Initial commit, basic html tests

mixu · Apr 6, 2013 · 5b44ef5 · 5b44ef5
commit 5b44ef5
Show file tree

Hide file tree

Showing 5 changed files with 166 additions and 0 deletions.
diff --git a/.gitignore b/.gitignore
@@ -0,0 +1 @@
+/node_modules/
diff --git a/index.js b/index.js
@@ -0,0 +1,49 @@
+var emptyTags = {
+  "area": 1,
+  "base": 1,
+  "basefont": 1,
+  "br": 1,
+  "col": 1,
+  "frame": 1,
+  "hr": 1,
+  "img": 1,
+  "input": 1,
+  "isindex": 1,
+  "link": 1,
+  "meta": 1,
+  "param": 1,
+  "embed": 1,
+  "?xml": 1
+};
+
+function html(item) {
+  // apply recursively to arrays
+  if(Array.isArray(item)) {
+    return item.map(html).join('');
+  }
+  switch(item.type) {
+    case 'text':
+      return item.data;
+    case 'directive':
+      return;
+    case 'comment':
+      return '<!--'+item.data+'-->';
+    case 'style':
+    case 'script':
+    case 'tag':
+      return '<'+item.name+
+               (item.attribs ?
+                  Object.keys(item.attribs).reduce(function(prev, key) {
+                    return prev + ' '+key+'="'+item.attribs[key]+'"'
+                  }, '') : ''
+               )+
+              '>'+
+              (item.children ? html(item.children) : '')+
+              (emptyTags[item.name] ? '' : '</'+item.name+'>');
+    case 'cdata':
+      return;
+  }
+
+}
+
+module.exports = html;
diff --git a/package.json b/package.json
@@ -0,0 +1,26 @@
+{
+  "name": "htmlparser-to-html",
+  "version": "0.0.0",
+  "description": "Converts the JSON that the htmlparser/htmlparser2 package produces back to HTML.",
+  "main": "index.js",
+  "scripts": {
+    "test": "node test.js"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/mixu/htmlparser-to-html.git"
+  },
+  "keywords": [
+    "html",
+    "parser",
+    "htmlparser",
+    "htmlparser2"
+  ],
+  "author": "Mikito Takada <mikito.takada@gmail.com>",
+  "license": "BSD",
+  "readmeFilename": "readme.md",
+  "devDependencies": {
+    "mocha": "~1.9.0",
+    "htmlparser": "~1.7.6"
+  }
+}
diff --git a/readme.md b/readme.md
@@ -0,0 +1,34 @@
+# htmlparser-to-html
+
+Converts the JSON that [htmlparser](https://npmjs.org/package/htmlparser) (and probably [htmlparser2](https://npmjs.org/package/htmlparser2)) produces back to HTML.
+
+Useful if you're doing some sort of transformation.
+
+Tests are based on reversing the parser tests in htmlparser, so they are quite comprehensive.
+
+## Usage
+
+    var html = require('htmlparser-to-html');
+
+    console.log(html([
+            {   type: 'tag'
+              , name: 'html'
+              , children:
+                 [ { type: 'tag'
+                   , name: 'title'
+                   , children: [ { data: 'The Title', type: 'text' } ]
+                   }
+                 , { type: 'tag'
+                   , name: 'body'
+                   , children: [ { data: 'Hello world', type: 'text' } ]
+                   }
+                 ]
+              }
+            ]));
+
+    // outputs: <html><title>The Title</title><body>Hello world</body></html>
+
+Of course, you probably want to generate the array from htmlparser.
+
+
+
diff --git a/test.js b/test.js
@@ -0,0 +1,56 @@
+var util = require('util'),
+    assert = require('assert'),
+
+    html = require('./index.js'),
+    htmlparser = require('htmlparser');
+
+// test cases are auto-generated  since we just check that
+// html(htmlparser(original)) == original is true
+var cases = {
+  "Basic test": "<html><title>The Title</title><body>Hello world</body></html>",
+  "Single Tag 1": "<br>text",
+  "Unescaped chars in script": "<head><script language=\"Javascript\">var foo = \"<bar>\"; alert(2 > foo); var baz = 10 << 2; var zip = 10 >> 1; var yap = \"<<>>>><<\";</script></head>",
+  "Special char in comment": "<head><!-- commented out tags <title>Test</title>--></head>",
+  "Script source in comment": "<script><!--var foo = 1;--></script>",
+  "Unescaped chars in style": "<style type=\"text/css\">\n body > p\n  { font-weight: bold; }</style>",
+  "Singular attribute": "<option value=\"foo\" selected=\"selected\"></option>",
+  "Text outside tags": "Line one\n<br>\nline two",
+  "Only text": "this is the text",
+  "Comment within text": "this is <!-- the comment --> the text",
+  "Comment within text within script": "<script>this is <!-- the comment --> the text</script>",
+  "XML Namespace": "<ns:tag>text</ns:tag>",
+};
+
+function parse(html) {
+  var handler = new htmlparser.DefaultHandler(function(error, dom) {
+      if(error) throw error;
+    }, { verbose: false, ignoreWhitespace: true });
+  var parser = new htmlparser.Parser(handler);
+  parser.parseComplete(html);
+  // silly async/sync hybrid
+  return handler.dom;
+}
+
+// this generates the exports tests
+Object.keys(cases).forEach(function(testName) {
+  var expected, original;
+  // if the output will differ from the input due to the permissive parsing, store the real expected value
+  if(Array.isArray(cases[testName])) {
+    original = cases[testName][0];
+    expected = cases[testName][1];
+  } else {
+    expected = original = cases[testName];
+  }
+  exports[testName] = function() {
+
+//    console.log(util.inspect(parse(original), false, 10, true));
+    assert.equal(html(parse(original)), original);
+  }
+});
+
+// if this module is the script being run, then run the tests:
+if (module == require.main) {
+  var mocha = require('child_process').spawn('./node_modules/.bin/mocha', [ '--colors', '--ui', 'exports', '--reporter', 'spec', __filename ]);
+  mocha.stdout.pipe(process.stdout);
+  mocha.stderr.pipe(process.stderr);
+}