@stll/folio-core 0.54.2 → 0.56.0

package/dist/docx/storyPlainText.d.ts

@@ -0,0 +1,20 @@
+import { document_d_exports } from "../types/document.js";
+//#region src/docx/storyPlainText.d.ts
+/** Read the document body with pending revisions accepted. */
+declare const getDocumentText: (body: document_d_exports.DocumentBody) => string;
+/** Approximate word count of the accepted body text. */
+declare const getWordCount: (body: document_d_exports.DocumentBody) => number;
+/** Character count of the accepted body text. */
+declare const getCharacterCount: (body: document_d_exports.DocumentBody) => number;
+/** Read an accepted table, including nested blocks and tracked structure. */
+declare const getTableText: (table: document_d_exports.Table) => string;
+/** Read accepted text-box content for search and indexing. */
+declare const getTextBoxText: (textBox: document_d_exports.TextBox) => string;
+/** Read an accepted header or footer through the same walk as body and notes. */
+declare const getHeaderFooterText: (headerFooter: document_d_exports.HeaderFooter) => string;
+/** Read accepted footnote content, including nested blocks. */
+declare const getFootnoteText: (footnote: document_d_exports.Footnote) => string;
+/** Read accepted endnote content, including nested blocks. */
+declare const getEndnoteText: (endnote: document_d_exports.Endnote) => string;
+//#endregion
+export { getCharacterCount, getDocumentText, getEndnoteText, getFootnoteText, getHeaderFooterText, getTableText, getTextBoxText, getWordCount };

package/dist/docx/storyPlainText.js

@@ -0,0 +1,23 @@
+import { blockPlainText } from "./blockPlainText.js";
+//#region src/docx/storyPlainText.ts
+/** Read the document body with pending revisions accepted. */
+const getDocumentText = (body) => blockPlainText(body.content);
+/** Approximate word count of the accepted body text. */
+const getWordCount = (body) => {
+	const words = getDocumentText(body).trim().split(/\s+/u);
+	return words.length > 0 && words[0] !== "" ? words.length : 0;
+};
+/** Character count of the accepted body text. */
+const getCharacterCount = (body) => getDocumentText(body).length;
+/** Read an accepted table, including nested blocks and tracked structure. */
+const getTableText = (table) => blockPlainText([table]);
+/** Read accepted text-box content for search and indexing. */
+const getTextBoxText = (textBox) => blockPlainText(textBox.content);
+/** Read an accepted header or footer through the same walk as body and notes. */
+const getHeaderFooterText = (headerFooter) => blockPlainText(headerFooter.content);
+/** Read accepted footnote content, including nested blocks. */
+const getFootnoteText = (footnote) => blockPlainText(footnote.content);
+/** Read accepted endnote content, including nested blocks. */
+const getEndnoteText = (endnote) => blockPlainText(endnote.content);
+//#endregion
+export { getCharacterCount, getDocumentText, getEndnoteText, getFootnoteText, getHeaderFooterText, getTableText, getTextBoxText, getWordCount };

package/dist/docx/streamingXmlParser.d.ts

@@ -6,6 +6,12 @@ type ParseXmlResult = {
 } | {
   status: "unsupported";
 };
+type XmlSourceRange = {
+  readonly start: number;
+  readonly end: number;
+};
+/** Exact element source ranges, available only from the opt-in source-range reader. */
+declare const getXmlSourceRange: (element: XmlElement) => XmlSourceRange | undefined;
 /**
  * Parse ordinary OOXML into Folio's existing element representation in one
  * pass. Unsupported or malformed constructs return a sentinel so callers can
@@ -17,11 +23,20 @@ type AttributeValueSpan = {
 };
 type OpenTagVisitor = (element: XmlElement, attributeValueSpans: ReadonlyMap<string, AttributeValueSpan>) => ReadonlyMap<string, string> | null;
 type OpenTagScanVisitor = (...args: Parameters<OpenTagVisitor>) => null;
+type ElementScanOptions = {
+  element: XmlElement;
+  attributeValueSpans: ReadonlyMap<string, AttributeValueSpan>;
+  nameEnd: number;
+  parent: XmlElement | undefined;
+};
+type ElementScanVisitor = (options: ElementScanOptions) => void;
 /**
  * @param inheritedNamespaceScope the bindings in scope around `xml`, for a
  *   fragment captured out of a part whose root declared them.
  */
 declare const parseStreamingXml: (xml: string, inheritedNamespaceScope?: XmlNamespaceScope) => ParseXmlResult;
+/** Retain source spans for every element without changing the parsed namespace tree. */
+declare const parseStreamingXmlWithSourceRanges: (xml: string, inheritedNamespaceScope?: XmlNamespaceScope) => ParseXmlResult;
 /** Retain the body tree while collecting identities for import repair. */
 declare const parseStreamingXmlWithIdentityVisitor: (xml: string, visitOpenTag: OpenTagScanVisitor) => ParseXmlResult;
 /** Rewrite selected decimal attribute values while preserving every other source byte. */
@@ -37,5 +52,11 @@ declare const scanStreamingXmlNumericIdAttributes: (xml: string, visitOpenTag: O
 } | {
   status: "scanned";
 };
+/** Visit namespace-resolved start tags and source spans without building a tree. */
+declare const scanStreamingXmlElements: (xml: string, visitElement: ElementScanVisitor) => {
+  status: "unsupported";
+} | {
+  status: "scanned";
+};
 //#endregion
-export { parseStreamingXml, parseStreamingXmlWithIdentityVisitor, rewriteStreamingXmlDecimalAttributes, scanStreamingXmlNumericIdAttributes };
+export { getXmlSourceRange, parseStreamingXml, parseStreamingXmlWithIdentityVisitor, parseStreamingXmlWithSourceRanges, rewriteStreamingXmlDecimalAttributes, scanStreamingXmlElements, scanStreamingXmlNumericIdAttributes };

