fast-xml-parser 4.0.0-beta.4 → 4.0.0-beta.8

package/CHANGELOG.md

@@ -1,5 +1,22 @@
 Note: If you find missing information about particular minor version, that version must have been changed without any functional change in this library.
 
+** 4.0.0-beta.8 / 2021-12-13**
+* call tagValueProcessor for stop nodes
+
+** 4.0.0-beta.7 / 2021-12-09**
+* fix Validator bug when an attribute has no value but '=' only
+* XML Builder should suppress unpaired tags by default.
+* documents update for missing features
+* refactoring to use Object.assign
+* refactoring to remove repeated code
+
+** 4.0.0-beta.6 / 2021-12-05**
+* Support PI Tags processing
+* Support `suppressBooleanAttributes` by XML Builder for attributes with value `true`.
+
+** 4.0.0-beta.5 / 2021-12-04**
+* fix: when a tag with name "attributes"
+
 ** 4.0.0-beta.4 / 2021-12-02**
 * Support HTML document parsing
 * skip stop nodes parsing when building the XML from JS object

package/README.md

@@ -55,6 +55,7 @@ Check the list of all known users [here](./USERs.md);
 * Supports comments
 * It can preserve Order of tags in JS object
 * You can control if a single tag should be parsed into array.
+* Supports parsing of PI (Processing Instruction) tags with XML declaration tags
 * And many more other features.
 
 ## How to use
@@ -108,6 +109,7 @@ In a HTML page
 4.  [XML Validator](./docs/v4/4.XMLValidator.md)
 5.  [Entites](./docs/5.Entities.md)
 6.  [HTML Document Parsing](./docs/6.HTMLParsing.md)
+7. [PI Tag processing](./docs/7.PITags.md)
 ## Performance
 
 ### XML Parser

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "fast-xml-parser",
-  "version": "4.0.0-beta.4",
+  "version": "4.0.0-beta.8",
   "description": "Validate XML, Parse XML, Build XML without C/C++ based libraries",
   "main": "./src/fxp.js",
   "scripts": {
@@ -47,6 +47,7 @@
     "@babel/preset-env": "^7.13.10",
     "@babel/register": "^7.13.8",
     "babel-loader": "^8.2.2",
+    "cytorus": "^0.2.9",
     "eslint": "^8.3.0",
     "he": "^1.2.0",
     "jasmine": "^3.6.4",

package/src/fxp.d.ts

@@ -44,6 +44,7 @@ type XmlBuilderOptions = {
   indentBy: string;
   arrayNodeName: string;
   suppressEmptyNode: boolean;
+  suppressBooleanAttributes: boolean;
   preserveOrder: boolean;
   unpairedTags: string[];
   stopNodes: string[];

package/src/util.js

@@ -67,23 +67,6 @@ exports.getValue = function(v) {
 // const fakeCall = function(a) {return a;};
 // const fakeCallNoReturn = function() {};
 
-const buildOptions = function(options, defaultOptions, props) {
-  let newOptions = {};
-  if (!options) {
-    return defaultOptions; //if there are not options
-  }
-
-  for (let i = 0; i < props.length; i++) {
-    if (options[props[i]] !== undefined) {
-      newOptions[props[i]] = options[props[i]];
-    } else {
-      newOptions[props[i]] = defaultOptions[props[i]];
-    }
-  }
-  return newOptions;
-};
-
-exports.buildOptions = buildOptions;
 exports.isName = isName;
 exports.getAllMatches = getAllMatches;
 exports.nameRegexp = nameRegexp;

package/src/validator.js

@@ -7,14 +7,9 @@ const defaultOptions = {
   unpairedTags: []
 };
 
-const props = [
-  'allowBooleanAttributes',
-  'unpairedTags'
-];
-
 //const tagsPattern = new RegExp("<\\/?([\\w:\\-_\.]+)\\s*\/?>","g");
 exports.validate = function (xmlData, options) {
-  options = util.buildOptions(options, defaultOptions, props);
+  options = Object.assign({}, defaultOptions, options);
 
