fast-xml-parser 4.3.4 → 4.3.6

package/src/v5/XmlPartReader.js

@@ -0,0 +1,212 @@
+'use strict';
+
+/**
+ * find paired tag for a stop node
+ * @param {string} xmlDoc 
+ * @param {string} tagName 
+ * @param {number} i : start index
+ */
+function readStopNode(xmlDoc, tagName, i){
+    const startIndex = i;
+    // Starting at 1 since we already have an open tag
+    let openTagCount = 1;
+  
+    for (; i < xmlDoc.length; i++) {
+      if( xmlDoc[i] === "<"){ 
+        if (xmlDoc[i+1] === "/") {//close tag
+            const closeIndex = findSubStrIndex(xmlDoc, ">", i, `${tagName} is not closed`);
+            let closeTagName = xmlDoc.substring(i+2,closeIndex).trim();
+            if(closeTagName === tagName){
+              openTagCount--;
+              if (openTagCount === 0) {
+                return {
+                  tagContent: xmlDoc.substring(startIndex, i),
+                  i : closeIndex
+                }
+              }
+            }
+            i=closeIndex;
+          } else if(xmlDoc[i+1] === '?') { 
+            const closeIndex = findSubStrIndex(xmlDoc, "?>", i+1, "StopNode is not closed.")
+            i=closeIndex;
+          } else if(xmlDoc.substr(i + 1, 3) === '!--') { 
+            const closeIndex = findSubStrIndex(xmlDoc, "-->", i+3, "StopNode is not closed.")
+            i=closeIndex;
+          } else if(xmlDoc.substr(i + 1, 2) === '![') { 
+            const closeIndex = findSubStrIndex(xmlDoc, "]]>", i, "StopNode is not closed.") - 2;
+            i=closeIndex;
+          } else {
+            const tagData = readTagExp(xmlDoc, i, '>')
+  
+            if (tagData) {
+              const openTagName = tagData && tagData.tagName;
+              if (openTagName === tagName && tagData.tagExp[tagData.tagExp.length-1] !== "/") {
+                openTagCount++;
+              }
+              i=tagData.closeIndex;
+            }
+          }
+        }
+    }//end for loop
+}
+
+/**
+ * Read closing tag name
+ * @param {Source} source 
+ * @returns tag name
+ */
+function readClosingTagName(source){
+  let text = ""; //temporary data
+  while(source.canRead()){
+    let ch = source.readCh();
+    // if (ch === null || ch === undefined) break;
+    // source.updateBuffer();
+
+    if (ch === ">") return text.trimEnd();
+    else text += ch;
+  }
+  throw new Error(`Unexpected end of source. Reading '${substr}'`);
+}
+
+/**
+ * Read XML tag and build attributes map
+ * This function can be used to read normal tag, pi tag.
+ * This function can't be used to read comment, CDATA, DOCTYPE.
+ * Eg <tag attr = ' some"' attr= ">" bool>
+ * @param {string} xmlDoc 
+ * @param {number} startIndex starting index
+ * @returns tag expression includes tag name & attribute string 
+ */
+function readTagExp(parser) {
+  let inSingleQuotes = false;
+  let inDoubleQuotes = false;
+  let i;
+  let EOE = false;
+
+  for (i = 0; parser.source.canRead(i); i++) {
+    const char = parser.source.readChAt(i);
+
+    if (char === "'" && !inDoubleQuotes) {
+      inSingleQuotes = !inSingleQuotes;
+    } else if (char === '"' && !inSingleQuotes) {
+      inDoubleQuotes = !inDoubleQuotes;
+    } else if (char === '>' && !inSingleQuotes && !inDoubleQuotes) {
+      // If not inside quotes, stop reading at '>'
+      EOE = true;