package/dist/docx/streamingXmlParser.js

@@ -2,6 +2,9 @@ import { EMPTY_NAMESPACE_SCOPE, attachXmlNamespaceContext } from "./xmlNamespace
 import { FOLIO_XML_RESOURCE_LIMITS } from "./xmlResourceLimits.js";
 import { isOoxmlNumericIdAttributeName } from "@stll/docx-core";
 //#region src/docx/streamingXmlParser.ts
+const sourceRanges = /* @__PURE__ */ new WeakMap();
+/** Exact element source ranges, available only from the opt-in source-range reader. */
+const getXmlSourceRange = (element) => sourceRanges.get(element);
 const BUILT_IN_ENTITIES = {
 	amp: "&",
 	apos: "'",
@@ -14,6 +17,7 @@ const parseStreamingXmlInternal = (options) => {
 	const visitOpenTag = options.visitOpenTag;
 	let spanMode = "none";
 	if (visitOpenTag !== void 0) spanMode = options.mode === "tree" ? "identities" : "all";
+	if (options.visitElement !== void 0) spanMode = "all";
 	const openTagOptions = {
 		xml,
 		start: 0,
@@ -67,6 +71,10 @@ const parseStreamingXmlInternal = (options) => {
 			const name = xml.slice(open + 2, close).trim();
 			const frame = stack.pop();
 			if (!frame || frame.name !== name) return { status: "unsupported" };
+			if (frame.start !== void 0) sourceRanges.set(frame.element, {
+				start: frame.start,
+				end: close + 1
+			});
 			cursor = close + 1;
 			mergeAdjacentText = false;
 			continue;
@@ -79,6 +87,12 @@ const parseStreamingXmlInternal = (options) => {
 		const scope = parent?.namespaceScope ?? inheritedNamespaceScope;
 		if (options.mode === "attributes" && options.retainAttribute !== void 0 && parsedTag.element.attributes === void 0) parsedTag.element.namespaceScope = scope;
 		else attachXmlNamespaceContext(parsedTag.element, scope);
+		options.visitElement?.({
+			element: parsedTag.element,
+			attributeValueSpans: parsedTag.attributeValueSpans ?? /* @__PURE__ */ new Map(),
+			nameEnd: parsedTag.nameEnd,
+			parent
+		});
 		const rewritten = (options.mode === "tree" ? parsedTag.attributeValueSpans !== void 0 : options.retainAttribute === void 0 || parsedTag.element.attributes !== void 0) ? visitOpenTag?.(parsedTag.element, parsedTag.attributeValueSpans ?? /* @__PURE__ */ new Map()) : null;
 		if (rewritten) for (const [attributeName, value] of rewritten) {
 			const span = parsedTag.attributeValueSpans?.get(attributeName);
@@ -91,11 +105,18 @@ const parseStreamingXmlInternal = (options) => {
 		if (options.mode === "tree") appendElement(parent ?? root, parsedTag.element);
 		if (!parsedTag.selfClosing) {
 			if (stack.length >= FOLIO_XML_RESOURCE_LIMITS.maxDepth) return { status: "unsupported" };
-			stack.push({
+			stack.push(options.mode === "tree" && options.sourceRanges === "tracked" ? {
+				element: parsedTag.element,
+				name: parsedTag.name,
+				start: open
+			} : {
 				element: parsedTag.element,
 				name: parsedTag.name
 			});
-		}
+		} else if (options.mode === "tree" && options.sourceRanges === "tracked") sourceRanges.set(parsedTag.element, {
+			start: open,
+			end: close + 1
+		});
 		cursor = close + 1;
 		mergeAdjacentText = false;
 	}
@@ -121,6 +142,19 @@ const parseStreamingXml = (xml, inheritedNamespaceScope = EMPTY_NAMESPACE_SCOPE)
 		value: parsed.value
 	} : { status: "unsupported" };
 };
+/** Retain source spans for every element without changing the parsed namespace tree. */
+const parseStreamingXmlWithSourceRanges = (xml, inheritedNamespaceScope = EMPTY_NAMESPACE_SCOPE) => {
+	const parsed = parseStreamingXmlInternal({
+		xml,
+		mode: "tree",
+		inheritedNamespaceScope,
+		sourceRanges: "tracked"
+	});
+	return parsed.status === "parsed" ? {
+		status: "parsed",
+		value: parsed.value
+	} : { status: "unsupported" };
+};
 /** Retain the body tree while collecting identities for import repair. */
 const parseStreamingXmlWithIdentityVisitor = (xml, visitOpenTag) => {
 	const parsed = parseStreamingXmlInternal({
@@ -170,12 +204,23 @@ const scanStreamingXmlNumericIdAttributes = (xml, visitOpenTag) => {
 	});
 	return parsed.status === "parsed" ? { status: "scanned" } : parsed;
 };
+/** Visit namespace-resolved start tags and source spans without building a tree. */
+const scanStreamingXmlElements = (xml, visitElement) => {
+	const parsed = parseStreamingXmlInternal({
+		xml,
+		mode: "attributes",
+		visitOpenTag: () => null,
+		visitElement
+	});
+	return parsed.status === "parsed" ? { status: "scanned" } : parsed;
+};
 const parseOpenTag = ({ xml, start, close, spanMode, retainAttribute }) => {
 	let cursor = skipWhitespace(xml, start, close);
 	const nameStart = cursor;
 	cursor = scanName(xml, cursor, close);
 	if (cursor === nameStart) return { status: "unsupported" };
 	const name = xml.slice(nameStart, cursor);
+	const nameEnd = cursor;
 	if (isUnsafePropertyName(name)) return { status: "unsupported" };
 	let attributes;
 	let attributeValueSpans;
@@ -232,6 +277,7 @@ const parseOpenTag = ({ xml, start, close, spanMode, retainAttribute }) => {
 		status: "parsed",
 		element,
 		name,
+		nameEnd,
 		selfClosing,
 		attributeValueSpans
 	};
@@ -358,4 +404,4 @@ const isUnsafePropertyName = (name) => {
 	}
 };
 //#endregion
-export { parseStreamingXml, parseStreamingXmlWithIdentityVisitor, rewriteStreamingXmlDecimalAttributes, scanStreamingXmlNumericIdAttributes };
+export { getXmlSourceRange, parseStreamingXml, parseStreamingXmlWithIdentityVisitor, parseStreamingXmlWithSourceRanges, rewriteStreamingXmlDecimalAttributes, scanStreamingXmlElements, scanStreamingXmlNumericIdAttributes };

