@stll/folio-core 0.54.2 → 0.56.0

package/dist/docx/paragraphParser.js

@@ -5,6 +5,7 @@ import { CAPTURE, dispatchChildrenWithContext, ownedElsewhere, transitionalNames
 import { resolveDefaultParagraphStyle } from "./defaultParagraphStyle.js";
 import { parseFieldType } from "./fieldParser.js";
 import { fieldStateOf, parseFieldState } from "./fieldState.js";
+import { foldListNumberFields, isListNumberField, isListNumberGapMarker, isTabOnlyRun } from "./foldedListNumberFields.js";
 import { HYPERLINK_CHILD_HANDLERS, parseHyperlink as parseHyperlink$1, parseHyperlinkShell } from "./hyperlinkParser.js";
 import { isHyperlinkContent, isInlineSdtContent, isTrackedChangeWrapperChild } from "./inlineWrapperContent.js";
 import { inlineWrapperOf } from "./inlineWrapperParser.js";
@@ -13,10 +14,13 @@ import { counterFormatOf, markerAlignmentForLevel, markerFormattingFromLevel, nu
 import { mergeParagraphNumbering as mergeParagraphNumbering$1, paragraphNumberingReferenceId as paragraphNumberingReferenceId$1, resolveParagraphNumbering as resolveParagraphNumbering$1 } from "./numberingReference.js";
 import { paraIdAttribute, textIdAttribute } from "./paraIdAttribute.js";
 import { paraIdInRange } from "./paraIdRangeNormalization.js";
+import { EMPTY_MARK_PROPERTIES_ATTRIBUTE, restoreEmptyMarkProperties } from "./paragraphMarkPropertyPresence.js";
 import { parseParagraphProperties } from "./paragraphProperties.js";
 import { assignParagraphPropertySource } from "./paragraphPropertySource.js";
 import { preserveInlineChild, preserveRunChild, preservedInlineCapture } from "./preservedRunContent.js";
 import { standalonePreviewLedger } from "./previewBudget.js";
+import { FOLIO_REVIEW_HISTORY_NAMESPACE } from "./reviewHistoryNamespace.js";
+import { ReviewResolutionProvenanceError, parseParagraphMarkResolutionJoin, parseResolutionJoins } from "./reviewResolutionProvenance.js";
 import { consolidateParagraphContent } from "./runConsolidator.js";
 import { parseRun } from "./runParser.js";
 import { runHoldsPayload } from "./runPayload.js";
@@ -31,15 +35,10 @@ import { WORDPROCESSINGML_NAMESPACE_URIS, cloneElement, findChild, findChildByNa
 import { panic } from "better-result";
 import { PARAGRAPH_MARK_CHANGE_KINDS, REVIEW_CARRIERS } from "@stll/docx-core/model";
 //#region src/docx/paragraphParser.ts
-const FOLIO_REVIEW_HISTORY_NAMESPACES = /* @__PURE__ */ new Set(["urn:stella:folio:review-history:1"]);
+const FOLIO_REVIEW_HISTORY_NAMESPACES = /* @__PURE__ */ new Set([FOLIO_REVIEW_HISTORY_NAMESPACE]);
 /**
 * Extract plain text from a math element (recursive text content extraction)
 */
-function extractPlainText(runs) {
-	let out = "";
-	for (const run of runs) for (const c of run.content) if (c.type === "text") out += c.text;
-	return out;
-}
 function extractMathText(el) {
 	let text = "";
 	if (el.type === "text" && typeof el.text === "string") return el.text;
@@ -200,17 +199,20 @@ function parseParagraphMarkChange(pPr) {
 		const element = findChildByNamespaceUri(rPr, WORDPROCESSINGML_NAMESPACE_URIS, kind);
 		if (element) return {
 			kind,
-			info: parseTrackedChangeInfo(element)
+			info: parseTrackedChangeInfo(element),
+			...parseParagraphMarkResolutionJoin(element)
 		};
 	}
 }
-function pushTrackedChangeWrapper({ contents, type, info, content, preserveEmpty = false }) {
+function pushTrackedChangeWrapper({ contents, type, info, resolutionJoins, content, preserveEmpty = false }) {
 	if (content.length === 0 && !preserveEmpty) return;
+	const provenance = resolutionJoins === void 0 ? {} : { resolutionJoins };
 	if (type === "insertion") {
 		contents.push({
 			type: "insertion",
 			info,
-			content: [...content]
+			content: [...content],
+			...provenance
 		});
 		return;
 	}
@@ -218,7 +220,8 @@ function pushTrackedChangeWrapper({ contents, type, info, content, preserveEmpty
 		contents.push({
 			type: "deletion",
 			info,
-			content: [...content]
+			content: [...content],
+			...provenance
 		});
 		return;
 	}
@@ -226,22 +229,30 @@ function pushTrackedChangeWrapper({ contents, type, info, content, preserveEmpty
 		contents.push({
 			type: "moveFrom",
 			info,
-			content: [...content]
+			content: [...content],
+			...provenance
 		});
 		return;
 	}
 	contents.push({
 		type: "moveTo",
 		info,
-		content: [...content]
+		content: [...content],
+		...provenance
 	});
 }
-function pushTrackedChangeSegments({ contents, type, info, parsedContent }) {
+function pushTrackedChangeSegments({ contents, type, info, resolutionJoins, parsedContent }) {
+	if (resolutionJoins !== void 0 && parsedContent.some((item) => !isTrackedChangeWrapperChild(item))) throw new ReviewResolutionProvenanceError({
+		attribute: "resolutionJoins",
+		reason: "invalid",
+		message: "Tracked-resolution provenance cannot bind a fragmented imported wrapper."
