropertySourceContract, getExplicitParagraphPropertySourceTransfers, getParagraphPropertySource, getParagraphPropertySourceCandidate, getParagraphPropertySourceToken, getParagraphPropertySourceTransferId, getProseDocumentParagraphPropertySourceContract, getProseParagraphPropertySourceToken, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };

package/dist/docx/paragraphPropertySource.js

@@ -662,6 +662,14 @@ const recreateProseNodeWithDetachedParagraphPropertySource = (source, options =
 		} : attrs
 	});
 };
+/**
+* A paragraph's attributes without its durable source token, for a copy of
+* them that is carried by another node and must not claim the paragraph.
+*/
+const proseParagraphAttrsWithoutPropertySource = (attrs) => {
+	const { [PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR]: _token, ...rest } = attrs;
+	return rest;
+};
 /** Replace paragraph markup without losing its private parser-owner link. */
 const setProseParagraphMarkupWithPropertySource = ({ attrs, ownership, pos, transaction }) => {
 	const source = transaction.doc.nodeAt(pos);
@@ -695,4 +703,4 @@ const linkParagraphPropertySourceCandidate = (target, source) => {
 const getParagraphPropertySourceCandidate = (paragraph) => paragraphPropertySourceCandidates.get(paragraph);
 const getParagraphPropertySourceTransferId = (paragraph) => paragraphPropertySourceTransferIds.get(paragraph);
 //#endregion
-export { PARAGRAPH_PROPERTY_SOURCE_VALIDATION_CODES, PROSE_PARAGRAPH_SOURCE_CONTRACT_ATTR, PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR, ParagraphPropertySourceValidationError, TABLE_CELL_PARAGRAPH_SOURCE_BINDING_ATTR, TABLE_CELL_PARAGRAPH_SOURCE_PAYLOAD_ERROR_CLASSIFICATIONS, assignDocumentParagraphPropertySourceContract, assignParagraphPropertySource, cloneDocumentWithParagraphPropertySources, cloneParagraphWithPropertySource, cloneParagraphWithoutPropertySource, cloneTableCellsWithParagraphPropertyCaptures, copyDocumentParagraphPropertySourceContract, copyDocumentParagraphPropertySources, copyParagraphPropertyCapture, copyParagraphPropertySource, createProseParagraphWithPropertySource, decodeTableCellParagraphSourcePayload, getDocumentParagraphPropertySourceContract, getExplicitParagraphPropertySourceTransfers, getParagraphPropertySource, getParagraphPropertySourceCandidate, getParagraphPropertySourceToken, getParagraphPropertySourceTransferId, getProseDocumentParagraphPropertySourceContract, getProseParagraphPropertySourceToken, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };
+export { PARAGRAPH_PROPERTY_SOURCE_VALIDATION_CODES, PROSE_PARAGRAPH_SOURCE_CONTRACT_ATTR, PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR, ParagraphPropertySourceValidationError, TABLE_CELL_PARAGRAPH_SOURCE_BINDING_ATTR, TABLE_CELL_PARAGRAPH_SOURCE_PAYLOAD_ERROR_CLASSIFICATIONS, assignDocumentParagraphPropertySourceContract, assignParagraphPropertySource, cloneDocumentWithParagraphPropertySources, cloneParagraphWithPropertySource, cloneParagraphWithoutPropertySource, cloneTableCellsWithParagraphPropertyCaptures, copyDocumentParagraphPropertySourceContract, copyDocumentParagraphPropertySources, copyParagraphPropertyCapture, copyParagraphPropertySource, createProseParagraphWithPropertySource, decodeTableCellParagraphSourcePayload, getDocumentParagraphPropertySourceContract, getExplicitParagraphPropertySourceTransfers, getParagraphPropertySource, getParagraphPropertySourceCandidate, getParagraphPropertySourceToken, getParagraphPropertySourceTransferId, getProseDocumentParagraphPropertySourceContract, getProseParagraphPropertySourceToken, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };

