@cjser/parse5 8.0.1-cjser.2 → 8.1.0-cjser.2

package/dist-cjser/index.cjs

@@ -16,10 +16,11 @@ var __copyProps = (to, from, except, desc) => {
 };
 var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/index.js
 var index_exports = {};
 __export(index_exports, {
   ErrorCodes: () => ERR,
+  InsertionMode: () => InsertionMode,
   Parser: () => Parser,
   Token: () => token_exports,
   Tokenizer: () => Tokenizer,
@@ -34,7 +35,7 @@ __export(index_exports, {
 });
 module.exports = __toCommonJS(index_exports);
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/unicode.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/unicode.js
 var UNDEFINED_CODE_POINTS = /* @__PURE__ */ new Set([
   65534,
   65535,
@@ -125,7 +126,7 @@ function isUndefinedCodePoint(cp) {
   return cp >= 64976 && cp <= 65007 || UNDEFINED_CODE_POINTS.has(cp);
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/error-codes.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/error-codes.js
 var ERR;
 (function(ERR2) {
   ERR2["controlCharacterInInputStream"] = "control-character-in-input-stream";
@@ -190,7 +191,7 @@ var ERR;
   ERR2["eofInElementThatCanContainOnlyText"] = "eof-in-element-that-can-contain-only-text";
 })(ERR || (ERR = {}));
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/tokenizer/preprocessor.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/tokenizer/preprocessor.js
 var DEFAULT_BUFFER_WATERLINE = 1 << 16;
 var Preprocessor = class {
   constructor(handler) {
@@ -213,8 +214,13 @@ var Preprocessor = class {
   get col() {
     return this.pos - this.lineStartPos + Number(this.lastGapPos !== this.pos);
   }
+  /** The index in `html` at which the current code point starts (the high surrogate of a pair). */
+  get codePointStart() {
+    return this.pos - Number(this.lastGapPos === this.pos);
+  }
+  /** The offset of the current code point, before the second unit of a surrogate pair. */
   get offset() {
-    return this.droppedBufferSize + this.pos;
+    return this.droppedBufferSize + this.codePointStart;
   }
   getError(code, cpOffset) {
     const { line, col, offset } = this;
@@ -260,12 +266,15 @@ var Preprocessor = class {
   }
   dropParsedChunk() {
     if (this.willDropParsedChunk()) {
-      this.html = this.html.substring(this.pos);
-      this.lineStartPos -= this.pos;
-      this.droppedBufferSize += this.pos;
-      this.pos = 0;
+      const droppedSize = this.offset - this.droppedBufferSize;
+      this.html = this.html.substring(droppedSize);
+      this.lineStartPos -= droppedSize;
+      this.droppedBufferSize += droppedSize;
+      this.pos -= droppedSize;
       this.lastGapPos = -2;
       this.gapStack.length = 0;
+      if (this.pos > 0)
+        this._addGap();
     }
   }
   write(chunk, isLastChunk) {
@@ -359,7 +368,7 @@ var Preprocessor = class {
   }
 };
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/token.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/token.js
 var token_exports = {};
 __export(token_exports, {
   TokenType: () => TokenType,
@@ -386,10 +395,10 @@ function getTokenAttr(token, attrName) {
   return null;
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/tokenizer/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/tokenizer/index.js
 var import_decode = require("@cjser/entities/decode");
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/html.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/html.js
 var html_exports = {};
 __export(html_exports, {
   ATTRS: () => ATTRS,
@@ -915,7 +924,7 @@ function hasUnescapedText(tn, scriptingEnabled) {
   return UNESCAPED_TEXT.has(tn) || scriptingEnabled && tn === TAG_NAMES.NOSCRIPT;
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/tokenizer/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/tokenizer/index.js
 var State;
 (function(State2) {
   State2[State2["DATA"] = 0] = "DATA";
@@ -1019,7 +1028,7 @@ function toAsciiLower(cp) {
   return cp + 32;
 }
 function isWhitespace(cp) {
-  return cp === CODE_POINTS.SPACE || cp === CODE_POINTS.LINE_FEED || cp === CODE_POINTS.TABULATION || cp === CODE_POINTS.FORM_FEED;
+  return cp === CODE_POINTS.SPACE || cp === CODE_POINTS.LINE_FEED || cp === CODE_POINTS.TABULATION || cp === CODE_POINTS.FORM_FEED || cp === CODE_POINTS.CARRIAGE_RETURN;
 }
 function isScriptDataDoubleEscapeSequenceEnd(cp) {
   return isWhitespace(cp) || cp === CODE_POINTS.SOLIDUS || cp === CODE_POINTS.GREATER_THAN_SIGN;
@@ -1054,11 +1063,17 @@ var Tokenizer = class {
     this.currentCharacterToken = null;
     this.currentToken = null;
     this.currentAttr = { name: "", value: "" };