+	});
 	if (!parsedContent.some(isTrackedChangeWrapperChild)) {
 		pushTrackedChangeWrapper({
 			contents,
 			type,
 			info,
+			resolutionJoins,
 			content: [],
 			preserveEmpty: true
 		});
@@ -258,6 +269,7 @@ function pushTrackedChangeSegments({ contents, type, info, parsedContent }) {
 			contents,
 			type,
 			info,
+			resolutionJoins,
 			content: segment
 		});
 		segment.length = 0;
@@ -267,6 +279,7 @@ function pushTrackedChangeSegments({ contents, type, info, parsedContent }) {
 		contents,
 		type,
 		info,
+		resolutionJoins,
 		content: segment
 	});
 }
@@ -679,6 +692,34 @@ const FIELD_INLINE_WRAPPER_HANDLERS = {
 	pPr: CAPTURE
 };
 /**
+* The `w:r` elements a complex field, or a run holding one tab, was read
+* from. A list paragraph draws its `LISTNUM` fields in the list marker and
+* keeps their markup in the content as it was read; these are where that
+* markup comes from.
+*/
+const fieldSourceElements = /* @__PURE__ */ new WeakMap();
+const tabRunSources = /* @__PURE__ */ new WeakMap();
+/**
+* Whether the content read so far ends in a list-number field whose markup is
+* kept, with nothing but range markers after it: the next tab is its tab.
+*/
+const followsListNumberField = (contents) => {
+	for (let index = contents.length - 1; index >= 0; index -= 1) {
+		const item = contents[index];
+		if (item === void 0 || isListNumberGapMarker(item)) continue;
+		return item.type === "complexField" && fieldSourceElements.has(item);
+	}
+	return false;
+};
+const sourceMarkupOf = (item) => {
+	if (item.type === "run") {
+		const element = tabRunSources.get(item);
+		return element ? captureVerbatimXml(element) : void 0;
+	}
+	const elements = fieldSourceElements.get(item);
+	return elements && elements.length > 0 ? elements.map((element) => captureVerbatimXml(element)).join("") : void 0;
+};
+/**
 * Whether the walk is currently inside some open field's code region — its
 * own `begin` has been read but not yet its `separate` — at any nesting
 * depth. `codeDepthStack` tracks every open field regardless of whether it
@@ -839,6 +880,7 @@ const PARAGRAPH_CONTENT_HANDLERS = {
 			scan.complexFieldCodeRuns = [];
 			scan.complexFieldResultRuns = [];
 			scan.complexFieldOpenRuns = [];
+			scan.complexFieldSourceRuns = [];
 			scan.complexFieldState = beginFieldState;
 			scan.complexFieldDataXml = beginFieldData ? captureVerbatimXml(beginFieldData) : void 0;
 			scan.complexFieldFallbackDisplay = beginFallbackDisplay;
@@ -849,6 +891,8 @@ const PARAGRAPH_CONTENT_HANDLERS = {
 		if (hasFieldEnd && scan.codeDepthStack.length > 0) scan.codeDepthStack.pop();
 		if (scan.inComplexField) {
 			scan.complexFieldOpenRuns.push(withOrphanFieldCharsPreserved(run, runElement));
+			if (commentReferenceId !== null || trackedContext === "deletion") scan.complexFieldSourceRuns = null;
+			else scan.complexFieldSourceRuns?.push(runElement);
 			if (instrText) scan.complexFieldInstr += instrText;
 			if ((!scan.complexFieldFormatting || Object.keys(scan.complexFieldFormatting).length === 0) && run.formatting && Object.keys(run.formatting).length > 0) scan.complexFieldFormatting = run.formatting;
 			let inFieldCodeRegion = !hasFieldBegin && !scan.afterSeparator;
@@ -897,6 +941,7 @@ const PARAGRAPH_CONTENT_HANDLERS = {
 				if (scan.complexFieldDataXml !== void 0) complexField.formFieldDataXml = scan.complexFieldDataXml;
 				if (resultIsFallback) complexField.fieldResultIsFallback = true;
 				if (scan.complexFieldFormatting) complexField.formatting = scan.complexFieldFormatting;
+				if (scan.complexFieldSourceRuns !== null && isListNumberField(complexField)) fieldSourceElements.set(complexField, scan.complexFieldSourceRuns);
 				contents.push(complexField);
 				if (commentReferenceId !== null) contents.push({
 					type: "commentReference",
@@ -926,7 +971,11 @@ const PARAGRAPH_CONTENT_HANDLERS = {
 			runElement,
 			rels,
 			media
-		})) contents.push(withOrphanFieldCharsPreserved(run, runElement));
+		})) {
+			const kept = withOrphanFieldCharsPreserved(run, runElement);
+			if (isTabOnlyRun(kept) && trackedContext !== "deletion" && followsListNumberField(contents)) tabRunSources.set(kept, runElement);
+			contents.push(kept);
