@usejunior/docx-core 0.0.1 → 0.1.1

package/dist/baselines/atomizer/xmlToWmlElement.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"xmlToWmlElement.d.ts","sourceRoot":"","sources":["../../../src/baselines/atomizer/xmlToWmlElement.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAIH;;;;;GAKG;AACH,wBAAgB,gBAAgB,CAAC,GAAG,EAAE,MAAM,GAAG,OAAO,CAGrD;AAED;;;;;GAKG;AACH,wBAAgB,QAAQ,CAAC,IAAI,EAAE,OAAO,GAAG,OAAO,GAAG,SAAS,CAG3D;AAED;;;;;GAKG;AACH,wBAAgB,YAAY,CAAC,IAAI,EAAE,OAAO,GAAG,OAAO,GAAG,SAAS,CAI/D;AAED;;;;;;GAMG;AACH,wBAAgB,WAAW,CACzB,IAAI,EAAE,OAAO,EACb,OAAO,EAAE,MAAM,GACd,OAAO,GAAG,SAAS,CAIrB;AAED;;;;;;GAMG;AACH,wBAAgB,eAAe,CAC7B,IAAI,EAAE,OAAO,EACb,OAAO,EAAE,MAAM,GACd,OAAO,EAAE,CAOX;AAED;;;;;GAKG;AACH,wBAAgB,cAAc,CAAC,OAAO,EAAE,OAAO,GAAG,QAAQ,GAAG,MAAM,CAElE;AAED;;;;;GAKG;AACH,wBAAgB,YAAY,CAAC,OAAO,EAAE,OAAO,GAAG,OAAO,CAEtD;AAED;;;;;GAKG;AACH,wBAAgB,wBAAwB,CACtC,KAAK,EAAE,OAAO,EACd,OAAO,CAAC,EAAE,OAAO,GAChB,IAAI,CAEN"}

package/dist/baselines/atomizer/xmlToWmlElement.js

@@ -0,0 +1,95 @@
+/**
+ * XML Parsing and Serialization
+ *
+ * Parses document.xml into a DOM tree using @xmldom/xmldom.
+ * Replaces the former fast-xml-parser + WmlElement POJO approach.
+ */
+import { DOMParser, XMLSerializer } from '@xmldom/xmldom';
+/**
+ * Parse document.xml string into a DOM Element tree.
+ *
+ * @param xml - The raw document.xml content
+ * @returns Root element (the Document's documentElement)
+ */
+export function parseDocumentXml(xml) {
+    const doc = new DOMParser().parseFromString(xml, 'application/xml');
+    return doc.documentElement;
+}
+/**
+ * Find the w:body element in the document tree.
+ *
+ * @param root - The document root element
+ * @returns The w:body element, or undefined if not found
+ */
+export function findBody(root) {
+    const bodies = root.getElementsByTagName('w:body');
+    return bodies.length > 0 ? bodies[0] : undefined;
+}
+/**
+ * Find the w:document element in the document tree.
+ *
+ * @param root - The document root element
+ * @returns The w:document element, or undefined if not found
+ */
+export function findDocument(root) {
+    if (root.tagName === 'w:document')
+        return root;
+    const docs = root.getElementsByTagName('w:document');
+    return docs.length > 0 ? docs[0] : undefined;
+}
+/**
+ * Find an element by tag name in the tree.
+ *
+ * @param node - The node to search from
+ * @param tagName - The tag name to find
+ * @returns The found element, or undefined
+ */
+export function findElement(node, tagName) {
+    if (node.tagName === tagName)
+        return node;
+    const results = node.getElementsByTagName(tagName);
+    return results.length > 0 ? results[0] : undefined;
+}
+/**
+ * Find all elements with a specific tag name.
+ *
+ * @param node - The node to search from
+ * @param tagName - The tag name to find
+ * @returns Array of matching elements
+ */
+export function findAllElements(node, tagName) {
+    const nodeList = node.getElementsByTagName(tagName);
+    const result = [];
+    for (let i = 0; i < nodeList.length; i++) {
+        result.push(nodeList[i]);
+    }
+    return result;
+}
+/**
+ * Serialize a DOM Element back to XML string.
+ *
+ * @param element - The element to serialize
+ * @returns XML string
+ */
+export function serializeToXml(element) {
+    return new XMLSerializer().serializeToString(element);
+}
+/**
+ * Clone a DOM Element tree (deep copy).
+ *
+ * @param element - The element to clone
+ * @returns A deep copy of the element
+ */
+export function cloneElement(element) {
+    return element.cloneNode(true);
+}
+/**
+ * Backfill parent references — NO-OP for DOM Elements.
+ *
+ * DOM Elements have native parentNode/parentElement. This function exists
+ * only to ease migration; callers should remove it over time.
+ */
+export function backfillParentReferences(_node, _parent) {
+    // No-op: DOM Elements have native parentNode
+}
+//# sourceMappingURL=xmlToWmlElement.js.map