+    this.currentAttrLocation = null;
+    this.runStart = 0;
+    this.runEnd = 0;
     this.preprocessor = new Preprocessor(handler);
     this.currentLocation = this.getCurrentLocation(-1);
-    this.entityDecoder = new import_decode.EntityDecoder(import_decode.htmlDecodeTree, (cp, consumed) => {
+    this.entityDecoder = new import_decode.HtmlEntityDecoder((cp, consumed) => {
       this.preprocessor.pos = this.entityStartPos + consumed - 1;
-      this._flushCodePointConsumedAsCharacterReference(cp);
+      if (!this._isCharacterReferenceInAttribute()) {
+        this._ensureCharacterToken(isWhitespace(cp) ? TokenType.WHITESPACE_CHARACTER : cp === CODE_POINTS.NULL ? TokenType.NULL_CHARACTER : TokenType.CHARACTER);
+      }
+      this._extendRunWith(String.fromCodePoint(cp), 0);
     }, handler.onParseError ? {
       missingSemicolonAfterCharacterReference: () => {
         this._err(ERR.missingSemicolonAfterCharacterReference, 1);
@@ -1106,6 +1121,92 @@ var Tokenizer = class {
     this.inLoop = false;
   }
   //API
+  /** Emit buffered character tokens without completing the current token or emitting EOF. */
+  flushCharacters() {
+    var _a;
+    if (this.inLoop) {
+      throw new Error("Cannot flush characters from a tokenizer callback");
+    }
+    const location = (_a = this.currentCharacterToken) === null || _a === void 0 ? void 0 : _a.location;
+    let nextLocation = null;
+    if (location) {
+      nextLocation = this.getCharacterTokenEndLocation(location);
+      this.currentLocation = nextLocation;
+    }
+    this.inLoop = true;
+    try {
+      this._emitCurrentCharacterToken(nextLocation);
+    } finally {
+      this.inLoop = false;
+    }
+    this._runParsingLoop();
+  }
+  getCharacterTokenEndLocation(location) {
+    const endOffset = this.getCharacterTokenEndOffset();
+    const html = this.preprocessor.html.slice(location.startOffset - this.preprocessor.droppedBufferSize, endOffset - this.preprocessor.droppedBufferSize);
+    let line = location.startLine;
+    let lastLineBreak = -1;
+    for (let i = html.indexOf("\n"); i !== -1; i = html.indexOf("\n", i + 1)) {
+      line++;
+      lastLineBreak = i;
+    }
+    for (let i = html.indexOf("\r"); i !== -1; i = html.indexOf("\r", i + 1)) {
+      if (html.charCodeAt(i + 1) !== CODE_POINTS.LINE_FEED) {
+        line++;
+        lastLineBreak = Math.max(lastLineBreak, i);
+      }
+    }
+    return {
+      startLine: line,
+      startCol: lastLineBreak === -1 ? location.startCol + html.length : html.length - lastLineBreak,
+      startOffset: endOffset,
+      endLine: -1,
+      endCol: -1,
+      endOffset: -1
+    };
+  }
+  // Return the input boundary before any unfinished token or buffered delimiter.
+  getCharacterTokenEndOffset() {
+    var _a;
+    const end = this.preprocessor.droppedBufferSize + this.preprocessor.pos + 1;
+    const tokenLocation = (_a = this.currentToken) === null || _a === void 0 ? void 0 : _a.location;
+    switch (this.state) {
+      case State.TAG_OPEN:
+      case State.RCDATA_LESS_THAN_SIGN:
+      case State.RAWTEXT_LESS_THAN_SIGN:
+      case State.SCRIPT_DATA_LESS_THAN_SIGN:
+      case State.SCRIPT_DATA_ESCAPED_LESS_THAN_SIGN:
+      case State.CDATA_SECTION_BRACKET: {
+        return end - 1;
+      }
+      case State.END_TAG_OPEN:
+      case State.MARKUP_DECLARATION_OPEN:
+      case State.RCDATA_END_TAG_OPEN:
+      case State.RCDATA_END_TAG_NAME:
+      case State.RAWTEXT_END_TAG_OPEN:
+      case State.RAWTEXT_END_TAG_NAME:
+      case State.SCRIPT_DATA_END_TAG_OPEN:
+      case State.SCRIPT_DATA_END_TAG_NAME:
+      case State.SCRIPT_DATA_ESCAPED_END_TAG_OPEN:
+      case State.SCRIPT_DATA_ESCAPED_END_TAG_NAME:
+      case State.CDATA_SECTION_END: {
+        return end - 2;
+      }
+      case State.CHARACTER_REFERENCE: {
+        return this._isCharacterReferenceInAttribute() && tokenLocation ? tokenLocation.startOffset : this.preprocessor.droppedBufferSize + this.entityStartPos;
+      }
+      case State.DOCTYPE:
+      case State.BEFORE_DOCTYPE_NAME: {
+        return this.currentLocation.startOffset;
+      }
+      default: {
+        if (tokenLocation && (this.state === State.TAG_NAME || this.state >= State.BEFORE_ATTRIBUTE_NAME && this.state <= State.BOGUS_COMMENT || this.state >= State.COMMENT_START && this.state <= State.BOGUS_DOCTYPE)) {
+          return tokenLocation.startOffset;
+        }
+        return end;
+      }
+    }
+  }
   pause() {
     this.paused = true;
   }
@@ -1164,6 +1265,7 @@ var Tokenizer = class {
   }
   //Token creation
   _createStartTagToken() {