   //xmlData = xmlData.replace(/(\r\n|\n|\r)/gm,"");//make it single line
   //xmlData = xmlData.replace(/(^\s*<\?xml.*?\?>)/g,"");//Remove XML starting tag
@@ -331,6 +326,8 @@ function validateAttributeString(attrStr, options) {
     if (matches[i][1].length === 0) {
       //nospace before attribute name: a="sd"b="saf"
       return getErrorObject('InvalidAttr', "Attribute '"+matches[i][2]+"' has no space in starting.", getPositionFromMatch(matches[i]))
+    } else if (matches[i][3] !== undefined && matches[i][4] === undefined) {
+      return getErrorObject('InvalidAttr', "Attribute '"+matches[i][2]+"' is without value.", getPositionFromMatch(matches[i]));
     } else if (matches[i][3] === undefined && !options.allowBooleanAttributes) {
       //independent attribute: ab
       return getErrorObject('InvalidAttr', "boolean attribute '"+matches[i][2]+"' is not allowed.", getPositionFromMatch(matches[i]));

package/src/xmlbuilder/json2xml.js

@@ -1,6 +1,5 @@
 'use strict';
 //parse Empty Node as self closing node
-const buildOptions = require('../util').buildOptions;
 const buildFromOrderedJs = require('./orderedJs2Xml');
 
 const defaultOptions = {
@@ -12,6 +11,7 @@ const defaultOptions = {
   format: false,
   indentBy: '  ',
   suppressEmptyNode: false,
+  suppressBooleanAttributes: true,
   tagValueProcessor: function(key, a) {
     return a;
   },
@@ -31,29 +31,8 @@ const defaultOptions = {
   stopNodes: []
 };
 
-const props = [
-  'attributeNamePrefix',
-  'attributesGroupName',
-  'textNodeName',
-  'ignoreAttributes',
-  'cdataPropName',
-  'format',
-  'indentBy',
-  'suppressEmptyNode',
-  'tagValueProcessor',
-  'attributeValueProcessor',
-  'arrayNodeName', //when array as root
-  'preserveOrder',
-  "commentPropName",
-  "unpairedTags",
-  "entities",
-  "processEntities",
-  "stopNodes",
-  // 'rootNodeName', //when jsObject have multiple properties on root level
-];
-
 function Builder(options) {
-  this.options = buildOptions(options, defaultOptions, props);
+  this.options = Object.assign({}, defaultOptions, options);
   if (this.options.ignoreAttributes || this.options.attributesGroupName) {
     this.isAttribute = function(/*a*/) {
       return false;

package/src/xmlbuilder/orderedJs2Xml.js

@@ -39,16 +39,18 @@ function arrToStr(arr, options, jPath, level){
         }else if( tagName === options.commentPropName){
             xmlStr += indentation + `<!--${tagObj[tagName][0][options.textNodeName]}-->`;
             continue;
+        }else if( tagName[0] === "?"){
+            const attStr = attr_to_str(tagObj[":@"], options);
+            xmlStr += indentation + `<${tagName} ${tagObj[tagName][0][options.textNodeName]} ${attStr}?>`;
+            continue;
         }
-        const attStr = attr_to_str(tagObj.attributes, options);
+        const attStr = attr_to_str(tagObj[":@"], options);
         let tagStart =  indentation + `<${tagName}${attStr}`;
         let tagValue = arrToStr(tagObj[tagName], options, newJPath, level + 1);
-        if( (!tagValue || tagValue.length === 0) && options.suppressEmptyNode){ 
-            if(options.unpairedTags.indexOf(tagName) !== -1){
-                xmlStr += tagStart + ">"; 
-            }else{
-                xmlStr += tagStart + "/>"; 
-            }
+        if(options.unpairedTags.indexOf(tagName) !== -1){
+            xmlStr += tagStart + ">"; 
+        }else if( (!tagValue || tagValue.length === 0) && options.suppressEmptyNode){ 
+            xmlStr += tagStart + "/>"; 
         }else{ 
             //TODO: node with only text value should not parse the text value in next line
             xmlStr += tagStart + `>${tagValue}${indentation}</${tagName}>` ;
@@ -62,7 +64,7 @@ function propName(obj){
     const keys = Object.keys(obj);
     for (let i = 0; i < keys.length; i++) {
       const key = keys[i];
-      if(key !== "attributes") return key;
+      if(key !== ":@") return key;
     }
   }
 
@@ -72,7 +74,11 @@ function attr_to_str(attrMap, options){
         for( attr in attrMap){
             let attrVal = options.attributeValueProcessor(attr, attrMap[attr]);
             attrVal = replaceEntitiesValue(attrVal, options);
-            attrStr+= ` ${attr.substr(options.attributeNamePrefix.length)}="${attrVal}"`;
+            if(attrVal === true && options.suppressBooleanAttributes){
+                attrStr+= ` ${attr.substr(options.attributeNamePrefix.length)}`;
+            }else{
+                attrStr+= ` ${attr.substr(options.attributeNamePrefix.length)}="${attrVal}"`;
+            }
         }
     }
     return attrStr;

package/src/xmlparser/OptionsBuilder.js

@@ -31,36 +31,9 @@ const defaultOptions = {
     htmlEntities: false,
 };
    