+		}
 	},
 	hyperlink: (child, { contents, styles, theme, rels, media, previews, inScopeXmlns }) => {
 		contents.push(...parseHyperlinkParagraphContents(child, rels, styles, theme, media, inScopeXmlns, previews));
@@ -983,35 +1032,47 @@ const PARAGRAPH_CONTENT_HANDLERS = {
 		}
 	},
 	ins: (child, { contents, styles, theme, rels, media, previews, inScopeXmlns }) => {
+		const insInfo = parseTrackedChangeInfo(child);
+		const insContent = parseParagraphContents(child, styles, theme, null, rels, media, previews, "default", inScopeXmlns);
 		pushTrackedChangeSegments({
 			contents,
 			type: "insertion",
-			info: parseTrackedChangeInfo(child),
-			parsedContent: parseParagraphContents(child, styles, theme, null, rels, media, previews, "default", inScopeXmlns)
+			info: insInfo,
+			resolutionJoins: parseResolutionJoins(child),
+			parsedContent: insContent
 		});
 	},
 	del: (child, { contents, styles, theme, rels, media, previews, inScopeXmlns }) => {
+		const delInfo = parseTrackedChangeInfo(child);
+		const delContent = parseParagraphContents(child, styles, theme, null, rels, media, previews, "deletion", inScopeXmlns);
 		pushTrackedChangeSegments({
 			contents,
 			type: "deletion",
-			info: parseTrackedChangeInfo(child),
-			parsedContent: parseParagraphContents(child, styles, theme, null, rels, media, previews, "deletion", inScopeXmlns)
+			info: delInfo,
+			resolutionJoins: parseResolutionJoins(child),
+			parsedContent: delContent
 		});
 	},
 	moveFrom: (child, { contents, styles, theme, rels, media, previews, inScopeXmlns }) => {
+		const moveFromInfo = parseTrackedChangeInfo(child);
+		const moveFromContent = parseParagraphContents(child, styles, theme, null, rels, media, previews, "deletion", inScopeXmlns);
 		pushTrackedChangeSegments({
 			contents,
 			type: "moveFrom",
-			info: parseTrackedChangeInfo(child),
-			parsedContent: parseParagraphContents(child, styles, theme, null, rels, media, previews, "deletion", inScopeXmlns)
+			info: moveFromInfo,
+			resolutionJoins: parseResolutionJoins(child),
+			parsedContent: moveFromContent
 		});
 	},
 	moveTo: (child, { contents, styles, theme, rels, media, previews, inScopeXmlns }) => {
+		const moveToInfo = parseTrackedChangeInfo(child);
+		const moveToContent = parseParagraphContents(child, styles, theme, null, rels, media, previews, "default", inScopeXmlns);
 		pushTrackedChangeSegments({
 			contents,
 			type: "moveTo",
-			info: parseTrackedChangeInfo(child),
-			parsedContent: parseParagraphContents(child, styles, theme, null, rels, media, previews, "default", inScopeXmlns)
+			info: moveToInfo,
+			resolutionJoins: parseResolutionJoins(child),
+			parsedContent: moveToContent
 		});
 	},
 	smartTag: paragraphInlineWrapper("smartTag"),
@@ -1068,6 +1129,7 @@ function parseParagraphContents(paraElement, styles, theme, _numbering, rels, me
 		complexFieldCodeRuns: [],
 		complexFieldResultRuns: [],
 		complexFieldOpenRuns: [],
+		complexFieldSourceRuns: null,
 		afterSeparator: false,
 		complexFieldState: {},
 		complexFieldDataXml: void 0,
@@ -1116,7 +1178,8 @@ function getCommentReferenceId(runElement) {
 const PARAGRAPH_ATTRIBUTES = /* @__PURE__ */ new Set([
 	"paraId",
 	"textId",
-	"reviewCarrier"
+	"reviewCarrier",
+	EMPTY_MARK_PROPERTIES_ATTRIBUTE
 ]);
 /**
 * Parse a paragraph element (w:p)
@@ -1158,8 +1221,10 @@ function parseParagraph(node, styles, theme, numbering, rels = null, media = nul
 		const sectPr = findChild(pPr, "w", "sectPr");
 		if (sectPr) paragraph.sectionProperties = parseSectionProperties(sectPr);
 	}
+	restoreEmptyMarkProperties(paragraph, node);
 	const rawContent = parseParagraphContents(node, styles, theme, numbering, rels, media, options?.previews ?? standalonePreviewLedger(), "default", options?.rootXmlns ?? {});
-	paragraph.content = options?.runConsolidation === "deferred" ? rawContent : consolidateParagraphContent(rawContent);
+	const consolidated = (content) => options?.runConsolidation === "deferred" ? content : consolidateParagraphContent(content);
+	paragraph.content = consolidated(rawContent);
 	const paragraphFormatting = paragraph.formatting;
 	const directNumPr = paragraphFormatting?.numPr;