package/dist/baselines/atomizer/xmlToWmlElement.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"xmlToWmlElement.js","sourceRoot":"","sources":["../../../src/baselines/atomizer/xmlToWmlElement.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,SAAS,EAAE,aAAa,EAAE,MAAM,gBAAgB,CAAC;AAE1D;;;;;GAKG;AACH,MAAM,UAAU,gBAAgB,CAAC,GAAW;IAC1C,MAAM,GAAG,GAAG,IAAI,SAAS,EAAE,CAAC,eAAe,CAAC,GAAG,EAAE,iBAAiB,CAAC,CAAC;IACpE,OAAO,GAAG,CAAC,eAAe,CAAC;AAC7B,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,QAAQ,CAAC,IAAa;IACpC,MAAM,MAAM,GAAG,IAAI,CAAC,oBAAoB,CAAC,QAAQ,CAAC,CAAC;IACnD,OAAO,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAE,MAAM,CAAC,CAAC,CAAa,CAAC,CAAC,CAAC,SAAS,CAAC;AAChE,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,YAAY,CAAC,IAAa;IACxC,IAAI,IAAI,CAAC,OAAO,KAAK,YAAY;QAAE,OAAO,IAAI,CAAC;IAC/C,MAAM,IAAI,GAAG,IAAI,CAAC,oBAAoB,CAAC,YAAY,CAAC,CAAC;IACrD,OAAO,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAE,IAAI,CAAC,CAAC,CAAa,CAAC,CAAC,CAAC,SAAS,CAAC;AAC5D,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,WAAW,CACzB,IAAa,EACb,OAAe;IAEf,IAAI,IAAI,CAAC,OAAO,KAAK,OAAO;QAAE,OAAO,IAAI,CAAC;IAC1C,MAAM,OAAO,GAAG,IAAI,CAAC,oBAAoB,CAAC,OAAO,CAAC,CAAC;IACnD,OAAO,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAE,OAAO,CAAC,CAAC,CAAa,CAAC,CAAC,CAAC,SAAS,CAAC;AAClE,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,eAAe,CAC7B,IAAa,EACb,OAAe;IAEf,MAAM,QAAQ,GAAG,IAAI,CAAC,oBAAoB,CAAC,OAAO,CAAC,CAAC;IACpD,MAAM,MAAM,GAAc,EAAE,CAAC;IAC7B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACzC,MAAM,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC,CAAY,CAAC,CAAC;IACtC,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,cAAc,CAAC,OAA2B;IACxD,OAAO,IAAI,aAAa,EAAE,CAAC,iBAAiB,CAAC,OAAO,CAAC,CAAC;AACxD,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,YAAY,CAAC,OAAgB;IAC3C,OAAO,OAAO,CAAC,SAAS,CAAC,IAAI,CAAY,CAAC;AAC5C,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,wBAAwB,CACtC,KAAc,EACd,OAAiB;IAEjB,6CAA6C;AAC/C,CAAC"}

package/dist/baselines/diffmatch/documentBuilder.d.ts

@@ -0,0 +1,44 @@
+/**
+ * Document builder for reconstructing DOCX with track changes.
+ *
+ * Takes alignment results and produces a new document.xml with
+ * insertions and deletions marked using OOXML track changes.
+ */
+import type { AlignmentResult } from '../../shared/ooxml/types.js';
+import { type ExtendedParagraphInfo } from './xmlParser.js';
+/**
+ * Build options for document generation.
+ */
+export interface BuildOptions {
+    author: string;
+    date: Date;
+}
+/**
+ * Represents a paragraph operation in the output document.
+ */
+export interface ParagraphOperation {
+    /** Type of operation */
+    type: 'unchanged' | 'modified' | 'deleted' | 'inserted';
+    /** Generated XML for this paragraph */
+    xml: string;
+    /** Index in original document (for ordering) */
+    originalIndex?: number;
+    /** Index in revised document (for ordering) */
+    revisedIndex?: number;
+}
+/**
+ * Generate paragraph operations from alignment result.
+ *
+ * This converts the alignment result into a list of XML paragraph operations
+ * that can be used to build the output document.
+ */
+export declare function generateParagraphOperations(alignment: AlignmentResult, originalParagraphs: ExtendedParagraphInfo[], revisedParagraphs: ExtendedParagraphInfo[], options: BuildOptions): ParagraphOperation[];
+/**
+ * Build a new document.xml with track changes applied.
+ *
+ * @param originalXml - The original document.xml content
+ * @param operations - Ordered list of paragraph operations
+ * @returns The new document.xml content
+ */
+export declare function buildDocumentWithTrackChanges(originalXml: string, operations: ParagraphOperation[]): string;
+//# sourceMappingURL=documentBuilder.d.ts.map

package/dist/baselines/diffmatch/documentBuilder.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"documentBuilder.d.ts","sourceRoot":"","sources":["../../../src/baselines/diffmatch/documentBuilder.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,KAAK,EAAE,eAAe,EAAW,MAAM,6BAA6B,CAAC;AAQ5E,OAAO,EAAiB,KAAK,qBAAqB,EAAE,MAAM,gBAAgB,CAAC;AAE3E;;GAEG;AACH,MAAM,WAAW,YAAY;IAC3B,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,IAAI,CAAC;CACZ;AAED;;GAEG;AACH,MAAM,WAAW,kBAAkB;IACjC,wBAAwB;IACxB,IAAI,EAAE,WAAW,GAAG,UAAU,GAAG,SAAS,GAAG,UAAU,CAAC;IACxD,uCAAuC;IACvC,GAAG,EAAE,MAAM,CAAC;IACZ,gDAAgD;IAChD,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,+CAA+C;IAC/C,YAAY,CAAC,EAAE,MAAM,CAAC;CACvB;AAED;;;;;GAKG;AACH,wBAAgB,2BAA2B,CACzC,SAAS,EAAE,eAAe,EAC1B,kBAAkB,EAAE,qBAAqB,EAAE,EAC3C,iBAAiB,EAAE,qBAAqB,EAAE,EAC1C,OAAO,EAAE,YAAY,GACpB,kBAAkB,EAAE,CAkItB;AAED;;;;;;GAMG;AACH,wBAAgB,6BAA6B,CAC3C,WAAW,EAAE,MAAM,EACnB,UAAU,EAAE,kBAAkB,EAAE,GAC/B,MAAM,CAkBR"}

