'tests'

packages/app-cli/tests/HtmlToMd.js

@@ -1,97 +1,83 @@
-/* eslint-disable no-unused-vars */
+"use strict";
-
+var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
-
+    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
-const os = require('os');
+    return new (P || (P = Promise))(function (resolve, reject) {
-const time = require('@joplin/lib/time').default;
+        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
-const { filename } = require('@joplin/lib/path-utils');
+        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
-const { fileContentEqual, setupDatabase, setupDatabaseAndSynchronizer, db, synchronizer, fileApi, sleep, clearDatabase, switchClient, syncTargetId, objectsEqual, checkThrowAsync } = require('./test-utils.js');
+        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
-const Folder = require('@joplin/lib/models/Folder.js');
+        step((generator = generator.apply(thisArg, _arguments || [])).next());
-const Note = require('@joplin/lib/models/Note.js');
+    });
-const BaseModel = require('@joplin/lib/BaseModel').default;
+};
-const shim = require('@joplin/lib/shim').default;
+Object.defineProperty(exports, "__esModule", { value: true });
-const HtmlToMd = require('@joplin/lib/HtmlToMd');
+const shim_1 = require("@joplin/lib/shim");
-const { enexXmlToMd } = require('@joplin/lib/import-enex-md-gen.js');
+const os = require('os');
-
+const { filename } = require('@joplin/lib/path-utils');
-describe('HtmlToMd', function() {
+const HtmlToMd = require('@joplin/lib/HtmlToMd');
-
+describe('HtmlToMd', function () {
-	beforeEach(async (done) => {
+    // beforeEach(async (done) => {
-		await setupDatabaseAndSynchronizer(1);
+    // 	await setupDatabaseAndSynchronizer(1);
-		await switchClient(1);
+    // 	await switchClient(1);
-		done();
+    // 	done();
-	});
+    // });
-
+    it('should convert from Html to Markdown', (() => __awaiter(this, void 0, void 0, function* () {
-	it('should convert from Html to Markdown', (async () => {
+        const basePath = `${__dirname}/html_to_md`;
-		const basePath = `${__dirname}/html_to_md`;
+        const files = yield shim_1.default.fsDriver().readDirStats(basePath);
-		const files = await shim.fsDriver().readDirStats(basePath);
+        const htmlToMd = new HtmlToMd();
-		const htmlToMd = new HtmlToMd();
+        for (let i = 0; i < files.length; i++) {
-
+            const htmlFilename = files[i].path;
-		for (let i = 0; i < files.length; i++) {
+            if (htmlFilename.indexOf('.html') < 0)
-			const htmlFilename = files[i].path;
+                continue;
-			if (htmlFilename.indexOf('.html') < 0) continue;
+            const htmlPath = `${basePath}/${htmlFilename}`;
-
+            const mdPath = `${basePath}/${filename(htmlFilename)}.md`;
-			const htmlPath = `${basePath}/${htmlFilename}`;
+            if (htmlFilename !== 'anchor_same_title_and_url.html')
-			const mdPath = `${basePath}/${filename(htmlFilename)}.md`;
+                continue;
-
+            // if (htmlFilename.indexOf('image_preserve_size') !== 0) continue;
-			// if (htmlFilename !== 'code_3.html') continue;
+            const htmlToMdOptions = {};
-
+            if (htmlFilename === 'anchor_local.html') {
-			// if (htmlFilename.indexOf('image_preserve_size') !== 0) continue;
+                // Normally the list of anchor names in the document are retrieved from the HTML code
-
+                // This is straightforward when the document is still in DOM format, as with the clipper,
-			const htmlToMdOptions = {};
+                // but otherwise it would need to be somehow parsed out from the HTML. Here we just
-
+                // hard code the anchors that we know are in the file.
-			if (htmlFilename === 'anchor_local.html') {
+                htmlToMdOptions.anchorNames = ['first', 'second', 'fourth'];
-				// Normally the list of anchor names in the document are retrieved from the HTML code
+            }
