clean-html - npm Package Compare versions

Comparing version 1.2.0 to 1.2.1

demo.js

@@ -14,5 +14,14 @@ var cleaner = require('./index.js'),
 
-    cleaner.clean(data, function (html) {
+    var options = {
+        'add-attr-to-remove': ['class'],
+        'add-tags-to-remove': ['b', 'i', 'u'],
+        'indent': '    ',
+        'remove-comments': true,
+        'remove-empty-paras': true,
+        'replace-nbsp': true
+    };
+
+    cleaner.clean(data, options, function (html) {
         console.log(html);
     });
 });

index.js

@@ -44,2 +44,3 @@ var htmlparser = require('htmlparser2'),
         'remove-empty-paras': false,
+        'replace-nbsp': false,
         'tags-to-remove': [
@@ -63,2 +64,3 @@ 'center',
     options['remove-empty-paras'] = opt['remove-empty-paras'] === true ? true : false;
+    options['replace-nbsp'] = opt['replace-nbsp'] === true ? true : false;
     options['tags-to-remove'] = opt['tags-to-remove'] || options['tags-to-remove'];
@@ -92,3 +94,10 @@
 function renderText(node) {
-    return node.data.replace(/\s+/g, ' ');
+    var text = node.data;
+
+    if (options['replace-nbsp']) {
+        text = text.replace(/ /g, ' ');
+    }
+
+    // replace all whitespace characters with a single space
+    return text.replace(/\s+/g, ' ');
 }
@@ -101,7 +110,9 @@
 
+    var comment = '<!--' + node.data + '-->';
+
     if (options['break-around-comments']) {
-        return '\n' + '<!--' + node.data + '-->' + '\n';
+        return '\n' + comment + '\n';
     }
 
-    return '<!--' + node.data + '-->';
+    return comment;
 }
@@ -122,4 +133,3 @@
 
-    var openTag = '<' + node.name,
-        closeTag;
+    var openTag = '<' + node.name;
 
@@ -142,3 +152,3 @@ for (var attrib in node.attribs) {
 
-    closeTag = '</' + node.name + '>';
+    var closeTag = '</' + node.name + '>';
 
@@ -179,6 +189,9 @@ if (options['block-tags'].indexOf(node.name) > -1) {
 
-    return html.replace(/\s{2,}/g, '\n');
+    // remove leading spaces and extra line breaks
+    html = html.replace(/\n\s+/g, '\n');
+
+    return html;
 }
 
-function indentLine(line, indentLevel) {
+function getIndent(indentLevel) {
     var indent = '';
@@ -190,3 +203,3 @@
 
-    return indent + line;
+    return indent;
 }
@@ -198,24 +211,31 @@
     return html.replace(/.*\n/g, function (line) {
-        var match = line.match(/<\/?(\w+).*?>/);
+        var openTags = [],
+            tagRegEx = /<\/?(\w+).*?>/g,
+            tag,
+            tagName,
+            result;
 
-        if (!match) {
-            return indentLine(line, indentLevel);
-        }
+        while (result = tagRegEx.exec(line)) {
+            tag = result[0];
+            tagName = result[1];
 
-        var tag = match[0],
-            tagName = match[1];
+            if (options['empty-tags'].indexOf(tagName) > -1) {
+                continue;
+            }
 
-        if (options['block-tags'].indexOf(tagName) > -1) {
             if (tag.indexOf('</') == -1) {
-                line = indentLine(line, indentLevel);
+                openTags.push(tag);
                 indentLevel++;
             } else {
+                openTags.pop();
                 indentLevel--;
-                line = indentLine(line, indentLevel);
             }
+        }
 
-            return line;
+        if (openTags.length) {
+            return getIndent(indentLevel - openTags.length)
+                + line.replace(openTags[0] + ' ', openTags[0] + '\n' + getIndent(indentLevel));
         }
 
-        return indentLine(line, indentLevel);
+        return getIndent(indentLevel) + line;
     });
@@ -222,0 +242,0 @@ }

package.json

 {
   "name": "clean-html",
-  "version": "1.2.0",
+  "version": "1.2.1",
   "description": "HTML cleaner and beautifier",
@@ -5,0 +5,0 @@ "main": "index.js",

readme.md

@@ -125,2 +125,9 @@ ## HTML cleaner and beautifier
 
+### replace-nbsp
+
+Replaces non-breaking white space entities (` `) with regular spaces.
+
+Type: Boolean  
+Default: `false`
+
 ### tags-to-remove
@@ -127,0 +134,0 @@

test.js

@@ -9,2 +9,10 @@ var assert = require('assert'),
 
+// test that non-breaking space is replaced
+cleaner.clean('Foo Bar', function (html) {
+    assert.equal(html, 'Foo Bar');
+});
+cleaner.clean('Foo Bar', {'replace-nbsp': true}, function (html) {
+    assert.equal(html, 'Foo Bar');
+});
+
 // test that extra whitespace is removed
@@ -11,0 +19,0 @@ cleaner.clean('Foo  Bar', function (html) {

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Major refactor

Supply chain risk

Package has recently undergone a major refactor. It may be unstable or indicate significant internal changes. Use caution when updating to versions that include significant changes.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

13648

increased by9.38%

Lines of code

306

increased by11.27%

Number of lines in readme file

170

increased by4.29%

Number of medium supply chain risk alerts

0

decreased by-100%

No dependency changes