Merge branch 'develop' into master

package.json

@@ -1,6 +1,6 @@
 {
   "name": "mfm-js",
-  "version": "0.14.1",
+  "version": "0.15.0",
   "description": "An MFM parser implementation with PEG.js",
   "main": "./built/index.js",
   "types": "./built/index.d.ts",
@@ -12,6 +12,7 @@
     "tsc": "tsc",
     "tsd": "tsd",
     "parse": "node ./built/cli/parse",
+    "parse-plain": "node ./built/cli/parsePlain",
     "test": "mocha -r ts-node/register 'test/**/*.ts' && npm run tsd"
   },
   "repository": {

src/cli/parsePlain.ts

@@ -0,0 +1,46 @@
+import { performance } from 'perf_hooks';
+import inputLine, { InputCanceledError } from './misc/inputLine';
+import { parsePlain } from '..';
+
+async function entryPoint() {
+	console.log('intaractive plain parser');
+
+	while (true) {
+		let input: string;
+		try {
+			input = await inputLine('> ');
+		}
+		catch (err) {
+			if (err instanceof InputCanceledError) {
+				console.log('bye.');
+				return;
+			}
+			throw err;
+		}
+
+		// replace special chars
+		input = input
+			.replace(/\\n/g, '\n')
+			.replace(/\\t/g, '\t')
+			.replace(/\\u00a0/g, '\u00a0');
+
+		try {
+			const parseTimeStart = performance.now();
+			const result = parsePlain(input);
+			const parseTimeEnd = performance.now();
+			console.log(JSON.stringify(result));
+			const parseTime = (parseTimeEnd - parseTimeStart).toFixed(3);
+			console.log(`parsing time: ${parseTime}ms`);
+		}
+		catch (err) {
+			console.log('parsing error:');
+			console.log(err);
+		}
+		console.log();
+	}
+}
+entryPoint()
+.catch(err => {
+	console.log(err);
+	process.exit(1);
+});

src/parser.pegjs

@@ -66,7 +66,7 @@ fullParser
 	= nodes:(&. n:(block / inline) { return n; })* { return mergeText(nodes); }
 
 plainParser
-	= nodes:(&. n:(emojiCode / unicodeEmoji / text) { return n; })* { return mergeText(nodes); }
+	= nodes:(&. n:(emojiCode / unicodeEmoji / plainText) { return n; })* { return mergeText(nodes); }
 
 inlineParser
 	= nodes:(&. n:inline { return n; })* { return mergeText(nodes); }
@@ -165,7 +165,7 @@ inline
 	/ url
 	/ link
 	/ fn
-	/ text
+	/ inlineText
 
 // inline: emoji code
 
@@ -219,16 +219,22 @@ small
 // inline: italic
 
 italic
+	= italicTag
+	/ italicAlt
+
+italicTag
 	= "<i>" content:(!"</i>" i:inline { return i; })+ "</i>"
 {
 	return ITALIC(mergeText(content));
 }
-	/ "*" content:$(!"*" ([a-z0-9]i / _))+ "*"
+
+italicAlt
+	= "*" content:$(!"*" ([a-z0-9]i / _))+ "*" &(EOF / LF / _)
 {
 	const parsedContent = applyParser(content, 'inlineParser');
 	return ITALIC(parsedContent);
 }
-	/ "_" content:$(!"_" ([a-z0-9]i / _))+ "_"
+	/ "_" content:$(!"_" ([a-z0-9]i / _))+ "_" &(EOF / LF / _)
 {
 	const parsedContent = applyParser(content, 'inlineParser');
 	return ITALIC(parsedContent);
@@ -289,7 +295,7 @@ mentionHostPart
 // inline: hashtag
 
 hashtag
-	= "#" content:hashtagContent
+	= "#" !("\uFE0F"? "\u20E3") content:hashtagContent
 {
 	return HASHTAG(content);
 }
@@ -382,7 +388,13 @@ fnArg
 
 // inline: text
 
-text
+inlineText
+	= !(LF / _) . &(hashtag / mention / italicAlt) . { return text(); } // hashtag, mention, italic ignore
+	/ . /* text node */
+
+// inline: text (for plainParser)
+
+plainText
 	= . /* text node */
 
 //

test/parser.ts

@@ -4,7 +4,29 @@ import {
 	TEXT, CENTER, FN, UNI_EMOJI, MENTION, EMOJI_CODE, HASHTAG, N_URL, BOLD, SMALL, ITALIC, STRIKE, QUOTE, MATH_BLOCK, SEARCH, CODE_BLOCK, LINK
 } from '../built/index';
 