+      break;
+    }
+
+  }
+  if(inSingleQuotes || inDoubleQuotes){
+    throw new Error("Invalid attribute expression. Quote is not properly closed");
+  }else if(!EOE) throw new Error("Unexpected closing of source. Waiting for '>'");
+  
+  
+  const exp = parser.source.readStr(i);
+  parser.source.updateBufferBoundary(i + 1);
+  return buildTagExpObj(exp, parser)
+}
+
+function readPiExp(parser) {
+  let inSingleQuotes = false;
+  let inDoubleQuotes = false;
+  let i;
+  let EOE = false;
+
+  for (i = 0; parser.source.canRead(i) ; i++) {
+    const currentChar = parser.source.readChAt(i);
+    const nextChar =  parser.source.readChAt(i+1);
+
+    if (currentChar === "'" && !inDoubleQuotes) {
+      inSingleQuotes = !inSingleQuotes;
+    } else if (currentChar === '"' && !inSingleQuotes) {
+      inDoubleQuotes = !inDoubleQuotes;
+    }
+
+    if (!inSingleQuotes && !inDoubleQuotes) {
+      if (currentChar === '?' && nextChar === '>') {
+        EOE = true;
+        break; // Exit the loop when '?>' is found
+      }
+    }
+  }
+  if(inSingleQuotes || inDoubleQuotes){
+    throw new Error("Invalid attribute expression. Quote is not properly closed in PI tag expression");
+  }else if(!EOE) throw new Error("Unexpected closing of source. Waiting for '?>'");
+  
+  if(!parser.options.attributes.ignore){
+    //TODO: use regex to verify attributes if not set to ignore
+  }
+
+  const exp = parser.source.readStr(i);
+  parser.source.updateBufferBoundary(i + 1);
+  return buildTagExpObj(exp, parser)
+}
+
+function buildTagExpObj(exp, parser){
+  const tagExp = {
+    tagName: "",
+    selfClosing: false
+  };
+  let attrsExp = "";
+
+  if(exp[exp.length -1] === "/") tagExp.selfClosing = true;
+
+  //separate tag name
+  let i = 0;
+  for (; i < exp.length; i++) {
+    const char = exp[i];
+    if(char === " "){
+      tagExp.tagName = exp.substring(0, i);
+      attrsExp = exp.substring(i + 1);
+      break;
+    }
+  }
+  //only tag
+  if(tagExp.tagName.length === 0 && i === exp.length)tagExp.tagName = exp;
+
+  tagExp.tagName = tagExp.tagName.trimEnd();
+
+  if(!parser.options.attributes.ignore && attrsExp.length > 0){
+    parseAttributesExp(attrsExp,parser)
+  }
+
+  return tagExp;
+}
+
+const attrsRegx = new RegExp('([^\\s=]+)\\s*(=\\s*([\'"])([\\s\\S]*?)\\3)?', 'gm');
+
+function parseAttributesExp(attrStr, parser) {
+  const matches = getAllMatches(attrStr, attrsRegx);
+  const len = matches.length; //don't make it inline
+  for (let i = 0; i < len; i++) {
+    let attrName = parser.processAttrName(matches[i][1]);
+    let attrVal = parser.replaceEntities(matches[i][4] || true);
+    
+    parser.outputBuilder.addAttribute(attrName, attrVal);
+  }
+}
+
+
+const getAllMatches = function(string, regex) {
+  const matches = [];
+  let match = regex.exec(string);
+  while (match) {
+    const allmatches = [];
+    allmatches.startIndex = regex.lastIndex - match[0].length;
+    const len = match.length;
+    for (let index = 0; index < len; index++) {
+      allmatches.push(match[index]);
+    }
+    matches.push(allmatches);
+    match = regex.exec(string);
+  }
+  return matches;
+};
+
+module.exports = {
+    readStopNode: readStopNode,
+    readClosingTagName: readClosingTagName,
+    readTagExp: readTagExp,
+    readPiExp: readPiExp,
+}

