@inshapardaz/likhari-converters 0.1.44 → 0.1.45-dev.334

package/dist/index.d.ts

@@ -4,4 +4,5 @@ export { plainTextConverter } from './plain-text/index.js';
 export { lexicalJsonConverter } from './lexical-json/index.js';
 export { markdownConverter } from './markdown/index.js';
 export { htmlConverter } from './html/index.js';
+export { mergePages, mergePagesToState, type MergePagesOptions, type PageInput } from './merge/index.js';
 //# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,SAAS,CAAC;AACxB,cAAc,YAAY,CAAC;AAC3B,OAAO,EAAE,kBAAkB,EAAE,MAAM,cAAc,CAAC;AAClD,OAAO,EAAE,oBAAoB,EAAE,MAAM,gBAAgB,CAAC;AACtD,OAAO,EAAE,iBAAiB,EAAE,MAAM,YAAY,CAAC;AAC/C,OAAO,EAAE,aAAa,EAAE,MAAM,QAAQ,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,SAAS,CAAC;AACxB,cAAc,YAAY,CAAC;AAC3B,OAAO,EAAE,kBAAkB,EAAE,MAAM,cAAc,CAAC;AAClD,OAAO,EAAE,oBAAoB,EAAE,MAAM,gBAAgB,CAAC;AACtD,OAAO,EAAE,iBAAiB,EAAE,MAAM,YAAY,CAAC;AAC/C,OAAO,EAAE,aAAa,EAAE,MAAM,QAAQ,CAAC;AACvC,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,KAAK,iBAAiB,EAAE,KAAK,SAAS,EAAE,MAAM,SAAS,CAAC"}

package/dist/index.js

@@ -4,3 +4,4 @@ export { plainTextConverter } from './plain-text/index.js';
 export { lexicalJsonConverter } from './lexical-json/index.js';
 export { markdownConverter } from './markdown/index.js';
 export { htmlConverter } from './html/index.js';
+export { mergePages, mergePagesToState } from './merge/index.js';

package/dist/merge/index.d.ts

@@ -0,0 +1,31 @@
+import type { SerializedEditorState } from 'lexical';
+import { type FormatRegistry } from '../registry.js';
+import type { ConverterContext, FormatId } from '../types.js';
+/** One page: its text, plus its format when that differs from `MergePagesOptions.inputFormat`. */
+export type PageInput = string | {
+    content: string;
+    format?: FormatId;
+};
+export interface MergePagesOptions {
+    /** The format of every page given as a bare string. Defaults to 'markdown'. */
+    inputFormat?: FormatId;
+    /** The format of the combined document. */
+    outputFormat: FormatId;
+    /** Passed to every parse and to the final serialize (image URL resolver and the like). */
+    context?: ConverterContext;
+    /** Defaults to the registry with all four built-in formats. */
+    registry?: FormatRegistry;
+}
+/** Parses each page and returns the combined document as Lexical state, without serializing it. */
+export declare function mergePagesToState(pages: PageInput[], options?: Omit<MergePagesOptions, 'outputFormat'>): SerializedEditorState;
+/**
+ * Merges pages (Markdown, Lexical JSON, or any other supported format) into one document,
+ * with a page break between each pair of pages, in the requested output format. Needs no
+ * editor or DOM, so it can run in a server, a worker or a build script.
+ *
+ * Footnote ids are only unique within a page, so a page whose footnote ids clash with an
+ * earlier page's gets them renamed (`1` becomes `1-2`), in references and notes alike; the
+ * notes of all pages end up in one list at the end of the document.
+ */
+export declare function mergePages(pages: PageInput[], options: MergePagesOptions): string;
+//# sourceMappingURL=index.d.ts.map