package/dist/docx/structuralXmlPatch.js

@@ -1,8 +1,8 @@
 import { canonicalJson } from "../utils/canonicalJson.js";
 import { parseDocumentBody } from "./documentParser.js";
 import { paraIdAttribute } from "./paraIdAttribute.js";
-import { spliceXml } from "./selectiveXmlPatch.js";
-import { readRootNamespaceBindings, serializePartElement } from "./serializer/partNamespaces.js";
+import { spliceSupportsRootNamespaces, spliceXml } from "./selectiveXmlPatch.js";
+import { materializeReplayFragment } from "./storyBlockReplay.js";
 import { NAMESPACES, WORDPROCESSINGML_NAMESPACE_URIS, getChildElements, getLocalName, getNamespaceUri, parseXmlDocument } from "./xmlParser.js";
 //#region src/docx/structuralXmlPatch.ts
 /** Conservative body-paragraph splices; source offsets never come from reserialization. */
@@ -111,21 +111,6 @@ const paragraphIds = (root) => {
 	}
 	return ids;
 };
-/** Bind fragment prefixes locally: the source root may use entirely different aliases. */
-const paragraphFragment = ({ xml, block, bindings }) => {
-	const fragment = xml.slice(block.start, block.end);
-	const openEnd = fragment.indexOf(">");
-	const name = block.element.name ?? "w:p";
-	const selfClosing = fragment[openEnd - 1] === "/";
-	return serializePartElement({
-		partPath: "word/document.xml",
-		rootName: name,
-		rootAttributes: fragment.slice(name.length + 1, selfClosing ? openEnd - 1 : openEnd).trim(),
-		baselinePrefixes: [],
-		sourceBindings: bindings,
-		body: selfClosing ? "" : fragment.slice(openEnd + 1, fragment.lastIndexOf("</"))
-	});
-};
 /**
 * Insert/delete direct body paragraphs between surviving paragraph/table anchors.
 * Tables are opaque barriers: their modeled content and order must stay identical.
@@ -168,11 +153,11 @@ const buildStructuralDocumentPatch = ({ originalXml, serializedXml, changedIds }
 	const survivingBefore = [...before.keys()].filter((key) => after.has(key));
 	const survivingAfter = [...after.keys()].filter((key) => before.has(key));
 	if (canonicalJson(survivingBefore) !== canonicalJson(survivingAfter)) return null;
-	const bindings = readRootNamespaceBindings(serializedXml);
-	const fragmentFor = (block) => paragraphFragment({
-		xml: serializedXml,
-		block,
-		bindings
+	const fragmentFor = (block) => materializeReplayFragment({
+		xml: serializedXml.slice(block.start, block.end),
+		element: block.element,
+		sourceNamespace: getNamespaceUri(source.root) ?? "",
+		generatedRoot: current.root
 	});
 	const splices = [];
 	for (const [key, block] of before) if (!after.has(key)) splices.push({
@@ -186,11 +171,14 @@ const buildStructuralDocumentPatch = ({ originalXml, serializedXml, changedIds }
 		if (!original) {
 			const id = paraIdAttribute(block.element)?.toUpperCase();
 			if (!id || allSourceIds.has(id)) return null;
-			pending.push(fragmentFor(block));
+			const fragment = fragmentFor(block);
+			if (fragment === null) return null;
+			pending.push(fragment);
 			continue;
 		}
 		const id = paraIdAttribute(block.element)?.toUpperCase();
 		const replacement = id !== void 0 && changed.has(id) ? fragmentFor(block) : originalXml.slice(original.start, original.end);
+		if (replacement === null) return null;
 		if (pending.length > 0 || replacement !== originalXml.slice(original.start, original.end)) splices.push({
 			start: original.start,
 			end: original.end,
@@ -203,6 +191,13 @@ const buildStructuralDocumentPatch = ({ originalXml, serializedXml, changedIds }
 		end: source.bodyEnd,
 		newXml: pending.join("")
 	});
+	if (!spliceSupportsRootNamespaces({
+		originalXml,
+		serializedXml,
+		fragments: splices.map(({ newXml }) => newXml).filter(Boolean),
+		originalRoot: source.root,
+		serializedRoot: current.root
+	})) return null;
 	return spliceXml(originalXml, splices);
 };
 //#endregion

package/dist/docx/styleParser.d.ts

@@ -8,6 +8,10 @@ type ParsedStylesPackage = {
   styleDefinitions: document_d_exports.StyleDefinitions;
   styles: StyleMap;
 };
+/**
+ * Resolve style inheritance chain
+ */
+declare function resolveStyleInheritance(style: document_d_exports.Style, styleMap: StyleMap, visited?: Set<string>): document_d_exports.Style;
 /**
  * Parse styles.xml content
  *
@@ -46,4 +50,4 @@ declare function getDefaultCharacterStyle(styleMap: StyleMap): document_d_export
  */
 declare function getStylesByType(styleMap: StyleMap, type: document_d_exports.StyleType): document_d_exports.Style[];
 //#endregion
-export { ParsedStylesPackage, StyleMap, getDefaultCharacterStyle, getDefaultParagraphStyle, getResolvedStyle, getStylesByType, parseStyleDefinitions, parseStyles, parseStylesPackage };
+export { ParsedStylesPackage, StyleMap, getDefaultCharacterStyle, getDefaultParagraphStyle, getResolvedStyle, getStylesByType, parseStyleDefinitions, parseStyles, parseStylesPackage, resolveStyleInheritance };

package/dist/docx/styleParser.js

@@ -554,4 +554,4 @@ function getStylesByType(styleMap, type) {
 	return result;
 }
 //#endregion
-export { getDefaultCharacterStyle, getDefaultParagraphStyle, getResolvedStyle, getStylesByType, parseStyleDefinitions, parseStyles, parseStylesPackage };
+export { getDefaultCharacterStyle, getDefaultParagraphStyle, getResolvedStyle, getStylesByType, parseStyleDefinitions, parseStyles, parseStylesPackage, resolveStyleInheritance };

