diff --git a/__tests__/ExpensiMark-Markdown-test.js b/__tests__/ExpensiMark-Markdown-test.js
index e6b4a12c..cfd9e288 100644
--- a/__tests__/ExpensiMark-Markdown-test.js
+++ b/__tests__/ExpensiMark-Markdown-test.js
@@ -5,9 +5,9 @@ const parser = new ExpensiMark();
test('Test bold HTML replacement', () => {
const boldTestStartString = 'This is a sentence, and it has some punctuation, words, and spaces. '
- + 'test * testing* test*test*test. * testing * *testing * '
- + 'This is a sentence, and it has some punctuation, words, and spaces. '
- + 'test * testing* test*test*test. * testing * *testing *';
+ + 'test * testing* test*test*test. * testing * *testing * '
+ + 'This is a sentence, and it has some punctuation, words, and spaces. '
+ + 'test * testing* test*test*test. * testing * *testing *';
const boldTestReplacedString = 'This is a *sentence,* and it has some *punctuation, words, and spaces*. '
+ '*test* * testing* test*test*test. * testing * *testing * '
+ 'This is a *sentence,* and it has some *punctuation, words, and spaces*. '
@@ -18,9 +18,9 @@ test('Test bold HTML replacement', () => {
test('Test italic HTML replacement', () => {
const italicTestStartString = 'This is a sentence, and it has some punctuation, words, and spaces. test _ testing_ test_test_test. _ test _ _test _ '
- + 'This is a sentence, and it has some punctuation, words, and spaces. test _ testing_ test_test_test. _ test _ _test _';
+ + 'This is a sentence, and it has some punctuation, words, and spaces. test _ testing_ test_test_test. _ test _ _test _';
const italicTestReplacedString = 'This is a _sentence,_ and it has some _punctuation, words, and spaces_. _test_ _ testing_ test_test_test. _ test _ _test _ '
- + 'This is a _sentence,_ and it has some _punctuation, words, and spaces_. _test_ _ testing_ test_test_test. _ test _ _test _';
+ + 'This is a _sentence,_ and it has some _punctuation, words, and spaces_. _test_ _ testing_ test_test_test. _ test _ _test _';
expect(parser.htmlToMarkdown(italicTestStartString)).toBe(italicTestReplacedString);
});
@@ -64,3 +64,53 @@ test('Test HTML string with attributes', () => {
expect(parser.htmlToMarkdown(testString)).toBe(resultString);
});
+
+test('Test HTML string with spcial Tags', () => {
+ const testString = '\n\ntest message\n\n\n';
+ const resultString = 'test message';
+
+ expect(parser.htmlToMarkdown(testString)).toBe(resultString);
+});
+
+
+test('Test HTML string with Internal Tags', () => {
+ const testString = `
+
+test message
`;
+ const resultString = 'test message';
+
+ expect(parser.htmlToMarkdown(testString)).toBe(resultString);
+});
+
+test('Test HTML string with encoded entities', () => {
+ const testString = 'Text Entity & "';
+ const resultString = 'Text Entity & "';
+
+ expect(parser.htmlToMarkdown(testString)).toBe(resultString);
+});
diff --git a/lib/ExpensiMark.js b/lib/ExpensiMark.js
index 7704e4b7..e5e683b0 100644
--- a/lib/ExpensiMark.js
+++ b/lib/ExpensiMark.js
@@ -1,3 +1,4 @@
+import _ from 'underscore';
import Str from './str';
import TLD_REGEX from './tlds';
@@ -33,7 +34,7 @@ export default class ExpensiMark {
// with the new lines here since they need to be converted into
. And we don't
// want to do this anywhere else since that would break HTML.
// will create styling issues so use
- replacement: (match, _, textWithinFences) => {
+ replacement: (match, __, textWithinFences) => {
const group = textWithinFences.replace(/(?:(?![\n\r])\s)/g, ' ');
return `${group}`;
},
@@ -170,6 +171,24 @@ export default class ExpensiMark {
* @type {Object[]}
*/
this.htmlToMarkdownRules = [
+ {
+ name: 'Strip Special Tags',
+ regex: /(\n|\r\n)?<\/?(html|body)(?:"[^"]*"|'[^']*'|[^'"><])*>(?![^<]*(<\/pre>|<\/code>))(\n|\r\n)?/gim,
+ replacement: ''
+ },
+
+ // Used to Exclude tags
+ {
+ name: 'exclude',
+ regex: new RegExp(
+ [
+ '<(script|style)(?:"[^"]*"|\'[^\']*\'|[^\'">])*>([\\s\\S]*?)<\\/\\1>',
+ '(?![^<]*(<\\/pre>|<\\/code>))(\n|\r\n)?'
+ ].join(''),
+ 'gim'
+ ),
+ replacement: '',
+ },
{
name: 'newline',
@@ -177,7 +196,7 @@ export default class ExpensiMark {
pre: inputString => inputString.replace('
', '
').replace('
', '
'),
// Include the immediately followed newline as `
\n` should be equal to one \n.
- regex: /])*>(?![^<]*(<\/pre>|<\/code>))\n?/gi,
+ regex: /<])*>(?![^<]*(<\/pre>|<\/code>))\n?/gi,
replacement: '\n'
},
{
@@ -298,7 +317,8 @@ export default class ExpensiMark {
* @returns {String}
*/
htmlToMarkdown(htmlString) {
- let generatedMarkdown = htmlString;
+ let generatedMarkdown = _.unescape(htmlString);
+
this.htmlToMarkdownRules.forEach((rule) => {
// Pre-processes input HTML before applying regex
if (rule.pre) {