+    this._flushRun("");
     this.currentToken = {
       type: TokenType.START_TAG,
       tagName: "",
@@ -1175,6 +1277,7 @@ var Tokenizer = class {
     };
   }
   _createEndTagToken() {
+    this._flushRun("");
     this.currentToken = {
       type: TokenType.END_TAG,
       tagName: "",
@@ -1186,6 +1289,7 @@ var Tokenizer = class {
     };
   }
   _createCommentToken(offset) {
+    this._flushRun("");
     this.currentToken = {
       type: TokenType.COMMENT,
       data: "",
@@ -1193,6 +1297,7 @@ var Tokenizer = class {
     };
   }
   _createDoctypeToken(initialName) {
+    this._flushRun("");
     this.currentToken = {
       type: TokenType.DOCTYPE,
       name: initialName,
@@ -1202,20 +1307,27 @@ var Tokenizer = class {
       location: this.currentLocation
     };
   }
-  _createCharacterToken(type, chars) {
+  _createCharacterToken(type) {
+    if (this.currentLocation) {
+      const start = this.currentLocation.startOffset - this.preprocessor.droppedBufferSize;
+      if (this.preprocessor.html[start] === "\n" && this.preprocessor.html[start - 1] === "\r") {
+        this.currentLocation.startOffset++;
+      }
+    }
     this.currentCharacterToken = {
       type,
-      chars,
+      chars: "",
       location: this.currentLocation
     };
   }
   //Tag attributes
   _createAttr(attrNameFirstCh) {
+    this._flushRun("");
     this.currentAttr = {
       name: attrNameFirstCh,
       value: ""
     };
-    this.currentLocation = this.getCurrentLocation(0);
+    this.currentAttrLocation = this.getCurrentLocation(0);
   }
   _leaveAttrName() {
     var _a;
@@ -1223,9 +1335,9 @@ var Tokenizer = class {
     const token = this.currentToken;
     if (getTokenAttr(token, this.currentAttr.name) === null) {
       token.attrs.push(this.currentAttr);
-      if (token.location && this.currentLocation) {
+      if (token.location && this.currentAttrLocation) {
         const attrLocations = (_a = (_b = token.location).attrs) !== null && _a !== void 0 ? _a : _b.attrs = /* @__PURE__ */ Object.create(null);
-        attrLocations[this.currentAttr.name] = this.currentLocation;
+        attrLocations[this.currentAttr.name] = this.currentAttrLocation;
         this._leaveAttrValue();
       }
     } else {
@@ -1233,11 +1345,66 @@ var Tokenizer = class {
     }
   }
   _leaveAttrValue() {
-    if (this.currentLocation) {
-      this.currentLocation.endLine = this.preprocessor.line;
-      this.currentLocation.endCol = this.preprocessor.col;
-      this.currentLocation.endOffset = this.preprocessor.offset;
+    if (this.currentAttrLocation) {
+      this.currentAttrLocation.endLine = this.preprocessor.line;
+      this.currentAttrLocation.endCol = this.preprocessor.col;
+      this.currentAttrLocation.endOffset = this.preprocessor.offset;
+    }
+  }
+  /**
+   * Append the pending run and `text` to the attribute value, comment data or character token being built.
+   * The run is a string of its own: V8 copies a slice of fewer than 13 characters but lets a longer one share
+   * the input's buffer, which would keep that buffer alive for as long as any attribute value or text cut from
+   * it is; a concatenation is flattened into a buffer of its own when it is sliced.
+   */
+  _flushRun(text) {
+    if (this.runEnd === this.runStart && text.length === 0) {
+      return;
+    }
+    const slice = this.preprocessor.html.slice(this.runStart, this.runEnd);
+    const run = (slice.length < 13 ? slice : ` ${slice}`.slice(1)) + text;
+    const token = this.currentToken;
+    this.runStart = this.runEnd;
+    if (token === null) {
+      this.currentCharacterToken.chars += run;
+    } else if (token.type === TokenType.COMMENT) {
+      token.data += run;
+    } else {
+      this.currentAttr.value += run;
+    }
+  }
+  /** Extend the pending run by the current input code point, read as `cp`, or flush it and start a new one. */
+  _extendRun(cp) {
+    const { preprocessor } = this;
+    if (cp === CODE_POINTS.LINE_FEED && preprocessor.html.charCodeAt(preprocessor.pos) !== CODE_POINTS.LINE_FEED) {
+      this._flushRun("\n");
+      this.runStart = preprocessor.pos + 1;
+    } else {
+      const start = preprocessor.pos - Number(cp > 65535);
+      if (start !== this.runEnd) {
+        this._flushRun("");
+        this.runStart = start;
+      }
+    }
+    this.runEnd = preprocessor.pos + 1;
+  }
+  /**
+   * Extend the pending run by `text`, the input that ends `cpOffset` code points after the current one, or flush
+   * the run and append `text` where the input differs (a NUL replacement, a decoded reference, a dropped chunk).