package/dist/docx/paragraphTextBoxEnrichment.d.ts

@@ -2,9 +2,10 @@ import { document_d_exports } from "../types/document.js";
 import { NumberingMap } from "./numberingParser.js";
 import { ParseContext } from "./parseContext.js";
 import { XmlElement } from "./xmlParser.js";
+import { PreviewLedger } from "./previewBudget.js";
 import { StyleMap } from "./styleParser.js";
 import { TableParserFn } from "./textBoxParser.js";
 //#region src/docx/paragraphTextBoxEnrichment.d.ts
-declare const enrichParagraphTextBoxes: (paragraph: document_d_exports.Paragraph, paraXml: XmlElement, styles: StyleMap | null, theme: document_d_exports.Theme | null, numbering: NumberingMap | null, rels: document_d_exports.RelationshipMap | null, media: Map<string, document_d_exports.MediaFile> | null, parseTable: TableParserFn, context?: ParseContext) => void;
+declare const enrichParagraphTextBoxes: (paragraph: document_d_exports.Paragraph, paraXml: XmlElement, styles: StyleMap | null, theme: document_d_exports.Theme | null, numbering: NumberingMap | null, rels: document_d_exports.RelationshipMap | null, media: Map<string, document_d_exports.MediaFile> | null, parseTable: TableParserFn, previews: PreviewLedger, context?: ParseContext) => void;
 //#endregion
 export { enrichParagraphTextBoxes };

