lark-docx2md 0.2.1-beta.2 → 0.3.0

package/README.md

@@ -141,8 +141,3 @@ ISC
 ## 🙏 致谢
 
 本项目开发过程中获得了 [LINUX DO](https://linux.do/latest) 社区佬友的帮助，本产品会在社区发布，感谢社区的支持。
-
-## TODO
-
-- [ ] 富文本转换
-- [x] 电子表格导出为 Markdown

package/dist/cli.js

@@ -1,5 +1,5 @@
 #!/usr/bin/env node
-import { r as setLogLevel, t as convert } from "./converter-FwY1m1jm.js";
+import { r as setLogLevel, t as convert } from "./converter-C5Nrkvfa.js";
 import { Command } from "commander";
 import { LoggerLevel } from "@larksuiteoapi/node-sdk";
 //#region src/cli.ts

package/dist/{converter-FwY1m1jm.js → converter-C5Nrkvfa.js}

@@ -835,7 +835,7 @@ const sheetResolvedSerializer = {
 		if (node.type !== "sheetResolved") return "";
 		let out = "";
 		for (const s of node.sheets) {
-			out += `## 工作表：${s.title}\n\n`;
+			if (node.showTitle) out += `## ${node.title}-${s.title}\n\n`;
 			if (s.error) {
 				out += `> ${s.error}\n\n`;
 				continue;
@@ -2548,19 +2548,19 @@ function whiteboardNodesToYaml(nodes) {
 	const nctx = prepareNodeContext(nodes);
 	const yamlNodes = nctx.rootNodes.map((n) => convertNode(n, nctx)).filter((n) => n !== null);
 	const imageTokens = [];
-	collectImageTokens$1(yamlNodes, imageTokens);
+	collectImageTokens(yamlNodes, imageTokens);
 	return {
 		yaml: serializeYaml({ whiteboard: { nodes: yamlNodes } }),
 		imageTokens
 	};
 }
 /** 递归收集所有 image 节点的 token */
-function collectImageTokens$1(nodes, out) {
+function collectImageTokens(nodes, out) {
 	for (const node of nodes) {
 		if (node.type === "image" && node.token) out.push(node.token);
-		if (Array.isArray(node.children)) collectImageTokens$1(node.children, out);
+		if (Array.isArray(node.children)) collectImageTokens(node.children, out);
 		if (Array.isArray(node.cells)) {
-			for (const cell of node.cells) if (Array.isArray(cell.children)) collectImageTokens$1(cell.children, out);
+			for (const cell of node.cells) if (Array.isArray(cell.children)) collectImageTokens(cell.children, out);
 		}
 	}
 }
@@ -2666,16 +2666,16 @@ function createLogger(module) {
 //#endregion
 //#region src/md-ast/transformer.ts
 const logger$1 = createLogger("transformer");
+const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
 var MdTransformer = class {
-	constructor(client, opts) {
+	constructor(client, opts, sourceType = "docx") {
 		this.client = client;
 		this.opts = opts;
+		this.sourceType = sourceType;
 	}
 	async transform(ast) {
-		const imageTokens = collectImageTokens(ast);
-		const whiteboardTokens = collectWhiteboardTokens(ast);
-		const sheetTokens = collectSheetTokens(ast);
-		replaceInAst(ast, await this.resolveImages(imageTokens), await this.resolveWhiteboards(whiteboardTokens), await this.resolveSheets(sheetTokens));
+		const { imageTokens, whiteboardTokens, sheetRefs } = collectTokens(ast);
+		replaceInAst(ast, await this.resolveImages(imageTokens), await this.resolveWhiteboards(whiteboardTokens), await this.resolveSheets(sheetRefs));
 	}
 	async resolveImages(tokens) {
 		const map = /* @__PURE__ */ new Map();
@@ -2814,17 +2814,18 @@ var MdTransformer = class {
 		}
 		return yamlContent;
 	}
-	async resolveSheets(tokens) {
+	async resolveSheets(refs) {
 		const map = /* @__PURE__ */ new Map();
-		const uniqueTokens = [...new Set(tokens)];
-		if (uniqueTokens.length === 0) return map;
-		for (const token of uniqueTokens) {
-			if (!token) continue;
+		const uniqueRefs = refs.filter((r, i, arr) => arr.findIndex((x) => x.raw === r.raw) === i);
+		if (uniqueRefs.length === 0) return map;
+		for (const { raw, spreadsheetToken, sheetId } of uniqueRefs) {
+			if (!spreadsheetToken) continue;
 			try {
-				const info = await this.client.getSpreadsheetInfo(token);
-				const list = await this.client.listSheets(token);
+				const info = await this.client.getSpreadsheetInfo(spreadsheetToken);
+				const list = await this.client.listSheets(spreadsheetToken);