+   */
+  _extendRunWith(text, cpOffset) {
+    const { preprocessor } = this;
+    const start = preprocessor.codePointStart + 1 + cpOffset - text.length;
+    if (start >= 0 && preprocessor.html.startsWith(text, start)) {
+      if (start !== this.runEnd) {
+        this._flushRun("");
+        this.runStart = start;
+      }
+    } else {
+      this._flushRun(text);
+      this.runStart = start + text.length;
     }
+    this.runEnd = start + text.length;
   }
   //Token emission
   prepareToken(ct) {
@@ -1279,6 +1446,7 @@ var Tokenizer = class {
     this.preprocessor.dropParsedChunk();
   }
   _emitCurrentCharacterToken(nextLocation) {
+    this._flushRun("");
     if (this.currentCharacterToken) {
       if (nextLocation && this.currentCharacterToken.location) {
         this.currentCharacterToken.location.endLine = nextLocation.startLine;
@@ -1322,27 +1490,39 @@ var Tokenizer = class {
   //1)TokenType.NULL_CHARACTER - \u0000-character sequences (e.g. '\u0000\u0000\u0000')
   //2)TokenType.WHITESPACE_CHARACTER - any whitespace/new-line character sequences (e.g. '\n  \r\t   \f')
   //3)TokenType.CHARACTER - any character sequence which don't belong to groups 1 and 2 (e.g. 'abcdef1234@@#$%^')
-  _appendCharToCurrentCharacterToken(type, ch) {
+  /** Make the current character token one of `type`, emitting the current one if it has another type. */
+  _ensureCharacterToken(type) {
     if (this.currentCharacterToken) {
       if (this.currentCharacterToken.type === type) {
-        this.currentCharacterToken.chars += ch;
         return;
-      } else {
-        this.currentLocation = this.getCurrentLocation(0);
-        this._emitCurrentCharacterToken(this.currentLocation);
-        this.preprocessor.dropParsedChunk();
       }
+      this.currentLocation = this.getCurrentLocation(0);
+      this._emitCurrentCharacterToken(this.currentLocation);
+      this.preprocessor.dropParsedChunk();
     }
-    this._createCharacterToken(type, ch);
+    this._createCharacterToken(type);
   }
+  /** Emit the current input code point, read as `cp`. */
   _emitCodePoint(cp) {
     const type = isWhitespace(cp) ? TokenType.WHITESPACE_CHARACTER : cp === CODE_POINTS.NULL ? TokenType.NULL_CHARACTER : TokenType.CHARACTER;
-    this._appendCharToCurrentCharacterToken(type, cp < 65536 ? String.fromCharCode(cp) : String.fromCodePoint(cp));
+    const token = this.currentCharacterToken;
+    if (token === null || token.type !== type) {
+      this._ensureCharacterToken(type);
+    }
+    this._extendRun(cp);
   }
   //NOTE: used when we emit characters explicitly.
   //This is always for non-whitespace and non-null characters, which allows us to avoid additional checks.
-  _emitChars(ch) {
-    this._appendCharToCurrentCharacterToken(TokenType.CHARACTER, ch);
+  // Negative `cpOffset` values locate text emitted before the current input character.
+  _emitChars(ch, cpOffset = 0) {
+    if (this.options.sourceCodeLocationInfo && cpOffset !== 0 && this.currentCharacterToken && this.currentCharacterToken.type !== TokenType.CHARACTER) {
+      this.currentLocation = this.getCurrentLocation(ch.length - 1 - cpOffset);
+      this._emitCurrentCharacterToken(this.currentLocation);
+      this._createCharacterToken(TokenType.CHARACTER);
+    } else {
+      this._ensureCharacterToken(TokenType.CHARACTER);
+    }
+    this._extendRunWith(ch, cpOffset);
   }
   // Character reference helpers
   _startCharacterReference() {
@@ -1356,7 +1536,7 @@ var Tokenizer = class {
   }
   _flushCodePointConsumedAsCharacterReference(cp) {
     if (this._isCharacterReferenceInAttribute()) {
-      this.currentAttr.value += String.fromCodePoint(cp);
+      this._extendRun(cp);
     } else {
       this._emitCodePoint(cp);
     }
@@ -1802,13 +1982,13 @@ var Tokenizer = class {
         }
         case CODE_POINTS.EOF: {
           this._err(ERR.eofBeforeTagName);
-          this._emitChars("<");
+          this._emitChars("<", -1);
           this._emitEOFToken();
           break;
         }
         default: {
           this._err(ERR.invalidFirstCharacterOfTagName);
-          this._emitChars("<");
+          this._emitChars("<", -1);
           this.state = State.DATA;
           this._stateData(cp);
         }
@@ -1830,7 +2010,7 @@ var Tokenizer = class {
         }
         case CODE_POINTS.EOF: {
           this._err(ERR.eofBeforeTagName);
-          this._emitChars("</");
+          this._emitChars("</", -1);
           this._emitEOFToken();
           break;
         }
@@ -1884,7 +2064,7 @@ var Tokenizer = class {
     if (cp === CODE_POINTS.SOLIDUS) {
       this.state = State.RCDATA_END_TAG_OPEN;
     } else {
-      this._emitChars("<");
+      this._emitChars("<", -1);
       this.state = State.RCDATA;
       this._stateRcdata(cp);
     }
@@ -1896,7 +2076,7 @@ var Tokenizer = class {
       this.state = State.RCDATA_END_TAG_NAME;
       this._stateRcdataEndTagName(cp);