-const props = [
-    'preserveOrder',
-    'attributeNamePrefix',
-    'attributesGroupName',
-    'textNodeName',
-    'ignoreAttributes',
-    'removeNSPrefix',
-    'allowBooleanAttributes',
-    'parseTagValue',
-    'parseAttributeValue',
-    'trimValues',
-    'cdataPropName',
-    'tagValueProcessor',
-    'attributeValueProcessor',
-    'numberParseOptions',
-    'stopNodes',
-    'alwaysCreateTextNode',
-    'isArray',
-    'commentPropName',
-    'unpairedTags',
-    'processEntities',
-    'htmlEntities'
-];
-  
-const util = require('../util');
-
 const buildOptions = function(options) {
-    return util.buildOptions(options, defaultOptions, props);
+    return Object.assign({}, defaultOptions, options);
 };
 
 exports.buildOptions = buildOptions;
-exports.defaultOptions = defaultOptions;
-exports.props = props;
+exports.defaultOptions = defaultOptions;

package/src/xmlparser/OrderedObjParser.js

@@ -1,4 +1,5 @@
 'use strict';
+///@ts-check
 
 const util = require('../util');
 const xmlNode = require('./xmlNode');
@@ -47,8 +48,8 @@ class OrderedObjParser{
     this.buildAttributesMap = buildAttributesMap;
     this.isItStopNode = isItStopNode;
     this.replaceEntitiesValue = replaceEntitiesValue;
-    this.readTagExp = readTagExp;
     this.readStopNodeData = readStopNodeData;
+    this.saveTextToParentTag = saveTextToParentTag;
   }
 
 }
@@ -71,14 +72,15 @@ function addExternalEntities(externalEntities){
  * @param {boolean} dontTrim
  * @param {boolean} hasAttributes
  * @param {boolean} isLeafNode
+ * @param {boolean} escapeEntities
  */
-function parseTextData(val, tagName, jPath, dontTrim, hasAttributes, isLeafNode) {
+function parseTextData(val, tagName, jPath, dontTrim, hasAttributes, isLeafNode, escapeEntities) {
   if (val !== undefined) {
     if (this.options.trimValues && !dontTrim) {
       val = val.trim();
     }
     if(val.length > 0){
-      val = this.replaceEntitiesValue(val);
+      if(!escapeEntities) val = this.replaceEntitiesValue(val);
       
       const newval = this.options.tagValueProcessor(tagName, val, jPath, hasAttributes, isLeafNode);
       if(newval === null || newval === undefined){
@@ -192,14 +194,7 @@ const parseXml = function(xmlData) {
         }
 
         if(currentNode){
-          textData = this.parseTextData(textData
-            , currentNode.tagname
-            , jPath
-            ,false
-            , currentNode.attributes ? Object.keys(currentNode.attributes).length !== 0 : false
-            , Object.keys(currentNode.child).length === 0);
-          if(textData !== undefined &&  textData !== "") currentNode.add(this.options.textNodeName, textData);
-          textData = "";
+          textData = this.saveTextToParentTag(textData, currentNode, jPath);
         }
 
         jPath = jPath.substr(0, jPath.lastIndexOf("."));
@@ -208,24 +203,26 @@ const parseXml = function(xmlData) {
         textData = "";
         i = closeIndex;
       } else if( xmlData[i+1] === '?') {
-        i = findClosingIndex(xmlData, "?>", i, "Pi Tag is not closed.")
+        let tagData = readTagExp(xmlData,i, false, "?>");
+        if(!tagData) throw new Error("Pi Tag is not closed.");
+        textData = this.saveTextToParentTag(textData, currentNode, jPath);
+
+        const childNode = new xmlNode(tagData.tagName);
+        childNode.add(this.options.textNodeName, "");
+        
+        if(tagData.tagName !== tagData.tagExp && tagData.attrExpPresent){
+          childNode[":@"] = this.buildAttributesMap(tagData.tagExp, jPath);
+        }
+        currentNode.addChild(childNode);
+
+        i = tagData.closeIndex + 1;
       } else if(xmlData.substr(i + 1, 3) === '!--') {
         const endIndex = findClosingIndex(xmlData, "-->", i, "Comment is not closed.")
         if(this.options.commentPropName){
           const comment = xmlData.substring(i + 4, endIndex - 2);
 