package/dist/docx/paragraphTextBoxEnrichment.js

@@ -1,9 +1,12 @@
 import { pixelsToEmu } from "../utils/units.js";
+import { groupTextContentFingerprint, groupXmlFingerprint } from "./drawingGroupChildren.js";
+import { groupPreviewTextBoxes } from "./groupDrawingParser.js";
 import { parseParagraph } from "./paragraphParser.js";
 import { consolidateParagraphContent } from "./runConsolidator.js";
-import { getTextBoxContentElement, parseTextBox, parseTextBoxContent, scanRunForTextBoxDrawings } from "./textBoxParser.js";
+import { captureShapeAlternateContent } from "./shapeAlternateContent.js";
+import { getTextBoxContentElement, parseTextBox, parseTextBoxContent, parseTextBoxFromShape, scanRunForTextBoxDrawings } from "./textBoxParser.js";
 import { isVmlPictParsedByRunParser } from "./vmlImageParser.js";
-import { findDeep, getAttribute, getChildElements, getLocalName } from "./xmlParser.js";
+import { findChildByLocalName, findDeep, getAttribute, getChildElements, getLocalName } from "./xmlParser.js";
 //#region src/docx/paragraphTextBoxEnrichment.ts
 const VML_HORIZONTAL_RELATIVES = /* @__PURE__ */ new Set([
 	"character",
@@ -121,7 +124,7 @@ const vmlWrapType = (positioned, zIndex) => {
 	if (Number.isFinite(zIndex) && zIndex < 0) return "behind";
 	return "inFront";
 };
-const enrichParagraphTextBoxes = (paragraph, paraXml, styles, theme, numbering, rels, media, parseTable, context) => {
+const enrichParagraphTextBoxes = (paragraph, paraXml, styles, theme, numbering, rels, media, parseTable, previews, context) => {
 	enrichTextBoxRuns({
 		content: paragraph.content,
 		xmlChildren: getChildElements(paraXml),
@@ -131,16 +134,129 @@ const enrichParagraphTextBoxes = (paragraph, paraXml, styles, theme, numbering,
 		rels,
 		media,
 		parseTable,
+		previews,
 		context
 	});
+	liftGroupTextBoxes(paragraph.content, {
+		styles,
+		theme,
+		numbering,
+		rels,
+		media,
+		parseTable,
+		previews
+	});
 	paragraph.content = consolidateParagraphContent(paragraph.content);
 };
+/** Previews whose text boxes were already lifted, so a second pass adds none. */
+const liftedGroupPreviews = /* @__PURE__ */ new WeakSet();
+/**
+* Lift the text boxes out of every group preview in the paragraph, each into
+* a text box shape right after the group's drawing in the same run.
+*
+* The group keeps drawing its other children; a lifted text box is laid out
+* and painted like any anchored text box, placed at the group's offset plus
+* the child's frame, and it records where in the group it came from so the
+* writer can put its text back there.
+*/
+const liftGroupTextBoxes = (content, parsers) => {
+	for (const item of content) switch (item.type) {
+		case "run":
+			liftGroupTextBoxesInRun(item, parsers);
+			break;
+		case "hyperlink":
+			liftGroupTextBoxes(item.children, parsers);
+			break;
+		case "inlineSdt":
+		case "insertion":
+		case "deletion":
+		case "moveFrom":
+		case "moveTo":
+			liftGroupTextBoxes(item.content, parsers);
+			break;
+		default: break;
+	}
+};
+const liftGroupTextBoxesInRun = (run, parsers) => {
+	if (!run.content.some((item) => item.type === "drawing")) return;
+	const lifted = [];
+	for (const item of run.content) {
+		lifted.push(item);
+		if (item.type !== "drawing" || item.rawXml === void 0) continue;
+		const frames = groupPreviewTextBoxes(item.image);
+		if (!frames || liftedGroupPreviews.has(item.image)) continue;
+		liftedGroupPreviews.add(item.image);
+		const group = groupXmlFingerprint(item.rawXml);
+		for (const frame of frames) {
+			const shape = groupTextBoxShape(item.image, frame, group, parsers);
+			if (shape) lifted.push({
+				type: "shape",
+				shape
+			});
+		}
+	}
+	run.content = lifted;
+};
+const groupTextBoxShape = (groupImage, frame, group, { styles, theme, numbering, rels, media, parseTable, previews }) => {
+	const position = groupImage.position;
+	const horizontalOffset = position?.horizontal.posOffset;
+	const verticalOffset = position?.vertical.posOffset;
+	if (!position || horizontalOffset === void 0 || verticalOffset === void 0) return;
+	const size = {
+		width: Math.round(frame.width),
+		height: Math.round(frame.height)
+	};
+	const textBox = parseTextBoxFromShape(frame.wsp, size);
+	const content = parseTextBoxContent(findChildByLocalName(findChildByLocalName(frame.wsp, "txbx"), "txbxContent"), parseParagraph, parseTable, styles, theme, numbering, rels, media, previews);
+	const transform = frame.rotation !== 0 || frame.flipH || frame.flipV ? {
+		...frame.rotation !== 0 ? { rotation: frame.rotation } : {},
+		...frame.flipH ? { flipH: true } : {},
+		...frame.flipV ? { flipV: true } : {}
+	} : void 0;
+	const shape = {
+		type: "shape",
+		shapeType: "textBox",
+		size,
+		...textBox?.name !== void 0 ? { name: textBox.name } : {},
+		...textBox?.alt !== void 0 ? { alt: textBox.alt } : {},
+		...textBox?.title !== void 0 ? { title: textBox.title } : {},
+		position: {
+			horizontal: {
+				relativeTo: position.horizontal.relativeTo,
+				posOffset: Math.round(horizontalOffset + frame.x)
+			},
+			vertical: {
+				relativeTo: position.vertical.relativeTo,
+				posOffset: Math.round(verticalOffset + frame.y)
+			}
+		},
+		wrap: { type: groupImage.wrap.type === "behind" ? "behind" : "inFront" },
+		...groupImage.anchor !== void 0 ? { anchor: { ...groupImage.anchor } } : {},
+		...textBox?.fill !== void 0 ? { fill: textBox.fill } : {},
+		...textBox?.outline !== void 0 ? { outline: textBox.outline } : {},
+		...transform !== void 0 ? { transform } : {},
+		textBody: {
+			content,
+			...textBox?.autoFit !== void 0 ? { autoFit: textBox.autoFit } : {},
+			...textBox?.textWrap !== void 0 ? { textWrap: textBox.textWrap } : {},
+			...textBox?.verticalAlign !== void 0 ? { anchor: textBox.verticalAlign } : {},
+			...textBox?.margins !== void 0 ? { margins: textBox.margins } : {}
+		},
+		groupChild: {
+			path: frame.path,
+			group,
+			content: groupTextContentFingerprint(content)
+		}
+	};
+	if (textBox?.id) shape.id = textBox.id;
+	return shape;
+};
 const trackedChangeTypeFromXml = (localName) => {
 	if (localName === "ins") return "insertion";
 	if (localName === "del") return "deletion";
 	if (localName === "moveFrom" || localName === "moveTo") return localName;
 };
-const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rels, media, parseTable, context }) => {
+const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rels, media, parseTable, previews, context }) => {
 	let parsedIndex = 0;
 	let lastConsumedRun;
 	for (const xmlChild of xmlChildren) {
@@ -157,6 +273,7 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 			rels,
 			media,
 			parseTable,
+			previews,
 			context
 		});
 		if (localName === "sdt" && parsedContent?.type === "inlineSdt") {
@@ -179,6 +296,7 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 				rels,
 				media,
 				parseTable,
+				previews,
 				context
 			});
 			parsedIndex = lastSegmentIndex + 1;
@@ -196,13 +314,13 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 		const targetRun = parsedRun ?? (hasNonTextBoxContent ? lastConsumedRun : void 0);
 		const targetRunMatchesXml = targetRun !== void 0 && (hasNonTextBoxContent || parsedRun?.content.length === 0);
 		const fillsEmptyCarrier = targetRunMatchesXml && !hasNonTextBoxContent && parsedRun !== void 0;
-		for (const runEl of textBoxDrawings) {
+		for (const { drawing: runEl, alternateContent: source } of textBoxDrawings) {
 			const textBox = parseTextBox(runEl, context);
 			if (!textBox) continue;
 			const wsp = findDeep(runEl, "wps", "wsp");
 			if (wsp) {
 				const txbxContentEl = getTextBoxContentElement(wsp);
-				if (txbxContentEl) textBox.content = parseTextBoxContent(txbxContentEl, parseParagraph, parseTable, styles, theme, numbering, rels, media);
+				if (txbxContentEl) textBox.content = parseTextBoxContent(txbxContentEl, parseParagraph, parseTable, styles, theme, numbering, rels, media, previews);
 			}
 			const shape = {
 				type: "shape",
@@ -226,9 +344,16 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 				}
 			};
 			if (textBox.id) shape.id = textBox.id;
+			const alternateContent = source && captureShapeAlternateContent({
+				shape,
+				alternateContent: source.element,
+				branch: source.branch,
+				drawing: runEl
+			});
 			const shapeContent = {
 				type: "shape",
-				shape
+				shape,
+				...alternateContent ? { alternateContent } : {}
 			};
 			if (targetRunMatchesXml) targetRun.content.push(shapeContent);
 			else {
@@ -242,7 +367,15 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 			}
 		}
 		for (const pictEl of vmlTextBoxes) {
-			const shape = parseVmlTextBoxShape(pictEl, styles, theme, numbering, rels, media, parseTable);
+			const shape = parseVmlTextBoxShape(pictEl, {
+				styles,
+				theme,
+				numbering,
+				rels,
+				media,
+				parseTable,
+				previews
+			});
 			if (!shape) continue;
 			const shapeContent = {
 				type: "shape",
@@ -265,7 +398,7 @@ const enrichTextBoxRuns = ({ content, xmlChildren, styles, theme, numbering, rel
 		}
 	}
 };
-const parseVmlTextBoxShape = (pictEl, styles, theme, numbering, rels, media, parseTable) => {
+const parseVmlTextBoxShape = (pictEl, { styles, theme, numbering, rels, media, parseTable, previews }) => {
 	const shapeEl = findDeep(pictEl, "v", "shape");
 	const textBoxEl = shapeEl ? findDeep(shapeEl, "v", "textbox") : null;
 	const contentEl = textBoxEl ? findDeep(textBoxEl, "w", "txbxContent") : null;
@@ -292,7 +425,7 @@ const parseVmlTextBoxShape = (pictEl, styles, theme, numbering, rels, media, par
 		...fill === void 0 ? {} : { fill },
 		wrap: { type: vmlWrapType(positioned, zIndex) },
 		textBody: {
-			content: parseTextBoxContent(contentEl, parseParagraph, parseTable, styles, theme, numbering, rels, media),
+			content: parseTextBoxContent(contentEl, parseParagraph, parseTable, styles, theme, numbering, rels, media, previews),
 			...margins === void 0 ? {} : { margins },
 			...anchor === void 0 ? {} : { anchor }
 		}