     } else {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.RCDATA;
       this._stateRcdata(cp);
     }
@@ -1930,6 +2110,7 @@ var Tokenizer = class {
         return false;
       }
       default: {
+        this.currentToken = null;
         return !this._ensureHibernation();
       }
     }
@@ -1938,7 +2119,7 @@ var Tokenizer = class {
   //------------------------------------------------------------------
   _stateRcdataEndTagName(cp) {
     if (this.handleSpecialEndTag(cp)) {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.RCDATA;
       this._stateRcdata(cp);
     }
@@ -1949,7 +2130,7 @@ var Tokenizer = class {
     if (cp === CODE_POINTS.SOLIDUS) {
       this.state = State.RAWTEXT_END_TAG_OPEN;
     } else {
-      this._emitChars("<");
+      this._emitChars("<", -1);
       this.state = State.RAWTEXT;
       this._stateRawtext(cp);
     }
@@ -1961,7 +2142,7 @@ var Tokenizer = class {
       this.state = State.RAWTEXT_END_TAG_NAME;
       this._stateRawtextEndTagName(cp);
     } else {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.RAWTEXT;
       this._stateRawtext(cp);
     }
@@ -1970,7 +2151,7 @@ var Tokenizer = class {
   //------------------------------------------------------------------
   _stateRawtextEndTagName(cp) {
     if (this.handleSpecialEndTag(cp)) {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.RAWTEXT;
       this._stateRawtext(cp);
     }
@@ -1989,7 +2170,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        this._emitChars("<");
+        this._emitChars("<", -1);
         this.state = State.SCRIPT_DATA;
         this._stateScriptData(cp);
       }
@@ -2002,7 +2183,7 @@ var Tokenizer = class {
       this.state = State.SCRIPT_DATA_END_TAG_NAME;
       this._stateScriptDataEndTagName(cp);
     } else {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.SCRIPT_DATA;
       this._stateScriptData(cp);
     }
@@ -2011,7 +2192,7 @@ var Tokenizer = class {
   //------------------------------------------------------------------
   _stateScriptDataEndTagName(cp) {
     if (this.handleSpecialEndTag(cp)) {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.SCRIPT_DATA;
       this._stateScriptData(cp);
     }
@@ -2136,11 +2317,11 @@ var Tokenizer = class {
     if (cp === CODE_POINTS.SOLIDUS) {
       this.state = State.SCRIPT_DATA_ESCAPED_END_TAG_OPEN;
     } else if (isAsciiLetter(cp)) {
-      this._emitChars("<");
+      this._emitChars("<", -1);
       this.state = State.SCRIPT_DATA_DOUBLE_ESCAPE_START;
       this._stateScriptDataDoubleEscapeStart(cp);
     } else {
-      this._emitChars("<");
+      this._emitChars("<", -1);
       this.state = State.SCRIPT_DATA_ESCAPED;
       this._stateScriptDataEscaped(cp);
     }
@@ -2152,7 +2333,7 @@ var Tokenizer = class {
       this.state = State.SCRIPT_DATA_ESCAPED_END_TAG_NAME;
       this._stateScriptDataEscapedEndTagName(cp);
     } else {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.SCRIPT_DATA_ESCAPED;
       this._stateScriptDataEscaped(cp);
     }
@@ -2161,7 +2342,7 @@ var Tokenizer = class {
   //------------------------------------------------------------------
   _stateScriptDataEscapedEndTagName(cp) {
     if (this.handleSpecialEndTag(cp)) {
-      this._emitChars("</");
+      this._emitChars("</", -1);
       this.state = State.SCRIPT_DATA_ESCAPED;
       this._stateScriptDataEscaped(cp);
     }
@@ -2447,7 +2628,7 @@ var Tokenizer = class {
       }
       case CODE_POINTS.NULL: {
         this._err(ERR.unexpectedNullCharacter);
-        this.currentAttr.value += REPLACEMENT_CHARACTER;
+        this._extendRunWith(REPLACEMENT_CHARACTER, 0);
         break;
       }
       case CODE_POINTS.EOF: {
@@ -2456,7 +2637,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        this.currentAttr.value += String.fromCodePoint(cp);
+        this._extendRun(cp);
       }
     }
   }
@@ -2474,7 +2655,7 @@ var Tokenizer = class {
       }
       case CODE_POINTS.NULL: {
         this._err(ERR.unexpectedNullCharacter);
-        this.currentAttr.value += REPLACEMENT_CHARACTER;
+        this._extendRunWith(REPLACEMENT_CHARACTER, 0);
         break;
       }
       case CODE_POINTS.EOF: {
@@ -2483,7 +2664,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        this.currentAttr.value += String.fromCodePoint(cp);
+        this._extendRun(cp);
       }
     }
   }
@@ -2511,7 +2692,7 @@ var Tokenizer = class {
       }
       case CODE_POINTS.NULL: {
         this._err(ERR.unexpectedNullCharacter);
-        this.currentAttr.value += REPLACEMENT_CHARACTER;
+        this._extendRunWith(REPLACEMENT_CHARACTER, 0);
         break;
       }
       case CODE_POINTS.QUOTATION_MARK:
@@ -2520,7 +2701,7 @@ var Tokenizer = class {
       case CODE_POINTS.EQUALS_SIGN:
       case CODE_POINTS.GRAVE_ACCENT: {
         this._err(ERR.unexpectedCharacterInUnquotedAttributeValue);
-        this.currentAttr.value += String.fromCodePoint(cp);
+        this._extendRun(cp);