package/dist/baselines/diffmatch/documentBuilder.js

@@ -0,0 +1,227 @@
+/**
+ * Document builder for reconstructing DOCX with track changes.
+ *
+ * Takes alignment results and produces a new document.xml with
+ * insertions and deletions marked using OOXML track changes.
+ */
+import { diffRuns } from './runDiff.js';
+import { renderTrackChanges, generateDeletedParagraph, generateInsertedParagraph, wrapInParagraph, } from './trackChangesRenderer.js';
+import { extractSectPr } from './xmlParser.js';
+/**
+ * Generate paragraph operations from alignment result.
+ *
+ * This converts the alignment result into a list of XML paragraph operations
+ * that can be used to build the output document.
+ */
+export function generateParagraphOperations(alignment, originalParagraphs, revisedParagraphs, options) {
+    const operations = [];
+    const { author, date } = options;
+    // Build lookup maps for original paragraphs by index
+    const originalByIndex = new Map();
+    for (const p of originalParagraphs) {
+        originalByIndex.set(p.originalIndex, p);
+    }
+    // Build lookup maps for revised paragraphs by index
+    const revisedByIndex = new Map();
+    for (const p of revisedParagraphs) {
+        revisedByIndex.set(p.originalIndex, p);
+    }
+    // Track which original paragraphs have been matched
+    const matchedOriginalIndices = new Set();
+    const matchedRevisedIndices = new Set();
+    // Build a map from revised index to the matching original paragraph
+    const revisedToOriginal = new Map();
+    for (const match of alignment.matched) {
+        const origPara = match.original;
+        const revPara = match.revised;
+        matchedOriginalIndices.add(origPara.originalIndex);
+        matchedRevisedIndices.add(revPara.originalIndex);
+        revisedToOriginal.set(revPara.originalIndex, origPara);
+    }
+    // Build output in revised document order
+    // For each position in the revised document, we output:
+    // 1. Any deleted paragraphs that came before the matching original
+    // 2. The matched/inserted paragraph
+    let lastOriginalIndex = -1;
+    for (let revIdx = 0; revIdx < revisedParagraphs.length; revIdx++) {
+        const revPara = revisedParagraphs[revIdx];
+        if (matchedRevisedIndices.has(revIdx)) {
+            // This is a matched paragraph
+            const origPara = revisedToOriginal.get(revIdx);
+            // First, output any deleted paragraphs between last and this
+            for (let i = lastOriginalIndex + 1; i < origPara.originalIndex; i++) {
+                if (!matchedOriginalIndices.has(i)) {
+                    const deletedPara = originalByIndex.get(i);
+                    if (deletedPara) {
+                        operations.push({
+                            type: 'deleted',
+                            xml: generateDeletedParagraph(deletedPara.runs, author, date, extractPPrContent(deletedPara.pPrXml)),
+                            originalIndex: i,
+                        });
+                    }
+                }
+            }
+            // Check if paragraphs are identical or modified
+            const similarity = alignment.matched.find(m => m.revised.originalIndex === revIdx)?.similarity ?? 0;
+            if (similarity >= 0.9999) {
+                // Unchanged paragraph - use revised content as-is
+                operations.push({
+                    type: 'unchanged',
+                    xml: generateParagraphXml(revPara),
+                    originalIndex: origPara.originalIndex,
+                    revisedIndex: revIdx,
+                });
+            }
+            else {
+                // Modified paragraph - diff the runs
+                const diffResult = diffRuns(origPara.runs, revPara.runs);
+                const trackChangesContent = renderTrackChanges(diffResult.mergedRuns, {
+                    author,
+                    date,
+                });
+                operations.push({
+                    type: 'modified',
+                    xml: wrapInParagraph(trackChangesContent, extractPPrContent(revPara.pPrXml)),
+                    originalIndex: origPara.originalIndex,
+                    revisedIndex: revIdx,
+                });
+            }
+            lastOriginalIndex = origPara.originalIndex;
+        }
+        else {
+            // This is an inserted paragraph
+            operations.push({
+                type: 'inserted',
+                xml: generateInsertedParagraph(revPara.runs, author, date, extractPPrContent(revPara.pPrXml)),
+                revisedIndex: revIdx,
+            });