-describe('parser', () => {
+describe('PlainParser', () => {
+	describe('text', () => {
+		it('basic', () => {
+			const input = 'abc';
+			const output = [TEXT('abc')];
+			assert.deepStrictEqual(mfm.parsePlain(input), output);
+		});
+
+		it('ignore hashtag', () => {
+			const input = 'abc#abc';
+			const output = [TEXT('abc#abc')];
+			assert.deepStrictEqual(mfm.parsePlain(input), output);
+		});
+
+		it('keycap number sign', () => {
+			const input = 'abc#️⃣abc';
+			const output = [TEXT('abc'), UNI_EMOJI('#️⃣'), TEXT('abc')];
+			assert.deepStrictEqual(mfm.parsePlain(input), output);
+		});
+	});
+});
+
+describe('FullParser', () => {
 	describe('text', () => {
 		it('普通のテキストを入力すると1つのテキストノードが返される', () => {
 			const input = 'abc';
@@ -222,6 +244,12 @@ describe('parser', () => {
 			const output = [TEXT('今起きた'), UNI_EMOJI('😇')];
 			assert.deepStrictEqual(mfm.parse(input), output);
 		});
+
+		it('keycap number sign', () => {
+			const input = 'abc#️⃣123';
+			const output = [TEXT('abc'), UNI_EMOJI('#️⃣'), TEXT('123')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
 	});
 
 	describe('big', () => {
@@ -338,7 +366,7 @@ describe('parser', () => {
 		});
 	});
 
-	describe('italic 1', () => {
+	describe('italic tag', () => {
 		it('basic', () => {
 			const input = '<i>abc</i>';
 			const output = [
@@ -376,7 +404,7 @@ describe('parser', () => {
 		});
 	});
 
-	describe('italic 2', () => {
+	describe('italic alt 1', () => {
 		it('basic', () => {
 			const input = '*abc*';
 			const output = [
@@ -386,6 +414,54 @@ describe('parser', () => {
 			];
 			assert.deepStrictEqual(mfm.parse(input), output);
 		});
+
+		it('basic 2', () => {
+			const input = 'before *abc* after';
+			const output = [
+				TEXT('before '),
+				ITALIC([
+					TEXT('abc')
+				]),
+				TEXT(' after')
+			];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('ignore a italic syntax if the before char is neither a space nor an LF', () => {
+			const input = 'before*abc*after';
+			const output = [TEXT('before*abc*after')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+	});
+
+	describe('italic alt 2', () => {
+		it('basic', () => {
+			const input = '_abc_';
+			const output = [
+				ITALIC([
+					TEXT('abc')
+				])
+			];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic 2', () => {
+			const input = 'before _abc_ after';
+			const output = [
+				TEXT('before '),
+				ITALIC([
+					TEXT('abc')
+				]),
+				TEXT(' after')
+			];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('ignore a italic syntax if the before char is neither a space nor an LF', () => {
+			const input = 'before_abc_after';
+			const output = [TEXT('before_abc_after')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
 	});
 
 	// strike
@@ -394,12 +470,73 @@ describe('parser', () => {
 
 	// mathInline
 
-	// mention
+	describe('mention', () => {
+		it('basic', () => {
+			const input = '@abc';
+			const output = [MENTION('abc', null, '@abc')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic 2', () => {
+			const input = 'before @abc after';
+			const output = [TEXT('before '), MENTION('abc', null, '@abc'), TEXT(' after')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic remote', () => {
+			const input = '@abc@misskey.io';
+			const output = [MENTION('abc', 'misskey.io', '@abc@misskey.io')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic remote 2', () => {
+			const input = 'before @abc@misskey.io after';
+			const output = [TEXT('before '), MENTION('abc', 'misskey.io', '@abc@misskey.io'), TEXT(' after')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic remote 3', () => {
+			const input = 'before\n@abc@misskey.io\nafter';
+			const output = [TEXT('before\n'), MENTION('abc', 'misskey.io', '@abc@misskey.io'), TEXT('\nafter')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('ignore format of mail address', () => {
+			const input = 'abc@example.com';
+			const output = [TEXT('abc@example.com')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+	});
 
 	describe('hashtag', () => {
-		it('and unicode emoji', () => {
-			const input = '#️⃣abc123#abc';
-			const output = [UNI_EMOJI('#️⃣'), TEXT('abc123'), HASHTAG('abc')];
+		it('basic', () => {
+			const input = '#abc';
+			const output = [HASHTAG('abc')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('basic 2', () => {
+			const input = 'before #abc after';
+			const output = [TEXT('before '), HASHTAG('abc'), TEXT(' after')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('with keycap number sign', () => {
+			const input = '#️⃣abc123 #abc';
+			const output = [UNI_EMOJI('#️⃣'), TEXT('abc123 '), HASHTAG('abc')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('with keycap number sign 2', () => {
+			const input = `abc
+#️⃣abc`;
+			const output = [TEXT('abc\n'), UNI_EMOJI('#️⃣'), TEXT('abc')];
+			assert.deepStrictEqual(mfm.parse(input), output);
+		});
+
+		it('ignore a hashtag if the before char is neither a space nor an LF', () => {
+			const input = 'abc#abc';
+			const output = [TEXT('abc#abc')];
 			assert.deepStrictEqual(mfm.parse(input), output);
 		});
 	});