         break;
       }
       case CODE_POINTS.EOF: {
@@ -2529,7 +2710,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        this.currentAttr.value += String.fromCodePoint(cp);
+        this._extendRun(cp);
       }
     }
   }
@@ -2563,6 +2744,7 @@ var Tokenizer = class {
       }
       default: {
         this._err(ERR.missingWhitespaceBetweenAttributes);
+        this._leaveAttrValue();
         this.state = State.BEFORE_ATTRIBUTE_NAME;
         this._stateBeforeAttributeName(cp);
       }
@@ -2608,11 +2790,11 @@ var Tokenizer = class {
       }
       case CODE_POINTS.NULL: {
         this._err(ERR.unexpectedNullCharacter);
-        token.data += REPLACEMENT_CHARACTER;
+        this._extendRunWith(REPLACEMENT_CHARACTER, 0);
         break;
       }
       default: {
-        token.data += String.fromCodePoint(cp);
+        this._extendRun(cp);
       }
     }
   }
@@ -2684,7 +2866,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        token.data += "-";
+        this._extendRunWith("-", -1);
         this.state = State.COMMENT;
         this._stateComment(cp);
       }
@@ -2700,13 +2882,13 @@ var Tokenizer = class {
         break;
       }
       case CODE_POINTS.LESS_THAN_SIGN: {
-        token.data += "<";
+        this._extendRunWith("<", 0);
         this.state = State.COMMENT_LESS_THAN_SIGN;
         break;
       }
       case CODE_POINTS.NULL: {
         this._err(ERR.unexpectedNullCharacter);
-        token.data += REPLACEMENT_CHARACTER;
+        this._extendRunWith(REPLACEMENT_CHARACTER, 0);
         break;
       }
       case CODE_POINTS.EOF: {
@@ -2716,22 +2898,21 @@ var Tokenizer = class {
         break;
       }
       default: {
-        token.data += String.fromCodePoint(cp);
+        this._extendRun(cp);
       }
     }
   }
   // Comment less-than sign state
   //------------------------------------------------------------------
   _stateCommentLessThanSign(cp) {
-    const token = this.currentToken;
     switch (cp) {
       case CODE_POINTS.EXCLAMATION_MARK: {
-        token.data += "!";
+        this._extendRunWith("!", 0);
         this.state = State.COMMENT_LESS_THAN_SIGN_BANG;
         break;
       }
       case CODE_POINTS.LESS_THAN_SIGN: {
-        token.data += "<";
+        this._extendRunWith("<", 0);
         break;
       }
       default: {
@@ -2785,7 +2966,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        token.data += "-";
+        this._extendRunWith("-", -1);
         this.state = State.COMMENT;
         this._stateComment(cp);
       }
@@ -2806,7 +2987,7 @@ var Tokenizer = class {
         break;
       }
       case CODE_POINTS.HYPHEN_MINUS: {
-        token.data += "-";
+        this._extendRunWith("-", -2);
         break;
       }
       case CODE_POINTS.EOF: {
@@ -2816,7 +2997,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        token.data += "--";
+        this._extendRunWith("--", -1);
         this.state = State.COMMENT;
         this._stateComment(cp);
       }
@@ -2828,7 +3009,7 @@ var Tokenizer = class {
     const token = this.currentToken;
     switch (cp) {
       case CODE_POINTS.HYPHEN_MINUS: {
-        token.data += "--!";
+        this._extendRunWith("--!", -1);
         this.state = State.COMMENT_END_DASH;
         break;
       }
@@ -2845,7 +3026,7 @@ var Tokenizer = class {
         break;
       }
       default: {
-        token.data += "--!";
+        this._extendRunWith("--!", -1);
         this.state = State.COMMENT;
         this._stateComment(cp);
       }
@@ -3469,7 +3650,7 @@ var Tokenizer = class {
     if (cp === CODE_POINTS.RIGHT_SQUARE_BRACKET) {
       this.state = State.CDATA_SECTION_END;
     } else {
-      this._emitChars("]");
+      this._emitChars("]", -1);
       this.state = State.CDATA_SECTION;
       this._stateCdataSection(cp);
     }
@@ -3483,11 +3664,11 @@ var Tokenizer = class {
         break;
       }
       case CODE_POINTS.RIGHT_SQUARE_BRACKET: {
-        this._emitChars("]");
+        this._emitChars("]", -2);
         break;
       }
       default: {
-        this._emitChars("]]");
+        this._emitChars("]]", -1);
         this.state = State.CDATA_SECTION;
         this._stateCdataSection(cp);
       }
@@ -3531,7 +3712,7 @@ var Tokenizer = class {
   }
 };
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/parser/open-element-stack.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/parser/open-element-stack.js
 var IMPLICIT_END_TAG_REQUIRED = /* @__PURE__ */ new Set([TAG_ID.DD, TAG_ID.DT, TAG_ID.LI, TAG_ID.OPTGROUP, TAG_ID.OPTION, TAG_ID.P, TAG_ID.RB, TAG_ID.RP, TAG_ID.RT, TAG_ID.RTC]);
 var IMPLICIT_END_TAG_REQUIRED_THOROUGHLY = /* @__PURE__ */ new Set([
   ...IMPLICIT_END_TAG_REQUIRED,
@@ -3848,7 +4029,7 @@ var OpenElementStack = class {
   }
 };
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/parser/formatting-element-list.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/parser/formatting-element-list.js