 	const styleNumPr = paragraphFormatting?.styleId && styles ? styles.get(paragraphFormatting.styleId)?.pPr?.numPr : void 0;
@@ -1206,41 +1271,23 @@ function parseParagraph(node, styles, theme, numbering, rels = null, media = nul
 				if (markerAlignment !== void 0) listRendering.markerAlignment = markerAlignment;
 				if (level.suffix) listRendering.markerSuffix = level.suffix;
 				if (level.pPr?.tabs?.length) listRendering.levelTabs = level.pPr.tabs;
-				let implicitChildLevelAdvances = 0;
-				const foldedMarkerSuffix = [];
-				const filteredContent = [];
-				let dropNextTab = false;
-				const isMetadataContent = (content) => content.type === "bookmarkStart" || content.type === "bookmarkEnd" || content.type === "commentRangeStart" || content.type === "commentRangeEnd" || content.type === "commentReference";
-				for (const content of paragraph.content) {
-					if (dropNextTab && isMetadataContent(content)) {
-						filteredContent.push(content);
-						continue;
-					}
-					if (dropNextTab) {
-						dropNextTab = false;
-						if (content.type === "run" && content.content.length === 1 && content.content[0]?.type === "tab") continue;
-					}
-					if (content.type === "complexField") {
-						if (content.fieldType === "LISTNUM" || content.instruction.trim().toUpperCase().startsWith("LISTNUM")) {
-							implicitChildLevelAdvances += 1;
-							const cached = extractPlainText(content.fieldResult);
-							if (cached) foldedMarkerSuffix.push(cached);
-							dropNextTab = true;
-							continue;
-						}
-					}
-					filteredContent.push(content);
-				}
-				if (foldedMarkerSuffix.length > 0) {
-					paragraph.content = filteredContent;
-					listRendering.marker = `${listRendering.marker}\t${foldedMarkerSuffix.join(" ")}`;
+				const fold = rawContent.some((item) => item.type === "complexField") ? foldListNumberFields(rawContent, sourceMarkupOf) : {
+					content: rawContent,
+					cached: [],
+					fieldCount: 0
+				};
+				const markerShowsFields = !listRendering.isBullet && level.lvlText !== "" && !level.lvlText.includes("	");
+				if (fold.cached.length > 0 && markerShowsFields) {
+					paragraph.content = consolidated(fold.content);
+					listRendering.foldedMarkerSuffix = fold.cached.join(" ");
+					listRendering.marker = `${listRendering.marker}\t${listRendering.foldedMarkerSuffix}`;
 					const nextLevel = numbering.getLevel(numId, ilvl + 1);
 					if (nextLevel?.pPr?.hangingIndent === true && nextLevel.pPr.indentFirstLine !== void 0) {
 						const hangingTwips = -nextLevel.pPr.indentFirstLine;
 						if (hangingTwips > 0) listRendering.markerSecondSlotOffsetTwips = hangingTwips;
 					}
 				}
-				if (implicitChildLevelAdvances > 0) listRendering.implicitChildLevelAdvances = implicitChildLevelAdvances;
+				if (fold.fieldCount > 0) listRendering.implicitChildLevelAdvances = fold.fieldCount;
 				paragraph.listRendering = listRendering;
 				const styleInd = styleChainInd(paragraph.formatting?.styleId ?? (styles ? resolveDefaultParagraphStyle(styles.values())?.styleId : void 0), styles);
 				const chainInd = numPrFromStyle ? styleInd : {
@@ -1304,6 +1351,7 @@ const getRunContentText = (content) => {
 		case "fieldChar":
 		case "footnoteRef":
 		case "instrText":
+		case "noteMarker":
 		case "renderedPageBreak":
 		case "shape":
 		case "symbol": return "";

package/dist/docx/paragraphPropertySource.d.ts

@@ -43,6 +43,20 @@ declare const getProseDocumentParagraphPropertySourceContract: (document: Node)
 declare const getProseParagraphPropertySourceToken: (paragraph: Node) => unknown;
 declare const copyDocumentParagraphPropertySourceContract: (target: document_d_exports.Document, source: document_d_exports.Document) => void;
 declare const getParagraphPropertySourceToken: (paragraph: document_d_exports.Paragraph) => string | undefined;
+type ParagraphSourceIdentity = {
+  type: "durable";
+  token: string;
+} | {
+  type: "capture-only";
+} | {
+  type: "authored";
+} | {
+  type: "missing-token";
+} | {
+  type: "invalid";
+};
+/** Byte captures can follow derivations; only explicit copies carry durable tokens. */
+declare const inspectParagraphSourceIdentity: (paragraph: document_d_exports.Paragraph, document: document_d_exports.Document) => ParagraphSourceIdentity;