package/dist/docx/tableParser.d.ts

@@ -342,13 +342,6 @@ declare function isCellMergeStart(cell: document_d_exports.TableCell): boolean;
  * @returns true if cell spans multiple columns
  */
 declare function isCellHorizontallyMerged(cell: document_d_exports.TableCell): boolean;
-/**
- * Get the plain text content of a table
- *
- * @param table - The table to extract text from
- * @returns Plain text content
- */
-declare function getTableText(table: document_d_exports.Table): string;
 /**
  * Check if table has header row
  *
@@ -371,4 +364,4 @@ declare function getHeaderRows(table: document_d_exports.Table): document_d_expo
  */
 declare function isFloatingTable(table: document_d_exports.Table): boolean;
 //#endregion
-export { CELL_CONTENT_HANDLERS, MAX_TABLE_COLUMNS, ROW_CONTENT_HANDLERS, TABLE_CONTENT_HANDLERS, getHeaderRows, getTableColumnCount, getTableRowCount, getTableText, hasHeaderRow, isCellHorizontallyMerged, isCellMergeContinuation, isCellMergeStart, isFloatingTable, isTableCellMergeRevisionContinuation, isTableCellMergeRevisionValue, parseCellMargins, parseConditionalFormatStyle, parseFloatingTableProperties, parseTable, parseTableBorders, parseTableCell, parseTableCellProperties, parseTableGrid, parseTableLook, parseTableMeasurement, parseTableProperties, parseTablePropertyExceptions, parseTableRow, parseTableRowProperties };
+export { CELL_CONTENT_HANDLERS, MAX_TABLE_COLUMNS, ROW_CONTENT_HANDLERS, TABLE_CONTENT_HANDLERS, getHeaderRows, getTableColumnCount, getTableRowCount, hasHeaderRow, isCellHorizontallyMerged, isCellMergeContinuation, isCellMergeStart, isFloatingTable, isTableCellMergeRevisionContinuation, isTableCellMergeRevisionValue, parseCellMargins, parseConditionalFormatStyle, parseFloatingTableProperties, parseTable, parseTableBorders, parseTableCell, parseTableCellProperties, parseTableGrid, parseTableLook, parseTableMeasurement, parseTableProperties, parseTablePropertyExceptions, parseTableRow, parseTableRowProperties };

package/dist/docx/tableParser.js

@@ -1406,33 +1406,6 @@ function isCellHorizontallyMerged(cell) {
 	return (cell.formatting?.gridSpan ?? 1) > 1;
 }
 /**
-* Get the plain text content of a table
-*
-* @param table - The table to extract text from
-* @returns Plain text content
-*/
-function getTableText(table) {
-	const rows = [];
-	for (const row of table.rows) {
-		const cells = [];
-		for (const cell of row.cells) {
-			const cellText = cell.content.filter((c) => c.type === "paragraph").map((p) => getParagraphText(p)).join("\n");
-			cells.push(cellText);
-		}
-		rows.push(cells.join("	"));
-	}
-	return rows.join("\n");
-}
-/**
-* Helper to get paragraph text (simplified)
-*/
-function getParagraphText(para) {
-	return para.content.filter((c) => "content" in c).flatMap((run) => {
-		if (!("content" in run) || !Array.isArray(run.content)) return [];
-		return run.content.filter((c) => typeof c === "object" && c !== null && "type" in c && c.type === "text" && "text" in c).map((c) => c.text);
-	}).join("");
-}
-/**
 * Check if table has header row
 *
 * @param table - The table to check
@@ -1461,4 +1434,4 @@ function isFloatingTable(table) {
 	return table.formatting?.floating !== void 0;
 }
 //#endregion
-export { CELL_CONTENT_HANDLERS, MAX_TABLE_COLUMNS, ROW_CONTENT_HANDLERS, TABLE_CONTENT_HANDLERS, getHeaderRows, getTableColumnCount, getTableRowCount, getTableText, hasHeaderRow, isCellHorizontallyMerged, isCellMergeContinuation, isCellMergeStart, isFloatingTable, isTableCellMergeRevisionContinuation, isTableCellMergeRevisionValue, parseCellMargins, parseConditionalFormatStyle, parseFloatingTableProperties, parseTable, parseTableBorders, parseTableCell, parseTableCellProperties, parseTableGrid, parseTableLook, parseTableMeasurement, parseTableProperties, parseTablePropertyExceptions, parseTableRow, parseTableRowProperties };
+export { CELL_CONTENT_HANDLERS, MAX_TABLE_COLUMNS, ROW_CONTENT_HANDLERS, TABLE_CONTENT_HANDLERS, getHeaderRows, getTableColumnCount, getTableRowCount, hasHeaderRow, isCellHorizontallyMerged, isCellMergeContinuation, isCellMergeStart, isFloatingTable, isTableCellMergeRevisionContinuation, isTableCellMergeRevisionValue, parseCellMargins, parseConditionalFormatStyle, parseFloatingTableProperties, parseTable, parseTableBorders, parseTableCell, parseTableCellProperties, parseTableGrid, parseTableLook, parseTableMeasurement, parseTableProperties, parseTablePropertyExceptions, parseTableRow, parseTableRowProperties };