 var NOAH_ARK_CAPACITY = 3;
 var EntryType;
 (function(EntryType2) {
@@ -3953,7 +4134,7 @@ var FormattingElementList = class {
   }
 };
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/tree-adapters/default.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/tree-adapters/default.js
 var defaultTreeAdapter = {
   //Node construction
   createDocument() {
@@ -4125,7 +4306,7 @@ var defaultTreeAdapter = {
   }
 };
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/doctype.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/doctype.js
 var VALID_DOCTYPE_NAME = "html";
 var VALID_SYSTEM_ID = "about:legacy-compat";
 var QUIRKS_MODE_SYSTEM_ID = "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd";
@@ -4234,7 +4415,7 @@ function getDocumentMode(token) {
   return DOCUMENT_MODE.NO_QUIRKS;
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/common/foreign-content.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/common/foreign-content.js
 var foreign_content_exports = {};
 __export(foreign_content_exports, {
   SVG_TAG_NAMES_ADJUSTMENT_MAP: () => SVG_TAG_NAMES_ADJUSTMENT_MAP,
@@ -4464,7 +4645,7 @@ function isIntegrationPoint(tn, ns, attrs, foreignNS) {
   return (!foreignNS || foreignNS === NS.HTML) && isHtmlIntegrationPoint(tn, ns, attrs) || (!foreignNS || foreignNS === NS.MATHML) && isMathMLTextIntegrationPoint(tn, ns);
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/parser/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/parser/index.js
 var HIDDEN_INPUT_TYPE = "hidden";
 var AA_OUTER_LOOP_ITER = 8;
 var AA_INNER_LOOP_ITER = 3;
@@ -4695,7 +4876,7 @@ var Parser = class {
     }
   }
   /** @protected */
-  _attachElementToTree(element, location) {
+  _attachElementToTree(element, location, insertionLocation) {
     if (this.options.sourceCodeLocationInfo) {
       const loc = location && {
         ...location,
@@ -4703,50 +4884,75 @@ var Parser = class {
       };
       this.treeAdapter.setNodeSourceCodeLocation(element, loc);
     }
-    if (this._shouldFosterParentOnInsertion()) {
-      this._fosterParentElement(element);
+    if (insertionLocation.beforeElement) {
+      this.treeAdapter.insertBefore(insertionLocation.parent, element, insertionLocation.beforeElement);
     } else {
-      const parent = this.openElements.currentTmplContentOrNode;
-      this.treeAdapter.appendChild(parent !== null && parent !== void 0 ? parent : this.document, element);
+      this.treeAdapter.appendChild(insertionLocation.parent, element);
     }
   }
+  /** @protected */
+  _getInsertionLocation() {
+    var _a;
+    if (this._shouldFosterParentOnInsertion()) {
+      return this._findFosterParentingLocation();
+    }
+    return {
+      parent: (_a = this.openElements.currentTmplContentOrNode) !== null && _a !== void 0 ? _a : this.document,
+      beforeElement: null
+    };
+  }
+  /** @protected */
+  _createElement(tagName, namespaceURI, attrs, intendedParent) {
+    var _a, _b;
+    const element = this.treeAdapter.createElement(tagName, namespaceURI, attrs);
+    (_b = (_a = this.treeAdapter).onElementCreated) === null || _b === void 0 ? void 0 : _b.call(_a, element, {
+      intendedParent,
+      formElement: this.formElement,
+      isInTemplate: this.openElements.tmplCount > 0
+    });
+    return element;
+  }
   /**
    * For self-closing tags. Add an element to the tree, but skip adding it
    * to the stack.
    */
   /** @protected */
   _appendElement(token, namespaceURI) {
-    const element = this.treeAdapter.createElement(token.tagName, namespaceURI, token.attrs);
-    this._attachElementToTree(element, token.location);
+    const insertionLocation = this._getInsertionLocation();
+    const element = this._createElement(token.tagName, namespaceURI, token.attrs, insertionLocation.parent);
+    this._attachElementToTree(element, token.location, insertionLocation);
   }
   /** @protected */
   _insertElement(token, namespaceURI) {
-    const element = this.treeAdapter.createElement(token.tagName, namespaceURI, token.attrs);
-    this._attachElementToTree(element, token.location);
+    const insertionLocation = this._getInsertionLocation();
+    const element = this._createElement(token.tagName, namespaceURI, token.attrs, insertionLocation.parent);
+    this._attachElementToTree(element, token.location, insertionLocation);
     this.openElements.push(element, token.tagID);
   }
   /** @protected */
   _insertFakeElement(tagName, tagID) {
-    const element = this.treeAdapter.createElement(tagName, NS.HTML, []);
-    this._attachElementToTree(element, null);
+    const insertionLocation = this._getInsertionLocation();
+    const element = this._createElement(tagName, NS.HTML, [], insertionLocation.parent);
+    this._attachElementToTree(element, null, insertionLocation);
     this.openElements.push(element, tagID);
   }
   /** @protected */
   _insertTemplate(token) {