+        }
+    }
+    // Output any remaining deleted paragraphs at the end
+    for (let i = lastOriginalIndex + 1; i < originalParagraphs.length; i++) {
+        if (!matchedOriginalIndices.has(i)) {
+            const deletedPara = originalByIndex.get(i);
+            if (deletedPara) {
+                operations.push({
+                    type: 'deleted',
+                    xml: generateDeletedParagraph(deletedPara.runs, author, date, extractPPrContent(deletedPara.pPrXml)),
+                    originalIndex: i,
+                });
+            }
+        }
+    }
+    return operations;
+}
+/**
+ * Build a new document.xml with track changes applied.
+ *
+ * @param originalXml - The original document.xml content
+ * @param operations - Ordered list of paragraph operations
+ * @returns The new document.xml content
+ */
+export function buildDocumentWithTrackChanges(originalXml, operations) {
+    // Extract document structure
+    const { beforeBody, bodyContent, afterBody } = getDocumentParts(originalXml);
+    // Extract sectPr from body (must be preserved at end)
+    const { sectPr } = extractSectPr(bodyContent);
+    // Build new body content from operations
+    const paragraphXmls = operations.map(op => op.xml);
+    // Combine: paragraphs + sectPr (if present) + closing
+    let newBodyContent = paragraphXmls.join('\n');
+    if (sectPr) {
+        newBodyContent += '\n' + sectPr;
+    }
+    // Reconstruct document
+    return beforeBody + newBodyContent + afterBody;
+}
+/**
+ * Extract document parts for reconstruction.
+ */
+function getDocumentParts(documentXml) {
+    // Find w:body opening and closing
+    const bodyOpenMatch = documentXml.match(/<w:body[^>]*>/);
+    const bodyCloseMatch = documentXml.match(/<\/w:body>/);
+    if (!bodyOpenMatch || !bodyCloseMatch) {
+        return {
+            beforeBody: documentXml,
+            bodyContent: '',
+            afterBody: '',
+        };
+    }
+    const bodyOpenEnd = documentXml.indexOf(bodyOpenMatch[0]) + bodyOpenMatch[0].length;
+    const bodyCloseStart = documentXml.lastIndexOf('</w:body>');
+    return {
+        beforeBody: documentXml.slice(0, bodyOpenEnd),
+        bodyContent: documentXml.slice(bodyOpenEnd, bodyCloseStart),
+        afterBody: documentXml.slice(bodyCloseStart),
+    };
+}
+/**
+ * Generate XML for an unchanged paragraph.
+ */
+function generateParagraphXml(para) {
+    // Build paragraph from runs
+    const runsXml = para.runs.map(run => generateRunXml(run)).join('');
+    const pPr = para.pPrXml ? para.pPrXml : '';
+    return `<w:p>${pPr}${runsXml}</w:p>`;
+}
+/**
+ * Generate XML for a single run.
+ */
+function generateRunXml(run) {
+    const rPr = generateRunPropertiesXml(run.properties);
+    // Handle whitespace preservation
+    const needsSpace = run.text.startsWith(' ') || run.text.endsWith(' ') || run.text.includes('  ');
+    const spaceAttr = needsSpace ? ' xml:space="preserve"' : '';
+    return `<w:r>${rPr}<w:t${spaceAttr}>${escapeXml(run.text)}</w:t></w:r>`;
+}
+/**
+ * Generate run properties XML.
+ */
+function generateRunPropertiesXml(props) {
+    if (!props)
+        return '';
+    const parts = [];
+    if (props.bold)
+        parts.push('<w:b/>');
+    if (props.italic)
+        parts.push('<w:i/>');
+    if (props.underline)
+        parts.push(`<w:u w:val="${escapeXml(props.underline)}"/>`);
+    if (props.strikethrough)
+        parts.push('<w:strike/>');
+    if (props.highlight)
+        parts.push(`<w:highlight w:val="${escapeXml(props.highlight)}"/>`);
+    if (props.color)
+        parts.push(`<w:color w:val="${escapeXml(props.color)}"/>`);
+    if (props.fontSize !== undefined)
+        parts.push(`<w:sz w:val="${props.fontSize}"/>`);
+    if (props.fontFamily)
+        parts.push(`<w:rFonts w:ascii="${escapeXml(props.fontFamily)}" w:hAnsi="${escapeXml(props.fontFamily)}"/>`);
+    if (parts.length === 0)
+        return '';
+    return `<w:rPr>${parts.join('')}</w:rPr>`;
+}
+/**
+ * Extract inner pPr content from full pPr XML.
+ * Removes the outer <w:pPr> tags.
+ */
+function extractPPrContent(pPrXml) {
+    if (!pPrXml)
+        return undefined;
+    // Remove outer tags
+    const match = pPrXml.match(/<w:pPr[^>]*>([\s\S]*)<\/w:pPr>/);
+    return match ? match[1] : undefined;
+}
+/**
+ * Escape XML special characters.
+ */
+function escapeXml(text) {
+    return text
+        .replace(/&/g, '&amp;')
+        .replace(/</g, '&lt;')
+        .replace(/>/g, '&gt;')
+        .replace(/"/g, '&quot;')
+        .replace(/'/g, '&apos;');
+}
+//# sourceMappingURL=documentBuilder.js.map