package/dist/docx/textBoxParser.d.ts

@@ -103,10 +103,6 @@ declare function hasTextBoxOutline(textBox: document_d_exports.TextBox): boolean
  * Check if text box has content
  */
 declare function hasTextBoxContent(textBox: document_d_exports.TextBox): boolean;
-/**
- * Get plain text from text box (helper for search/indexing)
- */
-declare function getTextBoxText(textBox: document_d_exports.TextBox): string;
 /**
  * Resolve fill color to CSS color string
  */
@@ -151,4 +147,4 @@ type ScanRunForTextBoxDrawingsOptions = {
 };
 declare const scanRunForTextBoxDrawings: ({ xmlRun, claimedByRunParser }: ScanRunForTextBoxDrawingsOptions) => TextBoxRunScan;
 //#endregion
-export { ParagraphParserFn, ScanRunForTextBoxDrawingsOptions, TableParserFn, TextBoxDrawing, TextBoxRunScan, extractTextBoxContentElements, getTextBoxContentElement, getTextBoxDimensionsPx, getTextBoxHeightPx, getTextBoxMarginsPx, getTextBoxOutlineWidthPx, getTextBoxText, getTextBoxWidthPx, hasTextBoxContent, hasTextBoxFill, hasTextBoxOutline, isFloatingTextBox, isShapeTextBox, isTextBoxDrawing, parseTextBox, parseTextBoxContent, parseTextBoxFromShape, resolveTextBoxFillColor, resolveTextBoxOutlineColor, scanRunForTextBoxDrawings };
+export { ParagraphParserFn, ScanRunForTextBoxDrawingsOptions, TableParserFn, TextBoxDrawing, TextBoxRunScan, extractTextBoxContentElements, getTextBoxContentElement, getTextBoxDimensionsPx, getTextBoxHeightPx, getTextBoxMarginsPx, getTextBoxOutlineWidthPx, getTextBoxWidthPx, hasTextBoxContent, hasTextBoxFill, hasTextBoxOutline, isFloatingTextBox, isShapeTextBox, isTextBoxDrawing, parseTextBox, parseTextBoxContent, parseTextBoxFromShape, resolveTextBoxFillColor, resolveTextBoxOutlineColor, scanRunForTextBoxDrawings };

package/dist/docx/textBoxParser.js

@@ -285,25 +285,6 @@ function hasTextBoxContent(textBox) {
 	return textBox.content.length > 0;
 }
 /**
-* Get plain text from text box (helper for search/indexing)
-*/
-function getTextBoxText(textBox) {
-	return textBox.content.map(getTextBoxBlockText).join("\n");
-}
-const getTextBoxBlockText = (block) => {
-	if (block.type === "paragraph") {
-		const runTexts = [];
-		for (const item of block.content) {
-			if (item.type !== "run") continue;
-			for (const content of item.content) if (content.type === "text") runTexts.push(content.text);
-		}
-		return runTexts.join("");
-	}
-	if (block.type === "table") return block.rows.map((row) => row.cells.map((cell) => cell.content.map(getTextBoxBlockText).join("\n")).join("	")).join("\n");
-	if (block.type !== "blockSdt" && block.type !== "blockCustomXml") return "";
-	return block.content.map(getTextBoxBlockText).join("\n");
-};
-/**
 * Resolve fill color to CSS color string
 */
 function resolveTextBoxFillColor(textBox) {
@@ -380,4 +361,4 @@ const scanRunForTextBoxDrawings = ({ xmlRun, claimedByRunParser }) => {
 	};
 };
 //#endregion
-export { extractTextBoxContentElements, getTextBoxContentElement, getTextBoxDimensionsPx, getTextBoxHeightPx, getTextBoxMarginsPx, getTextBoxOutlineWidthPx, getTextBoxText, getTextBoxWidthPx, hasTextBoxContent, hasTextBoxFill, hasTextBoxOutline, isFloatingTextBox, isShapeTextBox, isTextBoxDrawing, parseTextBox, parseTextBoxContent, parseTextBoxFromShape, resolveTextBoxFillColor, resolveTextBoxOutlineColor, scanRunForTextBoxDrawings };
+export { extractTextBoxContentElements, getTextBoxContentElement, getTextBoxDimensionsPx, getTextBoxHeightPx, getTextBoxMarginsPx, getTextBoxOutlineWidthPx, getTextBoxWidthPx, hasTextBoxContent, hasTextBoxFill, hasTextBoxOutline, isFloatingTextBox, isShapeTextBox, isTextBoxDrawing, parseTextBox, parseTextBoxContent, parseTextBoxFromShape, resolveTextBoxFillColor, resolveTextBoxOutlineColor, scanRunForTextBoxDrawings };