package/dist/docx/parser.d.ts

@@ -1,6 +1,7 @@
 import { document_d_exports } from "../types/document.js";
 import { DocxInput } from "../utils/docxInput.js";
 import { DocxUnzipOptions } from "./unzip.js";
+import { PreviewBudgetOverrides } from "./previewBudget.js";
 //#region src/docx/parser.d.ts
 /**
  * Progress callback for tracking parsing stages
@@ -45,6 +46,11 @@ type ParseOptions = {
  * @throws {Error} if parsing fails
  */
 declare function parseDocx(input: DocxInput, options?: ParseOptions): Promise<document_d_exports.Document>;
+/**
+ * {@link parseDocx}, charging the package's previews against `previewBudget`
+ * in place of the default per-kind allowances.
+ */
+declare function parseDocxWithPreviewBudget(input: DocxInput, options: ParseOptions, previewBudget: PreviewBudgetOverrides): Promise<document_d_exports.Document>;
 declare const DocxParseError_base: import("better-result").TaggedErrorClass<"DocxParseError">;
 /** DOCX parsing failure: malformed package, unsupported feature, or
  *  upstream parser exception. Wraps the original cause for diagnostics. */
@@ -80,4 +86,4 @@ declare function getDocxSummary(buffer: ArrayBuffer): Promise<{
   variableCount: number;
 }>;
 //#endregion
-export { DocxParseError, MediaResolver, ParseOptions, ProgressCallback, fullParseDocx, getDocxSummary, getDocxVariables, parseDocx, quickParseDocx };
+export { DocxParseError, MediaResolver, ParseOptions, ProgressCallback, fullParseDocx, getDocxSummary, getDocxVariables, parseDocx, parseDocxWithPreviewBudget, quickParseDocx };