 type ParagraphCloneOverrides = Omit<Partial<document_d_exports.Paragraph>, "type">;
 /** Clone a paragraph while deliberately retaining its parsed `w:pPr` owner. */
 declare const cloneParagraphWithPropertySource: (paragraph: document_d_exports.Paragraph, overrides: ParagraphCloneOverrides) => document_d_exports.Paragraph;
@@ -136,4 +150,4 @@ 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, ParagraphPropertySourceTransfer, 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, markParagraphPropertySourceTransfers, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, 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, ParagraphPropertySourceTransfer, 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, inspectParagraphSourceIdentity, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, markParagraphPropertySourceTransfers, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };

package/dist/docx/paragraphPropertySource.js

@@ -1,11 +1,28 @@
 import { modelParagraphFormattingEmission } from "../internal/paragraphFormattingSerialization.js";
 import { canonicalJson } from "../utils/canonicalJson.js";
+import { copyHeaderFooterBaselineHandles, copyHeaderFooterPackageBaselines } from "./headerFooterVerbatim.js";
 import { visitDocxParagraphs } from "./paragraphTraversal.js";
 import { TaggedError, panic } from "better-result";
 import { PluginKey } from "prosemirror-state";
 //#region src/docx/paragraphPropertySource.ts
 const paragraphPropertySourceEmissionFingerprint = Symbol("paragraphPropertySourceEmissionFingerprint");
-const paragraphPropertySources = /* @__PURE__ */ new WeakMap();
+const paragraphPropertyCapture = Symbol("paragraphPropertyCapture");
+const paragraphPropertyCaptures = /* @__PURE__ */ new WeakMap();
+const paragraphSourceRecordKinds = /* @__PURE__ */ new WeakMap();
+const captureForParagraph = (paragraph) => {
+	const handle = Object.getOwnPropertyDescriptor(paragraph, paragraphPropertyCapture)?.value;
+	if (!Object.hasOwn(paragraph, paragraphPropertyCapture)) return paragraphPropertyCaptures.get(paragraph);
+	return typeof handle === "object" && handle !== null ? paragraphPropertyCaptures.get(handle) : void 0;
+};
+const attachParagraphCapture = (paragraph, capture) => {
+	const handle = Object.freeze({});
+	paragraphPropertyCaptures.set(handle, capture);
+	Object.defineProperty(paragraph, paragraphPropertyCapture, {
+		value: handle,
+		enumerable: true,
+		configurable: true
+	});
+};
 const paragraphFormattingEmissionFingerprint = (emission) => canonicalJson(emission);
 const ownedParagraphPropertySource = (paragraph, xml) => {
 	return Object.freeze({
@@ -13,7 +30,6 @@ const ownedParagraphPropertySource = (paragraph, xml) => {
 		[paragraphPropertySourceEmissionFingerprint]: paragraphFormattingEmissionFingerprint(modelParagraphFormattingEmission(paragraph.formatting))
 	});
 };
-const paragraphPropertySourceOwners = /* @__PURE__ */ new WeakMap();
 const proseParagraphSourceOwners = /* @__PURE__ */ new WeakMap();
 const paragraphPropertySourceCandidates = /* @__PURE__ */ new WeakMap();
 const paragraphPropertySourceTransferIds = /* @__PURE__ */ new WeakMap();
@@ -91,19 +107,23 @@ const visitDocumentStoryParagraphs = (content, visit) => {
 	visitDocxParagraphs({ documentBody: { content } }, visit);
 };
 const assignParagraphPropertySource = (paragraph, xml) => {
-	paragraphPropertySources.set(paragraph, ownedParagraphPropertySource(paragraph, xml));
-	paragraphPropertySourceOwners.set(paragraph, paragraph);
+	attachParagraphCapture(paragraph, {
+		source: ownedParagraphPropertySource(paragraph, xml),
+		owner: paragraph
+	});
 };
-const getParagraphPropertySource = (paragraph) => paragraphPropertySources.get(paragraph);
+const getParagraphPropertySource = (paragraph) => captureForParagraph(paragraph)?.source;
 const paragraphPropertySourceMatchesEmission = (source, emission) => source[paragraphPropertySourceEmissionFingerprint] === paragraphFormattingEmissionFingerprint(emission);
 /** Copy the captured `w:pPr` without claiming the source paragraph's durable identity. */
 const copyParagraphPropertyCapture = (target, source) => {