package/dist/docx/unzip.js

@@ -5,6 +5,7 @@ import { openDocxBuffer } from "./encryption/openEncryptedDocx.js";
 import { RASTER_MIME_TYPES, detectRasterMimeType } from "./rasterMime.js";
 import { decodeXmlBytes } from "./xmlEncoding.js";
 import { FOLIO_XML_RESOURCE_LIMITS, assertXmlResourceLimits, createXmlPackageBudget } from "./xmlResourceLimits.js";
+import { writeZipPart } from "@stll/docx-core/zip";
 import JSZip from "jszip";
 //#region src/docx/unzip.ts
 /**
@@ -282,7 +283,11 @@ const replaceRawDocxXmlParts = async (content, parts) => {
 			lowerPath: path.toLowerCase(),
 			content: xmlContent
 		});
-		content.originalZip.file(path, xmlContent);
+		writeZipPart({
+			zip: content.originalZip,
+			path,
+			data: xmlContent
+		});
 		identitiesChanged = true;
 	}
 	if (identitiesChanged) content.originalBuffer = await content.originalZip.generateAsync({

package/dist/docx/verbatimCapture.d.ts

@@ -1,3 +1,4 @@
+import { document_d_exports } from "../types/document.js";
 import { XmlElement, XmlNamespaceScope } from "./xmlParser.js";
 //#region src/docx/verbatimCapture.d.ts
 declare const UntranslatableStrictNamespaceError_base: import("better-result").TaggedErrorClass<"UntranslatableStrictNamespaceError">;
@@ -29,5 +30,6 @@ declare const isSafeCapturedXmlDocument: (xml: string) => boolean;
  * must therefore depend on nothing but the element they are given.
  */
 declare const createCapturedXmlSanitizer: (options: SanitizeCapturedXmlOptions) => ((xml: string | undefined) => string | null);
+declare const captureSourceProfileXml: (element: XmlElement, conformance: document_d_exports.DocxConformanceClass | undefined) => string;
 //#endregion
-export { UntranslatableStrictNamespaceError, captureVerbatimXml, createCapturedXmlSanitizer, isSafeCapturedXmlDocument };
+export { UntranslatableStrictNamespaceError, captureSourceProfileXml, captureVerbatimXml, createCapturedXmlSanitizer, isSafeCapturedXmlDocument };