package/src/v5/XmlSpecialTagsReader.js

@@ -0,0 +1,118 @@
+const {readPiExp} = require("./XmlPartReader");
+
+function readCdata(parser){
+  //<![ are already read till this point
+  let str = parser.source.readStr(6); //CDATA[
+  parser.source.updateBufferBoundary(6);
+
+  if(str !== "CDATA[") throw new Error(`Invalid CDATA expression at ${parser.source.line}:${parser.source.cols}`);
+
+  let text = parser.source.readUpto("]]>");
+  parser.outputBuilder.addCdata(text);
+}
+function readPiTag(parser){
+  //<? are already read till this point
+  let tagExp = readPiExp(parser, "?>");
+  if(!tagExp) throw new Error("Invalid Pi Tag expression.");
+
+  if (tagExp.tagName === "?xml") {//TODO: test if tagName is just xml
+    parser.outputBuilder.addDeclaration();
+  } else {
+    parser.outputBuilder.addPi("?"+tagExp.tagName);
+  }
+}
+
+function readComment(parser){
+  //<!- are already read till this point
+  let ch = parser.source.readCh();
+  if(ch !== "-") throw new Error(`Invalid comment expression at ${parser.source.line}:${parser.source.cols}`);
+
+  let text = parser.source.readUpto("-->");
+  parser.outputBuilder.addComment(text);
+}
+
+const DOCTYPE_tags = {
+  "EL":/^EMENT\s+([^\s>]+)\s+(ANY|EMPTY|\(.+\)\s*$)/m,
+  "AT":/^TLIST\s+[^\s]+\s+[^\s]+\s+[^\s]+\s+[^\s]+\s+$/m,
+  "NO":/^TATION.+$/m
+}
+function readDocType(parser){
+  //<!D are already read till this point
+  let str = parser.source.readStr(6); //OCTYPE
+  parser.source.updateBufferBoundary(6);
+
+  if(str !== "OCTYPE") throw new Error(`Invalid DOCTYPE expression at ${parser.source.line}:${parser.source.cols}`);
+
+  let hasBody = false, lastch = "";
+
+  while(parser.source.canRead()){
+    //TODO: use readChAt like used in partReader
+    let ch = parser.source.readCh();
+    if(hasBody){
+      if (ch === '<') { //Determine the tag type
+        let str = parser.source.readStr(2);
+        parser.source.updateBufferBoundary(2);
+        if(str === "EN"){ //ENTITY
+          let str = parser.source.readStr(4);
+          parser.source.updateBufferBoundary(4);
+          if(str !== "TITY") throw new Error("Invalid DOCTYPE ENTITY expression");
+
+          registerEntity(parser);
+        }else if(str === "!-") {//comment
+          readComment(parser);
+        }else{ //ELEMENT, ATTLIST, NOTATION
+          let dTagExp = parser.source.readUpto(">");
+          const regx = DOCTYPE_tags[str];
+          if(regx){
+            const match = dTagExp.match(regx);
+            if(!match) throw new Error("Invalid DOCTYPE");
+          }else throw new Error("Invalid DOCTYPE");
+        }
+      }else if( ch === '>' && lastch === "]"){//end of doctype
+        return;  
+      }
+    }else if( ch === '>'){//end of doctype
+      return;
+    }else if( ch === '['){
+      hasBody = true;
+    }else{
+      lastch = ch;
+    }
+  }//End While loop
+
+}
+
+function registerEntity(parser){
+  //read Entity
+  let attrBoundary="";
+  let name ="", val ="";
+  while(source.canRead()){
+    let ch = source.readCh();
+
+    if(attrBoundary){
+      if (ch === attrBoundary){
+        val = text;
+        text = ""
+      }
+    }else if(ch === " " || ch === "\t"){
+      if(!name){
+        name = text.trimStart();
+        text = "";
+      }
+    }else if (ch === '"' || ch === "'") {//start of attrBoundary
+      attrBoundary = ch;
+    }else if(ch === ">"){
+      parser.entityParser.addExternalEntity(name,val);
+      return;
+    }else{
+      text+=ch;
+    }
+  }
+}
+
+module.exports = {
+  readCdata: readCdata,
+  readComment:readComment,
+  readDocType:readDocType,
+  readPiTag:readPiTag
+}