-				// This is straightforward when the document is still in DOM format, as with the clipper,
+            if (htmlFilename.indexOf('image_preserve_size') === 0) {
-				// but otherwise it would need to be somehow parsed out from the HTML. Here we just
+                htmlToMdOptions.preserveImageTagsWithSize = true;
-				// hard code the anchors that we know are in the file.
+            }
-				htmlToMdOptions.anchorNames = ['first', 'second', 'fourth'];
+            const html = yield shim_1.default.fsDriver().readFile(htmlPath);
-			}
+            let expectedMd = yield shim_1.default.fsDriver().readFile(mdPath);
-
+            let actualMd = yield htmlToMd.parse(`<div>${html}</div>`, htmlToMdOptions);
-			if (htmlFilename.indexOf('image_preserve_size') === 0) {
+            if (os.EOL === '\r\n') {
-				htmlToMdOptions.preserveImageTagsWithSize = true;
+                expectedMd = expectedMd.replace(/\r\n/g, '\n');
-			}
+                actualMd = actualMd.replace(/\r\n/g, '\n');
-
+            }
-			const html = await shim.fsDriver().readFile(htmlPath);
+            if (actualMd !== expectedMd) {
-			let expectedMd = await shim.fsDriver().readFile(mdPath);
+                const result = [];
-
+                result.push('');
-			let actualMd = await htmlToMd.parse(`<div>${html}</div>`, htmlToMdOptions);
+                result.push(`Error converting file: ${htmlFilename}`);
-
+                result.push('--------------------------------- Got:');
-			if (os.EOL === '\r\n') {
+                result.push(actualMd.split('\n').map(l => `"${l}"`).join('\n'));
-				expectedMd = expectedMd.replace(/\r\n/g, '\n');
+                result.push('--------------------------------- Expected:');
-				actualMd = actualMd.replace(/\r\n/g, '\n');
+                result.push(expectedMd.split('\n').map(l => `"${l}"`).join('\n'));
-			}
+                result.push('--------------------------------------------');
-
+                result.push('');
-			if (actualMd !== expectedMd) {
+                console.info(result.join('\n'));
-				const result = [];
+                // console.info('');
-
+                // console.info(`Error converting file: ${htmlFilename}`);
-				result.push('');
+                // console.info('--------------------------------- Got:');
-				result.push(`Error converting file: ${htmlFilename}`);
+                // console.info(actualMd);
-				result.push('--------------------------------- Got:');
+                // console.info('--------------------------------- Raw:');
-				result.push(actualMd.split('\n').map(l => `"${l}"`).join('\n'));
+                // console.info(actualMd.split('\n'));
-				result.push('--------------------------------- Expected:');
+                // console.info('--------------------------------- Expected:');
-				result.push(expectedMd.split('\n').map(l => `"${l}"`).join('\n'));
+                // console.info(expectedMd.split('\n'));
-				result.push('--------------------------------------------');
+                // console.info('--------------------------------------------');
-				result.push('');
+                // console.info('');
-
+                expect(false).toBe(true);
-				console.info(result.join('\n'));
+                // return;
-
+            }
-
+            else {
-				// console.info('');
+                expect(true).toBe(true);
-				// console.info(`Error converting file: ${htmlFilename}`);
+            }
-				// console.info('--------------------------------- Got:');
+        }
-				// console.info(actualMd);
+    })));
-				// console.info('--------------------------------- Raw:');
+});
-				// console.info(actualMd.split('\n'));
+//# sourceMappingURL=HtmlToMd.js.map
-				// console.info('--------------------------------- Expected:');
-				// console.info(expectedMd.split('\n'));
-				// console.info('--------------------------------------------');
-				// console.info('');
-
-				expect(false).toBe(true);
-				// return;
-			} else {
-				expect(true).toBe(true);
-			}
-		}
-	}));
-
-});