package/dist/baselines/diffmatch/documentBuilder.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"documentBuilder.js","sourceRoot":"","sources":["../../../src/baselines/diffmatch/documentBuilder.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAGH,OAAO,EAAE,QAAQ,EAAE,MAAM,cAAc,CAAC;AACxC,OAAO,EACL,kBAAkB,EAClB,wBAAwB,EACxB,yBAAyB,EACzB,eAAe,GAChB,MAAM,2BAA2B,CAAC;AACnC,OAAO,EAAE,aAAa,EAA8B,MAAM,gBAAgB,CAAC;AAwB3E;;;;;GAKG;AACH,MAAM,UAAU,2BAA2B,CACzC,SAA0B,EAC1B,kBAA2C,EAC3C,iBAA0C,EAC1C,OAAqB;IAErB,MAAM,UAAU,GAAyB,EAAE,CAAC;IAC5C,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,GAAG,OAAO,CAAC;IAEjC,qDAAqD;IACrD,MAAM,eAAe,GAAG,IAAI,GAAG,EAAiC,CAAC;IACjE,KAAK,MAAM,CAAC,IAAI,kBAAkB,EAAE,CAAC;QACnC,eAAe,CAAC,GAAG,CAAC,CAAC,CAAC,aAAa,EAAE,CAAC,CAAC,CAAC;IAC1C,CAAC;IAED,oDAAoD;IACpD,MAAM,cAAc,GAAG,IAAI,GAAG,EAAiC,CAAC;IAChE,KAAK,MAAM,CAAC,IAAI,iBAAiB,EAAE,CAAC;QAClC,cAAc,CAAC,GAAG,CAAC,CAAC,CAAC,aAAa,EAAE,CAAC,CAAC,CAAC;IACzC,CAAC;IAED,oDAAoD;IACpD,MAAM,sBAAsB,GAAG,IAAI,GAAG,EAAU,CAAC;IACjD,MAAM,qBAAqB,GAAG,IAAI,GAAG,EAAU,CAAC;IAEhD,oEAAoE;IACpE,MAAM,iBAAiB,GAAG,IAAI,GAAG,EAAiC,CAAC;IAEnE,KAAK,MAAM,KAAK,IAAI,SAAS,CAAC,OAAO,EAAE,CAAC;QACtC,MAAM,QAAQ,GAAG,KAAK,CAAC,QAAiC,CAAC;QACzD,MAAM,OAAO,GAAG,KAAK,CAAC,OAAgC,CAAC;QAEvD,sBAAsB,CAAC,GAAG,CAAC,QAAQ,CAAC,aAAa,CAAC,CAAC;QACnD,qBAAqB,CAAC,GAAG,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC;QACjD,iBAAiB,CAAC,GAAG,CAAC,OAAO,CAAC,aAAa,EAAE,QAAQ,CAAC,CAAC;IACzD,CAAC;IAED,yCAAyC;IACzC,wDAAwD;IACxD,mEAAmE;IACnE,oCAAoC;IAEpC,IAAI,iBAAiB,GAAG,CAAC,CAAC,CAAC;IAE3B,KAAK,IAAI,MAAM,GAAG,CAAC,EAAE,MAAM,GAAG,iBAAiB,CAAC,MAAM,EAAE,MAAM,EAAE,EAAE,CAAC;QACjE,MAAM,OAAO,GAAG,iBAAiB,CAAC,MAAM,CAAE,CAAC;QAE3C,IAAI,qBAAqB,CAAC,GAAG,CAAC,MAAM,CAAC,EAAE,CAAC;YACtC,8BAA8B;YAC9B,MAAM,QAAQ,GAAG,iBAAiB,CAAC,GAAG,CAAC,MAAM,CAAE,CAAC;YAEhD,6DAA6D;YAC7D,KAAK,IAAI,CAAC,GAAG,iBAAiB,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,aAAa,EAAE,CAAC,EAAE,EAAE,CAAC;gBACpE,IAAI,CAAC,sBAAsB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC;oBACnC,MAAM,WAAW,GAAG,eAAe,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;oBAC3C,IAAI,WAAW,EAAE,CAAC;wBAChB,UAAU,CAAC,IAAI,CAAC;4BACd,IAAI,EAAE,SAAS;4BACf,GAAG,EAAE,wBAAwB,CAC3B,WAAW,CAAC,IAAI,EAChB,MAAM,EACN,IAAI,EACJ,iBAAiB,CAAC,WAAW,CAAC,MAAM,CAAC,CACtC;4BACD,aAAa,EAAE,CAAC;yBACjB,CAAC,CAAC;oBACL,CAAC;gBACH,CAAC;YACH,CAAC;YAED,gDAAgD;YAChD,MAAM,UAAU,GAAG,SAAS,CAAC,OAAO,CAAC,IAAI,CACvC,CAAC,CAAC,EAAE,CAAE,CAAC,CAAC,OAAiC,CAAC,aAAa,KAAK,MAAM,CACnE,EAAE,UAAU,IAAI,CAAC,CAAC;YAEnB,IAAI,UAAU,IAAI,MAAM,EAAE,CAAC;gBACzB,kDAAkD;gBAClD,UAAU,CAAC,IAAI,CAAC;oBACd,IAAI,EAAE,WAAW;oBACjB,GAAG,EAAE,oBAAoB,CAAC,OAAO,CAAC;oBAClC,aAAa,EAAE,QAAQ,CAAC,aAAa;oBACrC,YAAY,EAAE,MAAM;iBACrB,CAAC,CAAC;YACL,CAAC;iBAAM,CAAC;gBACN,qCAAqC;gBACrC,MAAM,UAAU,GAAG,QAAQ,CAAC,QAAQ,CAAC,IAAI,EAAE,OAAO,CAAC,IAAI,CAAC,CAAC;gBACzD,MAAM,mBAAmB,GAAG,kBAAkB,CAAC,UAAU,CAAC,UAAU,EAAE;oBACpE,MAAM;oBACN,IAAI;iBACL,CAAC,CAAC;gBAEH,UAAU,CAAC,IAAI,CAAC;oBACd,IAAI,EAAE,UAAU;oBAChB,GAAG,EAAE,eAAe,CAAC,mBAAmB,EAAE,iBAAiB,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC;oBAC5E,aAAa,EAAE,QAAQ,CAAC,aAAa;oBACrC,YAAY,EAAE,MAAM;iBACrB,CAAC,CAAC;YACL,CAAC;YAED,iBAAiB,GAAG,QAAQ,CAAC,aAAa,CAAC;QAC7C,CAAC;aAAM,CAAC;YACN,gCAAgC;YAChC,UAAU,CAAC,IAAI,CAAC;gBACd,IAAI,EAAE,UAAU;gBAChB,GAAG,EAAE,yBAAyB,CAC5B,OAAO,CAAC,IAAI,EACZ,MAAM,EACN,IAAI,EACJ,iBAAiB,CAAC,OAAO,CAAC,MAAM,CAAC,CAClC;gBACD,YAAY,EAAE,MAAM;aACrB,CAAC,CAAC;QACL,CAAC;IACH,CAAC;IAED,qDAAqD;IACrD,KAAK,IAAI,CAAC,GAAG,iBAAiB,GAAG,CAAC,EAAE,CAAC,GAAG,kBAAkB,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,IAAI,CAAC,sBAAsB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC;YACnC,MAAM,WAAW,GAAG,eAAe,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;YAC3C,IAAI,WAAW,EAAE,CAAC;gBAChB,UAAU,CAAC,IAAI,CAAC;oBACd,IAAI,EAAE,SAAS;oBACf,GAAG,EAAE,wBAAwB,CAC3B,WAAW,CAAC,IAAI,EAChB,MAAM,EACN,IAAI,EACJ,iBAAiB,CAAC,WAAW,CAAC,MAAM,CAAC,CACtC;oBACD,aAAa,EAAE,CAAC;iBACjB,CAAC,CAAC;YACL,CAAC;QACH,CAAC;IACH,CAAC;IAED,OAAO,UAAU,CAAC;AACpB,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,6BAA6B,CAC3C,WAAmB,EACnB,UAAgC;IAEhC,6BAA6B;IAC7B,MAAM,EAAE,UAAU,EAAE,WAAW,EAAE,SAAS,EAAE,GAAG,gBAAgB,CAAC,WAAW,CAAC,CAAC;IAE7E,sDAAsD;IACtD,MAAM,EAAE,MAAM,EAAE,GAAG,aAAa,CAAC,WAAW,CAAC,CAAC;IAE9C,yCAAyC;IACzC,MAAM,aAAa,GAAG,UAAU,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC;IAEnD,sDAAsD;IACtD,IAAI,cAAc,GAAG,aAAa,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAC9C,IAAI,MAAM,EAAE,CAAC;QACX,cAAc,IAAI,IAAI,GAAG,MAAM,CAAC;IAClC,CAAC;IAED,uBAAuB;IACvB,OAAO,UAAU,GAAG,cAAc,GAAG,SAAS,CAAC;AACjD,CAAC;AAED;;GAEG;AACH,SAAS,gBAAgB,CAAC,WAAmB;IAK3C,kCAAkC;IAClC,MAAM,aAAa,GAAG,WAAW,CAAC,KAAK,CAAC,eAAe,CAAC,CAAC;IACzD,MAAM,cAAc,GAAG,WAAW,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC;IAEvD,IAAI,CAAC,aAAa,IAAI,CAAC,cAAc,EAAE,CAAC;QACtC,OAAO;YACL,UAAU,EAAE,WAAW;YACvB,WAAW,EAAE,EAAE;YACf,SAAS,EAAE,EAAE;SACd,CAAC;IACJ,CAAC;IAED,MAAM,WAAW,GAAG,WAAW,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,GAAG,aAAa,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC;IACpF,MAAM,cAAc,GAAG,WAAW,CAAC,WAAW,CAAC,WAAW,CAAC,CAAC;IAE5D,OAAO;QACL,UAAU,EAAE,WAAW,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC;QAC7C,WAAW,EAAE,WAAW,CAAC,KAAK,CAAC,WAAW,EAAE,cAAc,CAAC;QAC3D,SAAS,EAAE,WAAW,CAAC,KAAK,CAAC,cAAc,CAAC;KAC7C,CAAC;AACJ,CAAC;AAED;;GAEG;AACH,SAAS,oBAAoB,CAAC,IAA2B;IACvD,4BAA4B;IAC5B,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC,cAAc,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;IACnE,MAAM,GAAG,GAAG,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC;IAE3C,OAAO,QAAQ,GAAG,GAAG,OAAO,QAAQ,CAAC;AACvC,CAAC;AAED;;GAEG;AACH,SAAS,cAAc,CAAC,GAAY;IAClC,MAAM,GAAG,GAAG,wBAAwB,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;IAErD,iCAAiC;IACjC,MAAM,UAAU,GAAG,GAAG,CAAC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,IAAI,GAAG,CAAC,IAAI,CAAC,QAAQ,CAAC,GAAG,CAAC,IAAI,GAAG,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC;IACjG,MAAM,SAAS,GAAG,UAAU,CAAC,CAAC,CAAC,uBAAuB,CAAC,CAAC,CAAC,EAAE,CAAC;IAE5D,OAAO,QAAQ,GAAG,OAAO,SAAS,IAAI,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,cAAc,CAAC;AAC1E,CAAC;AAED;;GAEG;AACH,SAAS,wBAAwB,CAAC,KAA2D;IAC3F,IAAI,CAAC,KAAK;QAAE,OAAO,EAAE,CAAC;IAEtB,MAAM,KAAK,GAAa,EAAE,CAAC;IAE3B,IAAI,KAAK,CAAC,IAAI;QAAE,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IACrC,IAAI,KAAK,CAAC,MAAM;QAAE,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IACvC,IAAI,KAAK,CAAC,SAAS;QAAE,KAAK,CAAC,IAAI,CAAC,eAAe,SAAS,CAAC,KAAK,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;IAChF,IAAI,KAAK,CAAC,aAAa;QAAE,KAAK,CAAC,IAAI,CAAC,aAAa,CAAC,CAAC;IACnD,IAAI,KAAK,CAAC,SAAS;QAAE,KAAK,CAAC,IAAI,CAAC,uBAAuB,SAAS,CAAC,KAAK,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;IACxF,IAAI,KAAK,CAAC,KAAK;QAAE,KAAK,CAAC,IAAI,CAAC,mBAAmB,SAAS,CAAC,KAAK,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC;IAC5E,IAAI,KAAK,CAAC,QAAQ,KAAK,SAAS;QAAE,KAAK,CAAC,IAAI,CAAC,gBAAgB,KAAK,CAAC,QAAQ,KAAK,CAAC,CAAC;IAClF,IAAI,KAAK,CAAC,UAAU;QAAE,KAAK,CAAC,IAAI,CAAC,sBAAsB,SAAS,CAAC,KAAK,CAAC,UAAU,CAAC,cAAc,SAAS,CAAC,KAAK,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC;IAElI,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,EAAE,CAAC;IAElC,OAAO,UAAU,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC;AAC5C,CAAC;AAED;;;GAGG;AACH,SAAS,iBAAiB,CAAC,MAAe;IACxC,IAAI,CAAC,MAAM;QAAE,OAAO,SAAS,CAAC;IAE9B,oBAAoB;IACpB,MAAM,KAAK,GAAG,MAAM,CAAC,KAAK,CAAC,gCAAgC,CAAC,CAAC;IAC7D,OAAO,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;AACtC,CAAC;AAED;;GAEG;AACH,SAAS,SAAS,CAAC,IAAY;IAC7B,OAAO,IAAI;SACR,OAAO,CAAC,IAAI,EAAE,OAAO,CAAC;SACtB,OAAO,CAAC,IAAI,EAAE,MAAM,CAAC;SACrB,OAAO,CAAC,IAAI,EAAE,MAAM,CAAC;SACrB,OAAO,CAAC,IAAI,EAAE,QAAQ,CAAC;SACvB,OAAO,CAAC,IAAI,EAAE,QAAQ,CAAC,CAAC;AAC7B,CAAC"}