-	const propertySource = paragraphPropertySources.get(source);
-	if (propertySource) {
-		const copiedSource = Object.freeze({ ...propertySource });
-		paragraphPropertySources.set(target, copiedSource);
-		paragraphPropertySourceOwners.set(target, paragraphPropertySourceOwners.get(source) ?? source);
-	}
+	const capture = captureForParagraph(source);
+	if (capture) attachParagraphCapture(target, capture);
+	const kind = paragraphPropertySourceTokens.has(source) || paragraphSourceRecordKinds.get(source) === "durable" ? "durable" : "capture-only";
+	if (!paragraphSourceRecordKinds.has(target)) paragraphSourceRecordKinds.set(target, kind);
+};
+const copyParagraphCaptureForTransport = (target, source) => {
+	const capture = captureForParagraph(source);
+	if (capture) paragraphPropertyCaptures.set(target, capture);
 };
 const copyParagraphPropertySource = (target, source) => {
 	copyParagraphPropertyCapture(target, source);
@@ -122,6 +142,7 @@ const assignDocumentParagraphPropertySourceContract = (document, sourceDigest) =
 	visitDocumentStoryParagraphs(document.package.document.content, (paragraph) => {
 		const token = tokenForOrdinal(contract, ordinal);
 		paragraphPropertySourceTokens.set(paragraph, token);
+		paragraphSourceRecordKinds.set(paragraph, "durable");
 		sources.set(token, paragraph);
 		ordinal += 1;
 	});
@@ -144,7 +165,7 @@ const copyDocumentParagraphPropertySources = (document) => {
 	return sources ? new Map(sources) : void 0;
 };
 const paragraphPropertySourceBelongsToDocument = (paragraph, document) => {
-	const owner = paragraphPropertySourceOwners.get(paragraph);
+	const owner = captureForParagraph(paragraph)?.owner;
 	const binding = getDocumentParagraphPropertySourceBinding(document);
 	return owner !== void 0 && binding !== void 0 && binding.sourceOwners.has(owner);
 };
@@ -158,6 +179,26 @@ const copyDocumentParagraphPropertySourceContract = (target, source) => {
 	if (binding) setDocumentParagraphPropertySourceBinding(target, binding);
 };
 const getParagraphPropertySourceToken = (paragraph) => paragraphPropertySourceTokens.get(paragraph);
+/** Byte captures can follow derivations; only explicit copies carry durable tokens. */
+const inspectParagraphSourceIdentity = (paragraph, document) => {
+	const binding = getDocumentParagraphPropertySourceBinding(document);
+	if (!binding) return { type: "invalid" };
+	const capture = captureForParagraph(paragraph);
+	if (Object.hasOwn(paragraph, paragraphPropertyCapture) && !capture) return { type: "invalid" };
+	const token = getParagraphPropertySourceToken(paragraph);
+	if (token !== void 0) {
+		if (capture && !binding.sourceOwners.has(capture.owner)) return { type: "invalid" };
+		return {
+			type: "durable",
+			token
+		};
+	}
+	const kind = paragraphSourceRecordKinds.get(paragraph);
+	if (kind === "durable" || binding.sourceOwners.has(paragraph)) return { type: "missing-token" };
+	if (!capture) return { type: "authored" };
+	if (kind === "capture-only" || binding.sourceOwners.has(capture.owner)) return { type: "capture-only" };
+	return { type: "invalid" };
+};
 /** Clone a paragraph while deliberately retaining its parsed `w:pPr` owner. */
 const cloneParagraphWithPropertySource = (paragraph, overrides) => {
 	const cloned = {
@@ -168,10 +209,14 @@ const cloneParagraphWithPropertySource = (paragraph, overrides) => {
 	return cloned;
 };
 /** Clone a paragraph whose formatting provenance is deliberately no longer applicable. */
-const cloneParagraphWithoutPropertySource = (paragraph, overrides) => ({
-	...paragraph,
-	...overrides
-});
+const cloneParagraphWithoutPropertySource = (paragraph, overrides) => {
+	const cloned = {
+		...paragraph,
+		...overrides
+	};
+	Reflect.deleteProperty(cloned, paragraphPropertyCapture);
+	return cloned;
+};
 const paragraphsIn = (document) => {
 	const paragraphs = [];
 	visitDocxParagraphs({
@@ -199,6 +244,8 @@ const cloneDocumentWithParagraphPropertySources = (document) => {
 		copyParagraphPropertySource(target, source);
 	}
 	copyDocumentParagraphPropertySourceContract(cloned, document);
+	copyHeaderFooterBaselineHandles(cloned, document);
+	copyHeaderFooterPackageBaselines(cloned, document);
 	return cloned;