packages/app-cli/tests/HtmlToMd.ts

@@ -0,0 +1,87 @@
+import shim from '@joplin/lib/shim';
+const os = require('os');
+const { filename } = require('@joplin/lib/path-utils');
+const HtmlToMd = require('@joplin/lib/HtmlToMd');
+
+describe('HtmlToMd', function() {
+
+	// beforeEach(async (done) => {
+	// 	await setupDatabaseAndSynchronizer(1);
+	// 	await switchClient(1);
+	// 	done();
+	// });
+
+	it('should convert from Html to Markdown', (async () => {
+		const basePath = `${__dirname}/html_to_md`;
+		const files = await shim.fsDriver().readDirStats(basePath);
+		const htmlToMd = new HtmlToMd();
+
+		for (let i = 0; i < files.length; i++) {
+			const htmlFilename = files[i].path;
+			if (htmlFilename.indexOf('.html') < 0) continue;
+
+			const htmlPath = `${basePath}/${htmlFilename}`;
+			const mdPath = `${basePath}/${filename(htmlFilename)}.md`;
+
+			if (htmlFilename !== 'anchor_same_title_and_url.html') continue;
+
+			// if (htmlFilename.indexOf('image_preserve_size') !== 0) continue;
+
+			const htmlToMdOptions:any = {};
+
+			if (htmlFilename === 'anchor_local.html') {
+				// Normally the list of anchor names in the document are retrieved from the HTML code
+				// This is straightforward when the document is still in DOM format, as with the clipper,
+				// but otherwise it would need to be somehow parsed out from the HTML. Here we just
+				// hard code the anchors that we know are in the file.
+				htmlToMdOptions.anchorNames = ['first', 'second', 'fourth'];
+			}
+
+			if (htmlFilename.indexOf('image_preserve_size') === 0) {
+				htmlToMdOptions.preserveImageTagsWithSize = true;
+			}
+
+			const html = await shim.fsDriver().readFile(htmlPath);
+			let expectedMd = await shim.fsDriver().readFile(mdPath);
+
+			let actualMd = await htmlToMd.parse(`<div>${html}</div>`, htmlToMdOptions);
+
+			if (os.EOL === '\r\n') {
+				expectedMd = expectedMd.replace(/\r\n/g, '\n');
+				actualMd = actualMd.replace(/\r\n/g, '\n');
+			}
+
+			if (actualMd !== expectedMd) {
+				const result = [];
+
+				result.push('');
+				result.push(`Error converting file: ${htmlFilename}`);
+				result.push('--------------------------------- Got:');
+				result.push(actualMd.split('\n').map(l => `"${l}"`).join('\n'));
+				result.push('--------------------------------- Expected:');
+				result.push(expectedMd.split('\n').map(l => `"${l}"`).join('\n'));
+				result.push('--------------------------------------------');
+				result.push('');
+
+				console.info(result.join('\n'));
+
+				// console.info('');
+				// console.info(`Error converting file: ${htmlFilename}`);
+				// console.info('--------------------------------- Got:');
+				// console.info(actualMd);
+				// console.info('--------------------------------- Raw:');
+				// console.info(actualMd.split('\n'));
+				// console.info('--------------------------------- Expected:');
+				// console.info(expectedMd.split('\n'));
+				// console.info('--------------------------------------------');
+				// console.info('');
+
+				expect(false).toBe(true);
+				// return;
+			} else {
+				expect(true).toBe(true);
+			}
+		}
+	}));
+
+});

packages/app-cli/tests/html_to_md/anchor_same_title_and_url.html

@@ -0,0 +1 @@
+<a href="https://example.com"/>https://example.com</a>

packages/app-cli/tests/html_to_md/anchor_same_title_and_url.md

@@ -0,0 +1 @@
+https://example.com

packages/turndown/src/commonmark-rules.js

@@ -227,6 +227,9 @@ rules.inlineLink = {
 
   replacement: function (content, node, options) {
     var href = filterLinkHref(node.getAttribute('href'))
+
+    console.info('RRRRRRRR', href);
+
     if (!href) {
       return getNamedAnchorFromLink(node, options) + filterLinkContent(content)
     } else {