-          //TODO: remove repeated code
-          if(textData){ //store previously collected data as textNode
-            textData = this.parseTextData(textData
-              , currentNode.tagname
-              , jPath
-              ,false
-              , currentNode.attributes ? Object.keys(currentNode.attributes).length !== 0 : false
-              , Object.keys(currentNode.child).length === 0);
-  
-            if(textData !== undefined &&  textData !== "") currentNode.add(this.options.textNodeName, textData);
-            textData = "";
-          }
+          textData = this.saveTextToParentTag(textData, currentNode, jPath);
+
           currentNode.add(this.options.commentPropName, [ { [this.options.textNodeName] : comment } ]);
         }
         i = endIndex;
@@ -237,17 +234,7 @@ const parseXml = function(xmlData) {
         const closeIndex = findClosingIndex(xmlData, "]]>", i, "CDATA is not closed.") - 2;
         const tagExp = xmlData.substring(i + 9,closeIndex);
 
-        if(textData){ //store previously collected data as textNode
-          textData = this.parseTextData(textData
-            , currentNode.tagname
-            , jPath
-            ,false
-            , currentNode.attributes ? Object.keys(currentNode.attributes).length !== 0 : false
-            , Object.keys(currentNode.child).length === 0);
-
-          if(textData !== undefined &&  textData !== "") currentNode.add(this.options.textNodeName, textData);
-          textData = "";
-        }
+        textData = this.saveTextToParentTag(textData, currentNode, jPath);
 
         //cdata should be set even if it is 0 length string
         if(this.options.cdataPropName){
@@ -263,7 +250,7 @@ const parseXml = function(xmlData) {
         i = closeIndex + 2;
       }else {//Opening tag
        
-        let result = this.readTagExp(xmlData,i);
+        let result = readTagExp(xmlData,i, this. options.removeNSPrefix);
         let tagName= result.tagName;
         let tagExp = result.tagExp;
         let attrExpPresent = result.attrExpPresent;
@@ -273,14 +260,7 @@ const parseXml = function(xmlData) {
         if (currentNode && textData) {
           if(currentNode.tagname !== '!xml'){
             //when nested tag is found
-            textData = this.parseTextData(textData
-              , currentNode.tagname
-              , jPath
-              , false
-              , currentNode.attributes ? Object.keys(currentNode.attributes).length !== 0 : false
-              , false);
-            if(textData !== undefined &&  textData !== "") currentNode.add(this.options.textNodeName, textData);
-            textData = "";
+            textData = this.saveTextToParentTag(textData, currentNode, jPath, false);
           }
         }
 
@@ -311,8 +291,12 @@ const parseXml = function(xmlData) {
 
           const childNode = new xmlNode(tagName);
           if(tagName !== tagExp && attrExpPresent){
-            childNode.attributes = this.buildAttributesMap(tagExp, jPath);
+            childNode[":@"] = this.buildAttributesMap(tagExp, jPath);
           }
+          if(tagContent) {
+            tagContent = this.parseTextData(tagContent, tagName, jPath, true, attrExpPresent, true, true);
+          }
+          
           jPath = jPath.substr(0, jPath.lastIndexOf("."));
           childNode.add(this.options.textNodeName, tagContent);
           
@@ -330,7 +314,7 @@ const parseXml = function(xmlData) {
 
             const childNode = new xmlNode(tagName);
             if(tagName !== tagExp && attrExpPresent){
-              childNode.attributes = this.buildAttributesMap(tagExp, jPath);
+              childNode[":@"] = this.buildAttributesMap(tagExp, jPath);
             }
             jPath = jPath.substr(0, jPath.lastIndexOf("."));
             currentNode.addChild(childNode);
@@ -341,7 +325,7 @@ const parseXml = function(xmlData) {
             this.tagsNodeStack.push(currentNode);
             