package/src/v5/inputSource/BufferSource.js

@@ -0,0 +1,118 @@
+const Constants = {
+  space: 32,
+  tab: 9
+}
+class BufferSource{
+  constructor(bytesArr){
+    this.line = 1;
+    this.cols = 0;
+    this.buffer = bytesArr;
+    this.startIndex = 0;
+  }
+
+
+
+  readCh() {
+    return String.fromCharCode(this.buffer[this.startIndex++]);
+  }
+
+  readChAt(index) {
+    return String.fromCharCode(this.buffer[this.startIndex+index]);
+  }
+
+  readStr(n,from){
+    if(typeof from === "undefined") from = this.startIndex;
+    return this.buffer.slice(from, from + n).toString();
+  }
+
+  readUpto(stopStr) {
+    const inputLength = this.buffer.length;
+    const stopLength = stopStr.length;
+    const stopBuffer = Buffer.from(stopStr);
+
+    for (let i = this.startIndex; i < inputLength; i++) {
+        let match = true;
+        for (let j = 0; j < stopLength; j++) {
+            if (this.buffer[i + j] !== stopBuffer[j]) {
+                match = false;
+                break;
+            }
+        }
+
+        if (match) {
+            const result = this.buffer.slice(this.startIndex, i).toString();
+            this.startIndex = i + stopLength;
+            return result;
+        }
+    }
+
+    throw new Error(`Unexpected end of source. Reading '${stopStr}'`);
+}
+
+readUptoCloseTag(stopStr) { //stopStr: "</tagname"
+    const inputLength = this.buffer.length;
+    const stopLength = stopStr.length;
+    const stopBuffer = Buffer.from(stopStr);
+    let stopIndex = 0;
+    //0: non-matching, 1: matching stop string, 2: matching closing
+    let match = 0;
+
+    for (let i = this.startIndex; i < inputLength; i++) {
+        if(match === 1){//initial part matched
+            if(stopIndex === 0) stopIndex = i;
+            if(this.buffer[i] === Constants.space || this.buffer[i] === Constants.tab) continue;
+            else if(this.buffer[i] === '>'){ //TODO: if it should be equivalent ASCII
+                match = 2;
+                //tag boundary found
+                // this.startIndex
+            }
+        }else{
+            match = 1;
+            for (let j = 0; j < stopLength; j++) {
+                if (this.buffer[i + j] !== stopBuffer[j]) {
+                    match = 0;
+                    break;
+                }
+            }
+        }
+        if (match === 2) {//matched closing part
+            const result = this.buffer.slice(this.startIndex, stopIndex - 1 ).toString();
+            this.startIndex = i + 1;
+            return result;
+        }
+    }
+
+    throw new Error(`Unexpected end of source. Reading '${stopStr}'`);
+}
+
+  readFromBuffer(n, shouldUpdate) {
+    let ch;
+    if (n === 1) {
+      ch = this.buffer[this.startIndex];
+      if (ch === 10) {
+        this.line++;
+        this.cols = 1;
+      } else {
+        this.cols++;
+      }
+      ch = String.fromCharCode(ch);
+    } else {
+      this.cols += n;
+      ch = this.buffer.slice(this.startIndex, this.startIndex + n).toString();
+    }
+    if (shouldUpdate) this.updateBuffer(n);
+    return ch;
+  }
+
+  updateBufferBoundary(n = 1) { //n: number of characters read
+    this.startIndex += n;
+  }
+
+  canRead(n){
+    n = n || this.startIndex;
+    return this.buffer.length - n + 1 > 0;
+  }
+  
+}
+
+module.exports = BufferSource;