package/dist/docx/verbatimCapture.js

@@ -1,10 +1,11 @@
 import { TRANSITIONAL_NAMESPACE_BY_STRICT_URI } from "./strictValueEncodings.gen.js";
 import { NUMBERS_PER_PERCENT, isStrictNamespaceUri, percentageSpelling, toTransitionalNamespaceUri, transitionalSlotEncoding } from "./transitionalSpelling.js";
 import { roundHalfAwayFromZero, universalMeasureAs } from "./universalMeasure.js";
-import { NAMESPACES, OOXML_NAMESPACE_SCOPE, cloneElement, elementToXml, getChildElements, getLocalName, getNamespacePrefix, parseXml } from "./xmlParser.js";
+import { NAMESPACES, OOXML_NAMESPACE_SCOPE, WORDPROCESSINGML_NAMESPACE_URIS, cloneElement, elementToXml, getAttributeByNamespaceUri, getChildElements, getLocalName, getNamespacePrefix, parseXml, resolveAttributeNamespaceUri } from "./xmlParser.js";
 import { assertXmlResourceLimits } from "./xmlResourceLimits.js";
 import { getDocxXmlSafetyIssue } from "./xmlSafety.js";
-import { TaggedError } from "better-result";
+import { TaggedError, panic } from "better-result";
+import { DOCX_CONFORMANCE_CLASSES } from "@stll/docx-core/model";
 //#region src/docx/verbatimCapture.ts
 /**
 * The one way to turn parsed markup into a string a rebuilt part replays.
@@ -75,14 +76,22 @@ const originOf = (element, inherited) => {
 	});
 	return TRANSITIONAL_URIS.has(uri) ? "transitional" : inherited;
 };
-const convertAttributes = (element, namespaceUri, localName, convertValues) => {
+/** A same-local-name extension attribute is not a slot in the element's vocabulary. */
+const attributeSlotEncoding = (element, name) => {
+	if (getNamespacePrefix(name) !== null) {
+		const attributeUri = resolveAttributeNamespaceUri(element, name);
+		if (attributeUri === void 0 || element.namespaceUri === void 0 || toTransitionalNamespaceUri(attributeUri) !== toTransitionalNamespaceUri(element.namespaceUri)) return void 0;
+	}
+	return transitionalSlotEncoding(element.namespaceUri, getLocalName(element.name), getLocalName(name));
+};
+const convertAttributes = (element, namespaceUri, convertValues) => {
 	const attributes = element.attributes ?? {};
 	let converted;
 	for (const [name, raw] of Object.entries(attributes)) {
 		if (typeof raw !== "string") continue;
 		let next = raw;
 		if (name === "xmlns" || name.startsWith("xmlns:")) next = toTransitionalNamespaceUri(raw);
-		else if (convertValues) next = transitionalValue(raw, namespaceUri, transitionalSlotEncoding(element.namespaceUri, localName, getLocalName(name)));
+		else if (convertValues) next = transitionalValue(raw, namespaceUri, attributeSlotEncoding(element, name));
 		if (next === raw) continue;
 		converted ??= { ...attributes };
 		converted[name] = next;
@@ -99,7 +108,7 @@ const toTransitional = (element, inherited) => {
 	const namespaceUri = element.namespaceUri === void 0 ? "" : toTransitionalNamespaceUri(element.namespaceUri);
 	const localName = getLocalName(element.name);
 	const textEncoding = transitionalSlotEncoding(element.namespaceUri, localName);
-	const attributes = element.attributes ? convertAttributes(element, namespaceUri, localName, origin === "strict") : void 0;
+	const attributes = element.attributes ? convertAttributes(element, namespaceUri, origin === "strict") : void 0;