+				const sheetsToProcess = this.sourceType === "sheet" ? list : list.filter((s) => s.sheet_id === sheetId);
 				const resolved = [];
-				for (const s of list) {
+				for (const s of sheetsToProcess) {
 					if (s.hidden) continue;
 					if (s.resource_type && s.resource_type !== "sheet") {
 						resolved.push({
@@ -2836,7 +2837,7 @@ var MdTransformer = class {
 						continue;
 					}
 					try {
-						const meta = await this.client.getSheetMeta(token, s.sheet_id);
+						const meta = await this.client.getSheetMeta(spreadsheetToken, s.sheet_id);
 						const { row_count = 0, column_count = 0 } = meta.grid_properties ?? {};
 						if (!row_count || !column_count) {
 							resolved.push({
@@ -2847,7 +2848,7 @@ var MdTransformer = class {
 							continue;
 						}
 						const endCol = columnIndexToLetter(column_count);
-						const trimmed = trimTrailingEmpty(expandMerges((await this.client.readSheetValues(token, `${s.sheet_id}!A1:${endCol}${row_count}`) ?? []).map((row) => row.map(cellToMd)), meta.merges ?? []));
+						const trimmed = trimTrailingEmpty(expandMerges((await this.client.readSheetValues(spreadsheetToken, `${s.sheet_id}!A1:${endCol}${row_count}`) ?? []).map((row) => row.map(cellToMd)), meta.merges ?? []));
 						resolved.push({
 							title: s.title ?? "",
 							kind: "grid",
@@ -2860,45 +2861,59 @@ var MdTransformer = class {
 							rows: [],
 							error: `读取失败：${e.message}`
 						});
+					} finally {
+						await sleep(600);
 					}
 				}
-				map.set(token, {
+				map.set(raw, {
 					type: "sheetResolved",
 					title: info.title ?? "",
-					sheets: resolved
+					sheets: resolved,
+					showTitle: this.sourceType === "sheet"
 				});
 			} catch (e) {
-				logger$1.warn(`Failed to render sheet ${token}:`, e.message);
-				map.set(token, {
+				logger$1.warn(`Failed to resolve sheet ${raw}:`, e.message);
+				map.set(raw, {
 					type: "sheetResolved",
 					title: "",
-					sheets: []
+					sheets: [],
+					showTitle: false
 				});
 			}
 		}
 		return map;
 	}
 };
-function collectImageTokens(node) {
-	const tokens = [];
-	traverseBlockAst(node, (n) => {
-		if (n.type === "image") tokens.push(n.src);
-	});
-	return tokens;
-}
-function collectWhiteboardTokens(node) {
-	const tokens = [];
-	traverseBlockAst(node, (n) => {
-		if (n.type === "whiteboard") tokens.push(n.token);
-	});
-	return tokens;
-}
-function collectSheetTokens(node) {
-	const tokens = [];
+function collectTokens(node) {
+	const imageTokens = [];
+	const whiteboardTokens = [];
+	const sheetRefs = [];
 	traverseBlockAst(node, (n) => {
-		if (n.type === "sheet") tokens.push(n.token);
+		switch (n.type) {
+			case "image":
+				imageTokens.push(n.src);
+				break;
+			case "whiteboard":
+				whiteboardTokens.push(n.token);
+				break;
+			case "sheet": {
+				const lastIdx = n.token.lastIndexOf("_");
+				const spreadsheetToken = lastIdx > 0 ? n.token.slice(0, lastIdx) : n.token;
+				const sheetId = lastIdx > 0 ? n.token.slice(lastIdx + 1) : void 0;
+				sheetRefs.push({
+					raw: n.token,
+					spreadsheetToken,
+					sheetId
+				});
+				break;
+			}
+		}
 	});
-	return tokens;
+	return {
+		imageTokens,
+		whiteboardTokens,
+		sheetRefs
+	};
 }
 function replaceInAst(node, imageMap, whiteboardMap, sheetMap) {
 	if (node.type === "image") {
@@ -2983,7 +2998,7 @@ async function convert(opts) {
 		registerBuiltinParsers(parser);
 		ast = parser.parse(doc, blocks);
 	}
-	await new MdTransformer(client, opts).transform(ast);
+	await new MdTransformer(client, opts, objType === "sheet" ? "sheet" : "docx").transform(ast);
 	const serializer = new MdSerializer();
 	registerBuiltinSerializers(serializer);
 	const markdown = serializer.serialize(ast);
@@ -3003,4 +3018,4 @@ async function convert(opts) {
 //#endregion
 export { parseWikiUrl as n, setLogLevel as r, convert as t };
 
-//# sourceMappingURL=converter-FwY1m1jm.js.map
+//# sourceMappingURL=converter-C5Nrkvfa.js.map