quoted bbcode params support, support escape backslash

packages/bbob-parser/README.md

@@ -1,2 +1,15 @@
-# bbob-parser
-Fast BB Code parser written in pure javascript, no dependencies
+# @bbob/parser
+
+Parses BBCode and returns AST Tree looks like
+
+```json
+[
+    {
+        tag: 'url',
+        attrs: {
+            url: 'https://github.com/JiLiZART/bbob/tree/master/packages/bbob-parser'
+        },
+        content: ['hello', ' ', 'world!']
+    }
+]
+```

packages/bbob-parser/lib/Token.js

@@ -61,6 +61,10 @@ class Token {
     this.row = Number(row);
   }
 
+  isEmpty() {
+    return !!this.type;
+  }
+
   isText() {
     return isTextToken(this);
   }

packages/bbob-parser/lib/Tokenizer.js

@@ -4,6 +4,7 @@ const {
   CLOSE_BRAKET, EQ, TAB, SPACE, N, QUOTEMARK,
   PLACEHOLDER_SPACE, PLACEHOLDER_SPACE_TAB,
   SLASH,
+  BACKSLASH,
 } = require('./char');
 const Token = require('./Token');
 
@@ -41,6 +42,15 @@ class Tokenizer {
     this.emitToken(token);
   }
 
+  skipChar(num) {
+    this.index += num;
+    this.colPos += num;
+  }
+
+  seekChar(num) {
+    return this.buffer.charCodeAt(this.index + num);
+  }
+
   nextCol() {
     this.colPos += 1;
   }
@@ -115,6 +125,7 @@ class Tokenizer {
     }
 
     if (this.attrValueToken[Token.TYPE_ID]) {
+      delete this.attrValueToken.quoted;
       this.attrTokens.push(this.attrValueToken);
       this.attrValueToken = this.dummyToken;
     }
@@ -129,12 +140,16 @@ class Tokenizer {
 
   charSPACE(charCode) {
     this.flushWord();
+    const spaceCode = charCode === TAB ? PLACEHOLDER_SPACE_TAB : PLACEHOLDER_SPACE;
 
-    if (this.tagToken[Token.TYPE_ID]) {
-      this.attrNameToken = this.createAttrNameToken('');
+    if (this.inTag()) {
+      if (this.inAttrValue() && this.attrValueToken.quoted) {
+        this.attrValueToken[Token.VALUE_ID] += spaceCode;
+      } else {
+        this.flushAttrNames();
+        this.attrNameToken = this.createAttrNameToken('');
+      }
     } else {
-      const spaceCode = charCode === TAB ? PLACEHOLDER_SPACE_TAB : PLACEHOLDER_SPACE;
-
       this.appendToken(this.createSpaceToken(spaceCode));
     }
     this.nextCol();
@@ -156,16 +171,23 @@ class Tokenizer {
   }
 
   charCLOSEBRAKET() {
+    this.nextCol();
     this.flushTag();
     this.flushAttrNames();
     this.flushAttrs();
-
-    this.nextCol();
   }
 
   charEQ(charCode) {
-    if (this.tagToken[Token.TYPE_ID]) {
+    const nextCharCode = this.seekChar(1);
+    const isNextQuotemark = nextCharCode === QUOTEMARK;
+
+    if (this.inTag()) {
       this.attrValueToken = this.createAttrValueToken('');
+
+      if (isNextQuotemark) {
+        this.attrValueToken.quoted = true;
+        this.skipChar(1);
+      }
     } else {
       this.wordToken[Token.VALUE_ID] += getChar(charCode);
     }
@@ -174,7 +196,13 @@ class Tokenizer {
   }
 
   charQUOTEMARK(charCode) {
-    if (this.attrValueToken[Token.TYPE_ID] && this.attrValueToken[Token.VALUE_ID] > 0) {
+    const prevCharCode = this.seekChar(-1);
+    const isPrevBackslash = prevCharCode === BACKSLASH;
+
+    if (this.inAttrValue() &&
+        this.attrValueToken[Token.VALUE_ID] &&
+        this.attrValueToken.quoted &&
+        !isPrevBackslash) {
       this.flushAttrNames();
     } else if (this.tagToken[Token.TYPE_ID] === '') {
       this.wordToken[Token.VALUE_ID] += getChar(charCode);
@@ -183,13 +211,31 @@ class Tokenizer {
     this.nextCol();
   }
 