package/dist/baselines/diffmatch/paragraphAlignment.d.ts

@@ -0,0 +1,75 @@
+/**
+ * Baseline B: Paragraph alignment using LCS (Longest Common Subsequence).
+ *
+ * Aligns paragraphs between original and revised documents to identify:
+ * - Matched paragraphs (same or similar content)
+ * - Inserted paragraphs (only in revised)
+ * - Deleted paragraphs (only in original)
+ * - Modified paragraphs (matched but with differences)
+ */
+import type { ParagraphInfo, AlignmentResult } from '../../shared/ooxml/types.js';
+/**
+ * Compute a hash for a paragraph's normalized text.
+ *
+ * Used for fast comparison during LCS.
+ */
+export declare function hashParagraph(text: string): string;
+/**
+ * Normalize paragraph text for comparison.
+ *
+ * - Trim whitespace
+ * - Collapse multiple spaces
+ * - Lowercase for case-insensitive comparison
+ */
+export declare function normalizeParagraphText(text: string): string;
+/**
+ * Compute similarity between two strings using Jaccard index on words.
+ *
+ * @returns Value between 0 (completely different) and 1 (identical)
+ */
+export declare function computeSimilarity(a: string, b: string): number;
+/**
+ * Compute the Longest Common Subsequence of two arrays using dynamic programming.
+ *
+ * @param a - First array
+ * @param b - Second array
+ * @param keyFn - Function to extract comparison key from elements
+ * @returns Array of [indexA, indexB] pairs representing the LCS
+ */
+export declare function lcs<T>(a: T[], b: T[], keyFn: (item: T) => string): Array<[number, number]>;
+/**
+ * Align paragraphs between original and revised documents.
+ *
+ * Uses hash-based LCS to find matching paragraphs, then classifies
+ * unmatched paragraphs as inserted or deleted.
+ *
+ * @param original - Paragraphs from original document
+ * @param revised - Paragraphs from revised document
+ * @param similarityThreshold - Minimum similarity to consider a match (default: 0.5)
+ */
+export declare function alignParagraphs(original: ParagraphInfo[], revised: ParagraphInfo[], similarityThreshold?: number): AlignmentResult;
+/**
+ * Classify paragraph matches into categories.
+ */
+export interface ParagraphClassification {
+    /** Identical paragraphs (similarity = 1.0) */
+    identical: Array<{
+        original: ParagraphInfo;
+        revised: ParagraphInfo;
+    }>;
+    /** Modified paragraphs (0 < similarity < 1.0) */
+    modified: Array<{
+        original: ParagraphInfo;
+        revised: ParagraphInfo;
+        similarity: number;
+    }>;
+    /** Deleted paragraphs */
+    deleted: ParagraphInfo[];
+    /** Inserted paragraphs */
+    inserted: ParagraphInfo[];
+}
+/**
+ * Classify alignment result into more granular categories.
+ */
+export declare function classifyAlignment(alignment: AlignmentResult): ParagraphClassification;
+//# sourceMappingURL=paragraphAlignment.d.ts.map

package/dist/baselines/diffmatch/paragraphAlignment.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"paragraphAlignment.d.ts","sourceRoot":"","sources":["../../../src/baselines/diffmatch/paragraphAlignment.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AAGH,OAAO,KAAK,EAAE,aAAa,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAC;AAElF;;;;GAIG;AACH,wBAAgB,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAGlD;AAED;;;;;;GAMG;AACH,wBAAgB,sBAAsB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAK3D;AAED;;;;GAIG;AACH,wBAAgB,iBAAiB,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,MAAM,GAAG,MAAM,CAe9D;AAED;;;;;;;GAOG;AACH,wBAAgB,GAAG,CAAC,CAAC,EACnB,CAAC,EAAE,CAAC,EAAE,EACN,CAAC,EAAE,CAAC,EAAE,EACN,KAAK,EAAE,CAAC,IAAI,EAAE,CAAC,KAAK,MAAM,GACzB,KAAK,CAAC,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC,CAqCzB;AAED;;;;;;;;;GASG;AACH,wBAAgB,eAAe,CAC7B,QAAQ,EAAE,aAAa,EAAE,EACzB,OAAO,EAAE,aAAa,EAAE,EACxB,mBAAmB,SAAM,GACxB,eAAe,CA6FjB;AAED;;GAEG;AACH,MAAM,WAAW,uBAAuB;IACtC,8CAA8C;IAC9C,SAAS,EAAE,KAAK,CAAC;QAAE,QAAQ,EAAE,aAAa,CAAC;QAAC,OAAO,EAAE,aAAa,CAAA;KAAE,CAAC,CAAC;IACtE,iDAAiD;IACjD,QAAQ,EAAE,KAAK,CAAC;QACd,QAAQ,EAAE,aAAa,CAAC;QACxB,OAAO,EAAE,aAAa,CAAC;QACvB,UAAU,EAAE,MAAM,CAAC;KACpB,CAAC,CAAC;IACH,yBAAyB;IACzB,OAAO,EAAE,aAAa,EAAE,CAAC;IACzB,0BAA0B;IAC1B,QAAQ,EAAE,aAAa,EAAE,CAAC;CAC3B;AAED;;GAEG;AACH,wBAAgB,iBAAiB,CAAC,SAAS,EAAE,eAAe,GAAG,uBAAuB,CAoBrF"}