 };
 const paragraphsInTableCells = (cells) => {
@@ -299,6 +346,7 @@ const tableCellRunContentTraversalByType = {
 	footnoteRef: "leaf",
 	instrText: "leaf",
 	noBreakHyphen: "leaf",
+	noteMarker: "leaf",
 	preservedXml: "leaf",
 	renderedPageBreak: "leaf",
 	shape: "shape",
@@ -549,7 +597,7 @@ const transportTableCellsWithParagraphPropertySources = (cells) => {
 	for (const [index, source] of sources.entries()) {
 		const target = targets.at(index);
 		if (!target) panic("The cloned table cells lost a paragraph owner.");
-		copyParagraphPropertyCapture(target, source);
+		copyParagraphCaptureForTransport(target, source);
 		const inspection = inspectTableCellParagraphPropertySourceBinding(source);
 		setTableCellParagraphPropertySourceBinding(target, tableCellParagraphPropertySourceBindingForTransport(source, inspection));
 	}
@@ -568,8 +616,11 @@ const restoreTableCellsWithParagraphPropertySources = (payload) => {
 		copyParagraphPropertyCapture(target, source);
 		if (!Reflect.deleteProperty(target, "_docxParagraphSourceBinding")) panic("Cannot detach paragraph-property transport identity from a table cell.");
 		switch (sourceEntry.binding.type) {
-			case "authored": break;
+			case "authored":
+				paragraphSourceRecordKinds.set(target, "capture-only");
+				break;
 			case "source":
+				paragraphSourceRecordKinds.set(target, "durable");
 				paragraphPropertySourceTokens.set(target, sourceEntry.binding.token);
 				break;
 			default: return sourceEntry.binding;
@@ -593,13 +644,14 @@ const cloneTableCellsWithParagraphPropertyCaptures = (payload) => {
 	for (const [index, sourceEntry] of sourceEntries.entries()) {
 		const target = targets.at(index);
 		if (!target) panic("The cloned table cells lost a paragraph owner.");
-		copyParagraphPropertyCapture(target, sourceEntry.paragraph);
+		copyParagraphCaptureForTransport(target, sourceEntry.paragraph);
 		setTableCellParagraphPropertySourceBinding(target, { type: "authored" });
 	}
 	return cloned;
 };
 const linkProseParagraphPropertySource = (proseParagraph, sourceParagraph) => {
-	if (paragraphPropertySources.has(sourceParagraph)) proseParagraphSourceOwners.set(proseParagraph, paragraphPropertySourceOwners.get(sourceParagraph) ?? sourceParagraph);
+	const capture = captureForParagraph(sourceParagraph);
+	if (capture) proseParagraphSourceOwners.set(proseParagraph, capture.owner);
 };
 /** Create a parsed paragraph with both its same-process owner and durable body token. */
 const createProseParagraphWithPropertySource = (nodeType, sourceParagraph, options = {}) => {
@@ -704,11 +756,14 @@ const transferProseParagraphPropertySource = (target, source, paraId) => {
 };
 const linkParagraphPropertySourceCandidate = (target, source) => {
 	const token = source.attrs[PROSE_PARAGRAPH_SOURCE_TOKEN_ATTR];
-	if (typeof token === "string") paragraphPropertySourceTokens.set(target, token);
+	if (typeof token === "string") {
+		paragraphPropertySourceTokens.set(target, token);
+		paragraphSourceRecordKinds.set(target, "durable");
+	} else paragraphSourceRecordKinds.set(target, "capture-only");
 	const sourceOwner = proseParagraphSourceOwners.get(source);
 	if (sourceOwner) paragraphPropertySourceCandidates.set(target, sourceOwner);
 };
 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, markParagraphPropertySourceTransfers, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, 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, inspectParagraphSourceIdentity, isParagraphPropertySourceToken, joinProseParagraphsWithRightPropertySource, linkParagraphPropertySourceCandidate, linkProseParagraphPropertySource, markParagraphPropertySourceTransfers, paragraphPropertySourceBelongsToDocument, paragraphPropertySourceMatchesEmission, paragraphPropertySourceTokenMatchesContract, proseParagraphAttrsWithoutPropertySource, recreateProseNodeWithDetachedParagraphPropertySource, recreateProseNodeWithParagraphPropertySource, restoreTableCellsWithParagraphPropertySources, setProseParagraphMarkupWithPropertySource, transferProseParagraphPropertySource, transportTableCellsWithParagraphPropertySources, visitDocumentStoryParagraphs, visitTableCellParagraphPropertySourceBindings };