+  charBACKSLASH() {
+    const nextCharCode = this.seekChar(1);
+    const isNextQuotemark = nextCharCode === QUOTEMARK;
+
+    if (this.inAttrValue() &&
+        this.attrValueToken[Token.VALUE_ID] &&
+        this.attrValueToken.quoted &&
+        isNextQuotemark
+    ) {
+      this.attrValueToken[Token.VALUE_ID] += getChar(nextCharCode);
+      this.skipChar(1);
+    }
+
+    this.nextCol();
+  }
+
   charWORD(charCode) {
-    if (this.tagToken[Token.TYPE_ID] && this.attrValueToken[Token.TYPE_ID]) {
-      this.attrValueToken[Token.VALUE_ID] += getChar(charCode);
-    } else if (this.tagToken[Token.TYPE_ID] && this.attrNameToken[Token.TYPE_ID]) {
-      this.attrNameToken[Token.VALUE_ID] += getChar(charCode);
-    } else if (this.tagToken[Token.TYPE_ID]) {
-      this.tagToken[Token.VALUE_ID] += getChar(charCode);
+    if (this.inTag()) {
+      if (this.inAttrValue()) {
+        this.attrValueToken[Token.VALUE_ID] += getChar(charCode);
+      } else if (this.inAttrName()) {
+        this.attrNameToken[Token.VALUE_ID] += getChar(charCode);
+      } else {
+        this.tagToken[Token.VALUE_ID] += getChar(charCode);
+      }
     } else {
       this.createWord();
 
@@ -214,11 +260,11 @@ class Tokenizer {
           break;
 
         case OPEN_BRAKET:
-          this.charOPENBRAKET();
+          this.charOPENBRAKET(charCode);
           break;
 
         case CLOSE_BRAKET:
-          this.charCLOSEBRAKET();
+          this.charCLOSEBRAKET(charCode);
           break;
 
         case EQ:
@@ -229,6 +275,10 @@ class Tokenizer {
           this.charQUOTEMARK(charCode);
           break;
 
+        case BACKSLASH:
+          this.charBACKSLASH(charCode);
+          break;
+
         default:
           this.charWORD(charCode);
           break;
@@ -245,6 +295,18 @@ class Tokenizer {
     return this.tokens;
   }
 
+  inTag() {
+    return this.tagToken[Token.TYPE_ID];
+  }
+
+  inAttrValue() {
+    return this.attrValueToken[Token.TYPE_ID];
+  }
+
+  inAttrName() {
+    return this.attrNameToken[Token.TYPE_ID];
+  }
+
   createWordToken(value = '', line = this.colPos, row = this.rowPos) {
     return createTokenOfType(Token.TYPE_WORD, value, line, row);
   }

packages/bbob-parser/lib/char.js

@@ -11,6 +11,7 @@ const OPEN_BRAKET = '['.charCodeAt(0);
 const CLOSE_BRAKET = ']'.charCodeAt(0);
 
 const SLASH = '/'.charCodeAt(0);
+const BACKSLASH = '\\'.charCodeAt(0);
 
 const PLACEHOLDER_SPACE_TAB = '    ';
 const PLACEHOLDER_SPACE = ' ';
@@ -31,4 +32,5 @@ module.exports = {
   SLASH,
   PLACEHOLDER_SPACE_TAB,
   PLACEHOLDER_SPACE,
+  BACKSLASH,
 };

packages/bbob-parser/test/Parser.test.js

@@ -36,4 +36,21 @@ describe('Parser', () => {
       },
     ]);
   });
+
+  test('parse tag with quoted param with spaces', () => {
+    const ast = parse('[url href=https://ru.wikipedia.org target=_blank text="Foo Bar"]Text[/url]');
+
+    expect(ast).toBeInstanceOf(Array);
+    expect(ast).toEqual([
+      {
+        tag: 'url',
+        attrs: {
+          href: 'https://ru.wikipedia.org',
+          target: '_blank',
+          text: 'Foo Bar',
+        },
+        content: ['Text'],
+      },
+    ]);
+  });
 });

packages/bbob-parser/test/Tokenizer.test.js

@@ -47,6 +47,34 @@ describe('Tokenizer', () => {
     expectOutput(output, tokens);
   });
 
+  test('tokenize tag with quotemark params with spaces', () => {
+    const input = '[url text="Foo Bar"]Text[/url]';
+    const tokens = tokenize(input);
+    const output = [
+      [TYPE.TAG, 'url', '0', '0'],
+      [TYPE.ATTR_NAME, 'text', '4', '0'],
+      [TYPE.ATTR_VALUE, 'Foo Bar', '9', '0'],
+      [TYPE.WORD, 'Text', '20', '0'],
+      [TYPE.TAG, '/url', '24', '0'],
+    ];
+
+    expectOutput(output, tokens);
+  });
+
+  test('tokenize tag with escaped quotemark param', () => {
+    const input = `[url text="Foo \\"Bar"]Text[/url]`;
+    const tokens = tokenize(input);
+    const output = [
+      [TYPE.TAG, 'url', '0', '0'],
+      [TYPE.ATTR_NAME, 'text', '4', '0'],
+      [TYPE.ATTR_VALUE, 'Foo "Bar', '9', '0'],
+      [TYPE.WORD, 'Text', '22', '0'],
+      [TYPE.TAG, '/url', '26', '0'],
+    ];
+
+    expectOutput(output, tokens);
+  });
+
   test('tokenize tag param without quotemarks', () => {
     const input = '[style color=#ff0000]Text[/style]';
     const tokens = tokenize(input);