package/src/v5/inputSource/StringSource.js

@@ -0,0 +1,123 @@
+const whiteSpaces = [" ", "\n", "\t"];
+
+
+class StringSource{
+  constructor(str){
+    this.line = 1;
+    this.cols = 0;
+    this.buffer = str;
+    //a boundary pointer to indicate where from the buffer dat should be read
+    // data before this pointer can be deleted to free the memory
+    this.startIndex = 0;
+  }
+
+  readCh() {
+    return this.buffer[this.startIndex++];
+  }
+
+  readChAt(index) {
+    return this.buffer[this.startIndex+index];
+  }
+
+  readStr(n,from){
+    if(typeof from === "undefined") from = this.startIndex;
+    return this.buffer.substring(from, from + n);
+  }
+
+  readUpto(stopStr) {
+    const inputLength = this.buffer.length;
+    const stopLength = stopStr.length;
+
+    for (let i = this.startIndex; i < inputLength; i++) {
+      let match = true;
+      for (let j = 0; j < stopLength; j++) {
+        if (this.buffer[i + j] !== stopStr[j]) {
+          match = false;
+          break;
+        }
+      }
+
+      if (match) {
+        const result = this.buffer.substring(this.startIndex, i);
+        this.startIndex = i + stopLength;
+        return result;
+      }
+    }
+
+    throw new Error(`Unexpected end of source. Reading '${stopStr}'`);
+  }
+
+  readUptoCloseTag(stopStr) { //stopStr: "</tagname"
+    const inputLength = this.buffer.length;
+    const stopLength = stopStr.length;
+    let stopIndex = 0;
+    //0: non-matching, 1: matching stop string, 2: matching closing
+    let match = 0;
+
+    for (let i = this.startIndex; i < inputLength; i++) {
+      if(match === 1){//initial part matched
+        if(stopIndex === 0) stopIndex = i;
+        if(this.buffer[i] === ' ' || this.buffer[i] === '\t') continue;
+        else if(this.buffer[i] === '>'){
+          match = 2;
+          //tag boundary found
+          // this.startIndex
+        }
+      }else{
+        match = 1;
+        for (let j = 0; j < stopLength; j++) {
+          if (this.buffer[i + j] !== stopStr[j]) {
+            match = 0;
+            break;
+          }
+        }
+      }
+      if (match === 2) {//matched closing part
+        const result = this.buffer.substring(this.startIndex, stopIndex - 1 );
+        this.startIndex = i + 1;
+        return result;
+      }
+    }
+
+    throw new Error(`Unexpected end of source. Reading '${stopStr}'`);
+  }
+
+  readFromBuffer(n, updateIndex){
+    let ch;
+    if(n===1){
+      ch = this.buffer[this.startIndex];
+      // if(ch === "\n") {
+      //   this.line++;
+      //   this.cols = 1;
+      // }else{
+      //   this.cols++;
+      // }
+    }else{
+      ch = this.buffer.substring(this.startIndex, this.startIndex + n);
+      // if("".indexOf("\n") !== -1){
+      //   //TODO: handle the scenario when there are multiple lines
+      //   //TODO: col should be set to number of chars after last '\n'
+      //   // this.cols = 1;
+      // }else{
+      //   this.cols += n;
+
+      // }
+    }
+    if(updateIndex) this.updateBufferBoundary(n);
+    return ch;
+  }
+
+  //TODO: rename to updateBufferReadIndex
+  
+  updateBufferBoundary(n = 1) { //n: number of characters read
+    this.startIndex += n;
+  }
+
+  canRead(n){
+    n = n || this.startIndex;
+    return this.buffer.length - n + 1 > 0;
+  }
+  
+}
+
+module.exports = StringSource;