package/dist/merge/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/merge/index.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,qBAAqB,EAAE,MAAM,SAAS,CAAC;AACrD,OAAO,EAAyB,KAAK,cAAc,EAAE,MAAM,aAAa,CAAC;AACzE,OAAO,KAAK,EAAE,gBAAgB,EAAE,QAAQ,EAAE,MAAM,UAAU,CAAC;AAG3D,kGAAkG;AAClG,MAAM,MAAM,SAAS,GAAG,MAAM,GAAG;IAAE,OAAO,EAAE,MAAM,CAAC;IAAC,MAAM,CAAC,EAAE,QAAQ,CAAA;CAAE,CAAC;AAExE,MAAM,WAAW,iBAAiB;IAChC,+EAA+E;IAC/E,WAAW,CAAC,EAAE,QAAQ,CAAC;IACvB,2CAA2C;IAC3C,YAAY,EAAE,QAAQ,CAAC;IACvB,0FAA0F;IAC1F,OAAO,CAAC,EAAE,gBAAgB,CAAC;IAC3B,+DAA+D;IAC/D,QAAQ,CAAC,EAAE,cAAc,CAAC;CAC3B;AAED,mGAAmG;AACnG,wBAAgB,iBAAiB,CAAC,KAAK,EAAE,SAAS,EAAE,EAAE,OAAO,GAAE,IAAI,CAAC,iBAAiB,EAAE,cAAc,CAAM,GAAG,qBAAqB,CA4BlI;AAED;;;;;;;;GAQG;AACH,wBAAgB,UAAU,CAAC,KAAK,EAAE,SAAS,EAAE,EAAE,OAAO,EAAE,iBAAiB,GAAG,MAAM,CAGjF"}

package/dist/merge/index.js

@@ -0,0 +1,65 @@
+import { defaultFormatRegistry } from '../registry.js';
+import { elementBase, makeState, rootChildren } from '../shared/serialized.js';
+/** Parses each page and returns the combined document as Lexical state, without serializing it. */
+export function mergePagesToState(pages, options = {}) {
+    const { inputFormat = 'markdown', context, registry = defaultFormatRegistry } = options;
+    const body = [];
+    const footnotes = [];
+    const takenFootnoteIds = new Set();
+    pages.forEach((page, index) => {
+        const { content, format = inputFormat } = typeof page === 'string' ? { content: page } : page;
+        let blocks;
+        try {
+            blocks = structuredClone(rootChildren(registry.parse(format, content, context)));
+        }
+        catch (error) {
+            const reason = error instanceof Error ? error.message : String(error);
+            throw new Error(`mergePages: page ${index + 1} is not valid ${format}: ${reason}`);
+        }
+        // Each page brings its own footnote list; they are combined into one at the end.
+        const pageLists = blocks.filter((block) => block.type === 'footnote-list');
+        const pageBody = blocks.filter((block) => block.type !== 'footnote-list');
+        uniquifyFootnoteIds([...pageBody, ...pageLists], takenFootnoteIds);
+        for (const list of pageLists)
+            footnotes.push(...(list.children ?? []));
+        if (index > 0)
+            body.push({ type: 'page-break', version: 1 });
+        body.push(...pageBody);
+    });
+    return makeState(footnotes.length > 0 ? [...body, elementBase('footnote-list', footnotes)] : body);
+}
+/**
+ * Merges pages (Markdown, Lexical JSON, or any other supported format) into one document,
+ * with a page break between each pair of pages, in the requested output format. Needs no
+ * editor or DOM, so it can run in a server, a worker or a build script.
+ *
+ * Footnote ids are only unique within a page, so a page whose footnote ids clash with an
+ * earlier page's gets them renamed (`1` becomes `1-2`), in references and notes alike; the
+ * notes of all pages end up in one list at the end of the document.
+ */
+export function mergePages(pages, options) {
+    const registry = options.registry ?? defaultFormatRegistry;
+    return registry.serialize(options.outputFormat, mergePagesToState(pages, options), options.context);
+}
+/** Renames this page's footnote ids that an earlier page already uses, then claims all its ids. */
+function uniquifyFootnoteIds(blocks, taken) {
+    const nodes = [];
+    const collect = (node) => {
+        if (typeof node.footnoteId === 'string')
+            nodes.push(node);
+        (node.children ?? []).forEach(collect);
+    };
+    blocks.forEach(collect);
+    const renamed = new Map();
+    for (const node of nodes) {
+        const id = node.footnoteId;
+        if (!renamed.has(id)) {
+            let next = id;
+            for (let n = 2; taken.has(next); n += 1)
+                next = `${id}-${n}`;
+            renamed.set(id, next);
+        }
+        node.footnoteId = renamed.get(id);
+    }
+    renamed.forEach((id) => taken.add(id));
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inshapardaz/likhari-converters",
-  "version": "0.1.44",
+  "version": "0.1.45-dev.334",
   "description": "Format transformers (Markdown/HTML/plain-text/Lexical JSON) against the canonical Lexical EditorState, isolated from UI.",
   "license": "MIT",
   "repository": {