 	let elements;
 	for (const [index, child] of (element.elements ?? []).entries()) {
 		const next = origin === "strict" && child.type === "text" && typeof child.text === "string" ? childWithValue(child, transitionalValue(child.text, namespaceUri, textEncoding)) : toTransitional(child, origin);
@@ -363,5 +372,39 @@ const createCapturedXmlSanitizer = (options) => {
 		return sanitized;
 	};
 };
+const STRICT_BY_TRANSITIONAL = new Map([...TRANSITIONAL_NAMESPACE_BY_STRICT_URI].map(([strict, transitional]) => [transitional, strict]));
+const strictPercentage = (value, unit) => /^-?\d+(?:\.\d+)?$/u.test(value) ? `${Number(value) / NUMBERS_PER_PERCENT[unit]}%` : value;
+/** Translate generated markup by namespace/slot, never by a user-text replacement. */
+const strictElement = (element) => {
+	const localName = getLocalName(element.name);
+	for (const [name, value] of Object.entries(element.attributes ?? {})) {
+		if (typeof value !== "string") continue;
+		if (name === "xmlns" || name.startsWith("xmlns:")) {
+			element.attributes[name] = STRICT_BY_TRANSITIONAL.get(value) ?? value;
+			continue;
+		}
+		if (localName === "graphicData" && toTransitionalNamespaceUri(element.namespaceUri ?? "") === NAMESPACES.a && name === "uri") {
+			element.attributes[name] = STRICT_BY_TRANSITIONAL.get(value) ?? value;
+			continue;
+		}
+		const encoding = attributeSlotEncoding(element, name);
+		if (encoding?.percent && (!encoding.measure || getAttributeByNamespaceUri(element, WORDPROCESSINGML_NAMESPACE_URIS, "type") === "pct")) element.attributes[name] = strictPercentage(value, encoding.percent);
+	}
+	const encoding = transitionalSlotEncoding(element.namespaceUri, localName);
+	for (const child of element.elements ?? []) if (child.type === "element") strictElement(child);
+	else if (child.type === "text" && typeof child.text === "string" && encoding?.percent) child.text = strictPercentage(child.text, encoding.percent);
+};
+/** Serialize a generated, scope-bound block in the source part's profile. */
+const cloneSourceProfileTree = (element) => cloneElement(element, {
+	...element.attributes === void 0 ? {} : { attributes: { ...element.attributes } },
+	...element.elements === void 0 ? {} : { elements: element.elements.map(cloneSourceProfileTree) }
+});
+const captureSourceProfileXml = (element, conformance) => {
+	const owned = cloneSourceProfileTree(element);
+	if (conformance === DOCX_CONFORMANCE_CLASSES.STRICT) strictElement(owned);
+	const xml = elementToXml(owned);
+	if (!isSafeCapturedXmlDocument(xml)) panic("Generated source-profile markup is unsafe.");
+	return xml;
+};
 //#endregion
-export { UntranslatableStrictNamespaceError, captureVerbatimXml, createCapturedXmlSanitizer, isSafeCapturedXmlDocument };
+export { UntranslatableStrictNamespaceError, captureSourceProfileXml, captureVerbatimXml, createCapturedXmlSanitizer, isSafeCapturedXmlDocument };

package/dist/i18n/messages/ar.gen.d.ts

@@ -0,0 +1,5 @@
+import { FolioMessages } from "../messages.js";
+//#region src/i18n/messages/ar.gen.d.ts
+declare const messages: FolioMessages;
+//#endregion
+export { messages as default, messages };