package/src/v5/valueParsers/EntitiesParser.js

@@ -0,0 +1,107 @@
+const ampEntity = { regex: /&(amp|#38|#x26);/g, val : "&"};
+const htmlEntities = {
+    "space": { regex: /&(nbsp|#160);/g, val: " " },
+    // "lt" : { regex: /&(lt|#60);/g, val: "<" },
+    // "gt" : { regex: /&(gt|#62);/g, val: ">" },
+    // "amp" : { regex: /&(amp|#38);/g, val: "&" },
+    // "quot" : { regex: /&(quot|#34);/g, val: "\"" },
+    // "apos" : { regex: /&(apos|#39);/g, val: "'" },
+    "cent" : { regex: /&(cent|#162);/g, val: "¢" },
+    "pound" : { regex: /&(pound|#163);/g, val: "£" },
+    "yen" : { regex: /&(yen|#165);/g, val: "¥" },
+    "euro" : { regex: /&(euro|#8364);/g, val: "€" },
+    "copyright" : { regex: /&(copy|#169);/g, val: "©" },
+    "reg" : { regex: /&(reg|#174);/g, val: "®" },
+    "inr" : { regex: /&(inr|#8377);/g, val: "₹" },
+    "num_dec": { regex: /&#([0-9]{1,7});/g, val : (_, str) => String.fromCharCode(Number.parseInt(str, 10)) },
+    "num_hex": { regex: /&#x([0-9a-fA-F]{1,6});/g, val : (_, str) => String.fromCharCode(Number.parseInt(str, 16)) },
+};
+
+class EntitiesParser{
+    constructor(replaceHtmlEntities) {
+      this.replaceHtmlEntities = replaceHtmlEntities;
+      this.docTypeEntities = {};
+      this.lastEntities = {
+        "apos" : { regex: /&(apos|#39|#x27);/g, val : "'"},
+        "gt" : { regex: /&(gt|#62|#x3E);/g, val : ">"},
+        "lt" : { regex: /&(lt|#60|#x3C);/g, val : "<"},
+        "quot" : { regex: /&(quot|#34|#x22);/g, val : "\""},
+      };
+    }
+
+    addExternalEntities(externalEntities){
+        const entKeys = Object.keys(externalEntities);
+        for (let i = 0; i < entKeys.length; i++) {
+          const ent = entKeys[i];
+          this.addExternalEntity(ent,externalEntities[ent])
+        }
+    }
+    addExternalEntity(key,val){
+      validateEntityName(key);
+      if(val.indexOf("&") !== -1) {
+        reportWarning(`Entity ${key} is not added as '&' is found in value;`)
+        return;
+      }else{
+        this.lastEntities[ent] = {
+          regex: new RegExp("&"+key+";","g"),
+          val : val
+        }
+      }
+    }
+
+    addDocTypeEntities(entities){
+        const entKeys = Object.keys(entities);
+        for (let i = 0; i < entKeys.length; i++) {
+          const ent = entKeys[i];
+          this.docTypeEntities[ent] = {
+             regex: new RegExp("&"+ent+";","g"),
+             val : entities[ent]
+          }
+        }
+    }
+
+    parse(val){
+        return this.replaceEntitiesValue(val)
+    }
+
+    /**
+     * 1. Replace DOCTYPE entities 
+     * 2. Replace external entities 
+     * 3. Replace HTML entities if asked
+     * @param {string} val 
+     */
+    replaceEntitiesValue(val){
+        if(typeof val === "string" && val.length > 0){
+            for(let entityName in this.docTypeEntities){
+                const entity = this.docTypeEntities[entityName];
+                val = val.replace( entity.regx, entity.val);
+              }
+              for(let entityName in this.lastEntities){
+                const entity = this.lastEntities[entityName];
+                val = val.replace( entity.regex, entity.val);
+              }
+              if(this.replaceHtmlEntities){
+                for(let entityName in htmlEntities){
+                  const entity = htmlEntities[entityName];
+                  val = val.replace( entity.regex, entity.val);
+                }
+              }
+            val = val.replace( ampEntity.regex, ampEntity.val);
+        }
+        return val;
+    }
+};
+
+//an entity name should not contains special characters that may be used in regex
+//Eg !?\\\/[]$%{}^&*()<>
+const specialChar = "!?\\\/[]$%{}^&*()<>|+";
+
+function validateEntityName(name){
+    for (let i = 0; i < specialChar.length; i++) {
+        const ch = specialChar[i];
+        if(name.indexOf(ch) !== -1) throw new Error(`Invalid character ${ch} in entity name`);
+    }
+    return name;
+}
+
+module.exports = EntitiesParser;

package/src/v5/valueParsers/booleanParser.js

@@ -0,0 +1,23 @@
+class boolParser{
+    constructor(trueList, falseList){
+        if(trueList)
+            this.trueList = trueList;
+        else
+            this.trueList = ["true"];
+        
+        if(falseList)
+            this.falseList = falseList;
+        else
+            this.falseList = ["false"];
+    }
+    parse(val){
+        if (typeof val === 'string') {
+            //TODO: performance: don't convert
+            const temp = val.toLowerCase();
+            if(this.trueList.indexOf(temp) !== -1) return true;
+            else if(this.falseList.indexOf(temp) !== -1 ) return false;
+        }
+        return val;
+    }
+}
+module.exports = boolParser;