-    const tmpl = this.treeAdapter.createElement(token.tagName, NS.HTML, token.attrs);
+    const insertionLocation = this._getInsertionLocation();
+    const tmpl = this._createElement(token.tagName, NS.HTML, token.attrs, insertionLocation.parent);
     const content = this.treeAdapter.createDocumentFragment();
     this.treeAdapter.setTemplateContent(tmpl, content);
-    this._attachElementToTree(tmpl, token.location);
+    this._attachElementToTree(tmpl, token.location, insertionLocation);
     this.openElements.push(tmpl, token.tagID);
     if (this.options.sourceCodeLocationInfo)
       this.treeAdapter.setNodeSourceCodeLocation(content, null);
   }
   /** @protected */
   _insertFakeRootElement() {
-    const element = this.treeAdapter.createElement(TAG_NAMES.HTML, NS.HTML, []);
+    const element = this._createElement(TAG_NAMES.HTML, NS.HTML, [], this.document);
     if (this.options.sourceCodeLocationInfo)
       this.treeAdapter.setNodeSourceCodeLocation(element, null);
-    this.treeAdapter.appendChild(this.openElements.current, element);
+    this.treeAdapter.appendChild(this.document, element);
     this.openElements.push(element, TAG_ID.HTML);
   }
   /** @protected */
@@ -5603,6 +5809,7 @@ function aaObtainFurthestBlock(p, formattingElementEntry) {
 function aaInnerLoop(p, furthestBlock, formattingElement) {
   let lastElement = furthestBlock;
   let nextElement = p.openElements.getCommonAncestor(furthestBlock);
+  const intendedParent = p.openElements.getCommonAncestor(formattingElement);
   for (let i = 0, element = nextElement; element !== formattingElement; i++, element = nextElement) {
     nextElement = p.openElements.getCommonAncestor(element);
     const elementEntry = p.activeFormattingElements.getElementEntry(element);
@@ -5614,7 +5821,7 @@ function aaInnerLoop(p, furthestBlock, formattingElement) {
       }
       p.openElements.remove(element);
     } else {
-      element = aaRecreateElementFromEntry(p, elementEntry);
+      element = aaRecreateElementFromEntry(p, elementEntry, intendedParent);
       if (lastElement === furthestBlock) {
         p.activeFormattingElements.bookmark = elementEntry;
       }
@@ -5625,9 +5832,9 @@ function aaInnerLoop(p, furthestBlock, formattingElement) {
   }
   return lastElement;
 }
-function aaRecreateElementFromEntry(p, elementEntry) {
+function aaRecreateElementFromEntry(p, elementEntry, intendedParent) {
   const ns = p.treeAdapter.getNamespaceURI(elementEntry.element);
-  const newElement = p.treeAdapter.createElement(elementEntry.token.tagName, ns, elementEntry.token.attrs);
+  const newElement = p._createElement(elementEntry.token.tagName, ns, elementEntry.token.attrs, intendedParent);
   p.openElements.replace(elementEntry.element, newElement);
   elementEntry.element = newElement;
   return newElement;
@@ -5648,7 +5855,7 @@ function aaInsertLastNodeInCommonAncestor(p, commonAncestor, lastElement) {
 function aaReplaceFormattingElement(p, furthestBlock, formattingElementEntry) {
   const ns = p.treeAdapter.getNamespaceURI(formattingElementEntry.element);
   const { token } = formattingElementEntry;
-  const newElement = p.treeAdapter.createElement(token.tagName, ns, token.attrs);
+  const newElement = p._createElement(token.tagName, ns, token.attrs, furthestBlock);
   p._adoptNodes(furthestBlock, newElement);
   p.treeAdapter.appendChild(furthestBlock, newElement);
   p.activeFormattingElements.insertElementAfterBookmark(newElement, token);
@@ -6556,7 +6763,7 @@ function genericEndTagInBody(p, token) {
   for (let i = p.openElements.stackTop; i > 0; i--) {
     const element = p.openElements.items[i];
     const elementId = p.openElements.tagIDs[i];
-    if (tid === elementId && (tid !== TAG_ID.UNKNOWN || p.treeAdapter.getTagName(element) === tn)) {
+    if (tid === elementId && p.treeAdapter.getNamespaceURI(element) === NS.HTML && (tid !== TAG_ID.UNKNOWN || p.treeAdapter.getTagName(element) === tn)) {
       p.openElements.generateImpliedEndTagsWithExclusion(tid);
       if (p.openElements.stackTop >= i)
         p.openElements.shortenToLength(i);
@@ -7489,7 +7696,7 @@ function endTagInForeignContent(p, token) {
   }
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/serializer/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/serializer/index.js
 var import_escape = require("@cjser/entities/escape");
 var VOID_ELEMENTS = /* @__PURE__ */ new Set([
   TAG_NAMES.AREA,
@@ -7602,7 +7809,7 @@ function serializeDocumentTypeNode(node, { treeAdapter }) {
   return `<!DOCTYPE ${treeAdapter.getDocumentTypeNodeName(node)}>`;
 }
 
-// packages/@cjser/parse5.tmp-26-1791908665740/dist/index.js
+// packages/@cjser/parse5.tmp-24-1793549660003/dist/index.js
 function parse(html, options) {
   return Parser.parse(html, options);
 }
@@ -7619,6 +7826,7 @@ function parseFragment(fragmentContext, html, options) {
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {
   ErrorCodes,
+  InsertionMode,
   Parser,
   Token,
   Tokenizer,