             if(tagName !== tagExp && attrExpPresent){
-              childNode.attributes = this.buildAttributesMap(tagExp, jPath);
+              childNode[":@"] = this.buildAttributesMap(tagExp, jPath);
             }
             currentNode.addChild(childNode);
             currentNode = childNode;
@@ -376,6 +360,24 @@ const replaceEntitiesValue = function(val){
   }
   return val;
 }
+function saveTextToParentTag(textData, currentNode, jPath, isLeafNode) {
+  if (textData) { //store previously collected data as textNode
+    if(isLeafNode === undefined) isLeafNode = Object.keys(currentNode.child).length === 0
+    
+    textData = this.parseTextData(textData,
+      currentNode.tagname,
+      jPath,
+      false,
+      currentNode[":@"] ? Object.keys(currentNode[":@"]).length !== 0 : false,
+      isLeafNode);
+
+    if (textData !== undefined && textData !== "")
+      currentNode.add(this.options.textNodeName, textData);
+    textData = "";
+  }
+  return textData;
+}
+
 //TODO: use jPath to simplify the logic
 /**
  * 
@@ -398,7 +400,7 @@ function isItStopNode(stopNodes, jPath, currentTagName){
  * @param {number} i starting index
  * @returns 
  */
-function tagExpWithClosingIndex(xmlData, i){
+function tagExpWithClosingIndex(xmlData, i, closingChar = ">"){
   let attrBoundary;
   let tagExp = "";
   for (let index = i; index < xmlData.length; index++) {
@@ -407,11 +409,20 @@ function tagExpWithClosingIndex(xmlData, i){
         if (ch === attrBoundary) attrBoundary = "";//reset
     } else if (ch === '"' || ch === "'") {
         attrBoundary = ch;
-    } else if (ch === '>') {
+    } else if (ch === closingChar[0]) {
+      if(closingChar[1]){
+        if(xmlData[index + 1] === closingChar[1]){
+          return {
+            data: tagExp,
+            index: index
+          }
+        }
+      }else{
         return {
           data: tagExp,
           index: index
         }
+      }
     } else if (ch === '\t') {
       ch = " "
     }
@@ -428,8 +439,9 @@ function findClosingIndex(xmlData, str, i, errMsg){
   }
 }
 
-function readTagExp(xmlData,i){
-  const result = tagExpWithClosingIndex(xmlData, i+1);
+function readTagExp(xmlData,i, removeNSPrefix, closingChar = ">"){
+  const result = tagExpWithClosingIndex(xmlData, i+1, closingChar);
+  if(!result) return;
   let tagExp = result.data;
   const closeIndex = result.index;
   const separatorIndex = tagExp.search(/\s/);
@@ -440,7 +452,7 @@ function readTagExp(xmlData,i){
     tagExp = tagExp.substr(separatorIndex + 1);
   }
 
-  if(this. options.removeNSPrefix){
+  if(removeNSPrefix){
     const colonIndex = tagName.indexOf(":");
     if(colonIndex !== -1){
       tagName = tagName.substr(colonIndex+1);

package/src/xmlparser/node2json.js

@@ -37,8 +37,8 @@ function compress(arr, options, jPath){
       let val = compress(tagObj[property], options, newJpath);
       const isLeaf = isLeafTag(val, options);
 
-      if(tagObj.attributes){
-        assignAttributes( val, tagObj.attributes, newJpath, options);
+      if(tagObj[":@"]){
+        assignAttributes( val, tagObj[":@"], newJpath, options);
       }else if(Object.keys(val).length === 1 && val[options.textNodeName] !== undefined && !options.alwaysCreateTextNode){
         val = val[options.textNodeName];
       }else if(Object.keys(val).length === 0){
@@ -74,7 +74,7 @@ function propName(obj){
   const keys = Object.keys(obj);
   for (let i = 0; i < keys.length; i++) {
     const key = keys[i];
-    if(key !== "attributes") return key;
+    if(key !== ":@") return key;
   }
 }
 

package/src/xmlparser/xmlNode.js

@@ -4,15 +4,15 @@ class XmlNode{
   constructor(tagname) {
     this.tagname = tagname;
     this.child = []; //nested tags, text, cdata, comments in order
-    this.attributes = {}; //attributes map
+    this[":@"] = {}; //attributes map
   }
   add(key,val){
     // this.child.push( {name : key, val: val, isCdata: isCdata });
     this.child.push( {[key]: val });
   }
   addChild(node) {
-    if(node.attributes && Object.keys(node.attributes).length > 0){
-      this.child.push( { [node.tagname]: node.child, attributes: node.attributes });
+    if(node[":@"] && Object.keys(node[":@"]).length > 0){
+      this.child.push( { [node.tagname]: node.child, [":@"]: node[":@"] });
     }else{
       this.child.push( { [node.tagname]: node.child });
     }