package/dist/docx/parser.js

@@ -15,7 +15,7 @@ import { parseFontTable } from "./fontTableParser.js";
 import { parseEndnotes, parseFootnotes } from "./footnoteParser.js";
 import { parseFooter, parseHeader } from "./headerFooterParser.js";
 import { DANGLING_FOOTER_REFERENCE_WARNING, DANGLING_HEADER_REFERENCE_WARNING, assignHeaderFooterRoles, normalizeHeaderFooterReferences } from "./headerFooterReferenceNormalization.js";
-import { assignHeaderFooterVerbatimXml, canReplayHeaderFooterVerbatim, refreshHeaderFooterVerbatimFingerprint } from "./headerFooterVerbatim.js";
+import { assignHeaderFooterVerbatimXml, canReplayHeaderFooterVerbatim, captureDocumentSourceBaseline, captureHeaderFooterPackageBaselines, refreshHeaderFooterVerbatimFingerprint } from "./headerFooterVerbatim.js";
 import { extractMetafileRaster, isMetafileMimeType } from "./metafileRaster.js";
 import { renderEmfSvg } from "./metafileSvg.js";
 import { DocxModelValidationError, formatDocumentModelIssues, validateFolioDocumentModel } from "./modelValidation.js";
@@ -473,6 +473,8 @@ async function parseDocxWithPreviewBudget(input, options, previewBudget) {
 			document.warnings = formatParseWarnings(parseWarnings);
 		}
 		onProgress("Complete", 100);
+		captureHeaderFooterPackageBaselines(document);
+		if (raw.documentXml) captureDocumentSourceBaseline(document, raw.documentXml);
 		return document;
 	} catch (error) {
 		if (error instanceof DocxEncryptionError) throw error;

package/dist/docx/removeHeaderFooterParts.js

@@ -4,6 +4,7 @@ import { captureVerbatimXml } from "./verbatimCapture.js";
 import { OFFICE_RELATIONSHIP_NAMESPACE_URIS, cloneElement, getAttribute, getLocalName, getNamespaceUri, parseXmlDocument } from "./xmlParser.js";
 import { panic } from "better-result";
 import { OOXML_NS } from "@stll/docx-utils";
+import { writeZipPart } from "@stll/docx-core/zip";
 //#region src/docx/removeHeaderFooterParts.ts
 const DOCUMENT_RELS_PATH = "word/_rels/document.xml.rels";
 const officeRelationshipHasType = (type, kind) => [...OFFICE_RELATIONSHIP_NAMESPACE_URIS].some((namespace) => type === `${namespace}/${kind}`);
@@ -54,9 +55,14 @@ const removeResolvedHeaderFooterParts = async ({ document, zip, removedReference
 		}
 	}
 	const compressionOptions = { level: compressionLevel };
-	zip.file(DOCUMENT_RELS_PATH, withoutChildren(relsXml, (child) => getNamespaceUri(child) === OOXML_NS.pr && getLocalName(child.name) === "Relationship" && removedIds.has(getAttribute(child, null, "Id") ?? "")), {
-		compression: "DEFLATE",
-		compressionOptions
+	writeZipPart({
+		zip,
+		path: DOCUMENT_RELS_PATH,
+		data: withoutChildren(relsXml, (child) => getNamespaceUri(child) === OOXML_NS.pr && getLocalName(child.name) === "Relationship" && removedIds.has(getAttribute(child, null, "Id") ?? "")),
+		options: {
+			compression: "DEFLATE",
+			compressionOptions
+		}
 	});
 	if (document.package.relationships) {
 		document.package.relationships = new Map(document.package.relationships);
@@ -78,9 +84,14 @@ const removeResolvedHeaderFooterParts = async ({ document, zip, removedReference
 	const contentTypes = zip.file("[Content_Types].xml");
 	if (!contentTypes) return panic("The package has no content types");
 	const contentTypesXml = await contentTypes.async("text");
-	zip.file("[Content_Types].xml", withoutChildren(contentTypesXml, (child) => getNamespaceUri(child) === OOXML_NS.ct && getLocalName(child.name) === "Override" && candidates.has((getAttribute(child, null, "PartName") ?? "").replace(/^\//u, ""))), {
-		compression: "DEFLATE",
-		compressionOptions
+	writeZipPart({
+		zip,
+		path: "[Content_Types].xml",
+		data: withoutChildren(contentTypesXml, (child) => getNamespaceUri(child) === OOXML_NS.ct && getLocalName(child.name) === "Override" && candidates.has((getAttribute(child, null, "PartName") ?? "").replace(/^\//u, ""))),
+		options: {
+			compression: "DEFLATE",
+			compressionOptions
+		}
 	});
 };
 //#endregion

package/dist/docx/reviewHistoryNamespace.d.ts

@@ -0,0 +1,4 @@
+//#region src/docx/reviewHistoryNamespace.d.ts
+declare const FOLIO_REVIEW_HISTORY_NAMESPACE = "urn:stella:folio:review-history:1";
+//#endregion
+export { FOLIO_REVIEW_HISTORY_NAMESPACE };