package/dist/baselines/diffmatch/paragraphAlignment.js

@@ -0,0 +1,206 @@
+/**
+ * Baseline B: Paragraph alignment using LCS (Longest Common Subsequence).
+ *
+ * Aligns paragraphs between original and revised documents to identify:
+ * - Matched paragraphs (same or similar content)
+ * - Inserted paragraphs (only in revised)
+ * - Deleted paragraphs (only in original)
+ * - Modified paragraphs (matched but with differences)
+ */
+import { createHash } from 'crypto';
+/**
+ * Compute a hash for a paragraph's normalized text.
+ *
+ * Used for fast comparison during LCS.
+ */
+export function hashParagraph(text) {
+    const normalized = normalizeParagraphText(text);
+    return createHash('sha1').update(normalized).digest('hex');
+}
+/**
+ * Normalize paragraph text for comparison.
+ *
+ * - Trim whitespace
+ * - Collapse multiple spaces
+ * - Lowercase for case-insensitive comparison
+ */
+export function normalizeParagraphText(text) {
+    return text
+        .trim()
+        .replace(/\s+/g, ' ')
+        .toLowerCase();
+}
+/**
+ * Compute similarity between two strings using Jaccard index on words.
+ *
+ * @returns Value between 0 (completely different) and 1 (identical)
+ */
+export function computeSimilarity(a, b) {
+    const wordsA = new Set(normalizeParagraphText(a).split(' ').filter(w => w.length > 0));
+    const wordsB = new Set(normalizeParagraphText(b).split(' ').filter(w => w.length > 0));
+    if (wordsA.size === 0 && wordsB.size === 0) {
+        return 1; // Both empty
+    }
+    if (wordsA.size === 0 || wordsB.size === 0) {
+        return 0; // One empty
+    }
+    const intersection = new Set([...wordsA].filter(x => wordsB.has(x)));
+    const union = new Set([...wordsA, ...wordsB]);
+    return intersection.size / union.size;
+}
+/**
+ * Compute the Longest Common Subsequence of two arrays using dynamic programming.
+ *
+ * @param a - First array
+ * @param b - Second array
+ * @param keyFn - Function to extract comparison key from elements
+ * @returns Array of [indexA, indexB] pairs representing the LCS
+ */
+export function lcs(a, b, keyFn) {
+    const m = a.length;
+    const n = b.length;
+    // Build DP table
+    const dp = Array.from({ length: m + 1 }, () => Array.from({ length: n + 1 }, () => 0));
+    for (let i = 1; i <= m; i++) {
+        for (let j = 1; j <= n; j++) {
+            if (keyFn(a[i - 1]) === keyFn(b[j - 1])) {
+                dp[i][j] = dp[i - 1][j - 1] + 1;
+            }
+            else {
+                dp[i][j] = Math.max(dp[i - 1][j], dp[i][j - 1]);
+            }
+        }
+    }
+    // Backtrack to find LCS
+    const result = [];
+    let i = m;
+    let j = n;
+    while (i > 0 && j > 0) {
+        if (keyFn(a[i - 1]) === keyFn(b[j - 1])) {
+            result.push([i - 1, j - 1]);
+            i--;
+            j--;
+        }
+        else if (dp[i - 1][j] > dp[i][j - 1]) {
+            i--;
+        }
+        else {
+            j--;
+        }
+    }
+    return result.reverse();
+}
+/**
+ * Align paragraphs between original and revised documents.
+ *
+ * Uses hash-based LCS to find matching paragraphs, then classifies
+ * unmatched paragraphs as inserted or deleted.
+ *
+ * @param original - Paragraphs from original document
+ * @param revised - Paragraphs from revised document
+ * @param similarityThreshold - Minimum similarity to consider a match (default: 0.5)
+ */
+export function alignParagraphs(original, revised, similarityThreshold = 0.5) {
+    // Compute hashes for all paragraphs
+    const originalHashes = original.map(p => ({
+        para: p,
+        hash: hashParagraph(p.text),
+    }));
+    const revisedHashes = revised.map(p => ({
+        para: p,
+        hash: hashParagraph(p.text),
+    }));
+    // Find LCS based on hashes
+    const lcsResult = lcs(originalHashes, revisedHashes, item => item.hash);
+    // Build sets of matched indices
+    const matchedOriginal = new Set(lcsResult.map(([i]) => i));
+    const matchedRevised = new Set(lcsResult.map(([, j]) => j));
+    // Build alignment result
+    const result = {
+        matched: [],
+        deleted: [],
+        inserted: [],
+    };
+    // Add matched paragraphs
+    for (const [origIdx, revIdx] of lcsResult) {
+        const origPara = original[origIdx];
+        const revPara = revised[revIdx];
+        const similarity = computeSimilarity(origPara.text, revPara.text);
+        result.matched.push({
+            original: origPara,
+            revised: revPara,
+            similarity,
+        });
+    }
+    // Add deleted paragraphs (in original but not matched)
+    for (let i = 0; i < original.length; i++) {
+        if (!matchedOriginal.has(i)) {
+            result.deleted.push(original[i]);
+        }
+    }
+    // Add inserted paragraphs (in revised but not matched)
+    for (let j = 0; j < revised.length; j++) {
+        if (!matchedRevised.has(j)) {
+            result.inserted.push(revised[j]);
+        }
+    }
+    // Try to find near-matches for unmatched paragraphs
+    // This helps with paragraphs that were modified significantly
+    const unmatchedOriginal = [...result.deleted];
+    const unmatchedRevised = [...result.inserted];
+    result.deleted = [];
+    result.inserted = [];
+    for (const origPara of unmatchedOriginal) {
+        let bestMatch = null;
+        for (const revPara of unmatchedRevised) {
+            const similarity = computeSimilarity(origPara.text, revPara.text);
+            if (similarity >= similarityThreshold) {
+                if (!bestMatch || similarity > bestMatch.similarity) {
+                    bestMatch = { para: revPara, similarity };
+                }
+            }
+        }
+        if (bestMatch) {
+            // Remove from unmatched revised
+            const idx = unmatchedRevised.indexOf(bestMatch.para);
+            if (idx !== -1) {
+                unmatchedRevised.splice(idx, 1);
+            }
+            result.matched.push({
+                original: origPara,
+                revised: bestMatch.para,
+                similarity: bestMatch.similarity,
+            });
+        }
+        else {
+            result.deleted.push(origPara);
+        }
+    }
+    // Remaining unmatched revised are insertions
+    result.inserted = unmatchedRevised;
+    return result;
+}
+/**
+ * Classify alignment result into more granular categories.
+ */
+export function classifyAlignment(alignment) {
+    const result = {
+        identical: [],
+        modified: [],
+        deleted: alignment.deleted,
+        inserted: alignment.inserted,
+    };
+    for (const match of alignment.matched) {
+        if (match.similarity >= 0.9999) {
+            result.identical.push({
+                original: match.original,
+                revised: match.revised,
+            });
+        }
+        else {
+            result.modified.push(match);
+        }
+    }
+    return result;
+}
+//# sourceMappingURL=paragraphAlignment.js.map