burkeen / pagedown Goto Github PK

What steps will reproduce the problem?

Attempt to enter an h4, h5, or h6 like so:

#### Test
##### Test
###### Test

What is the expected output? What do you see instead?

Expect to convert to <h4>Test</h4> for example. Converts to unmarked-up text 
instead.


What version of the product are you using? On what operating system?

Using trunk code in Chrome 19, Mac OS X Snow Leopard


Please provide any additional information below.

What steps will reproduce the problem?
1. Enter three times the hash sign #
2. the output is no h3-tag


What is the expected output? What do you see instead?
Three # should generate a h3 heading (<h3> my headline </h3>
According to http://daringfireball.net/projects/markdown/syntax#header markdown 
should support html headers h1 down to h6 be entering n times the # sign.

What version of the product are you using? On what operating system?
The current version (july 2012).

Please provide any additional information below.
Take a look at the markdown page at daringfireball.net

PHP Markdown Extra (http://michelf.com/projects/php-markdown/extra/) is a great 
extension to markdown which supports table, definition list, header id and 
abbreviation, tables and definition lists are especially usefull when writing 
blogs and other types of documents, so it would be nice to support it.

When clicking the image button, the wmd-prompt-background appears. If the image 
insert is canceled using callback(null); in JavaScript, then the background 
disappears. However, if the user clicks the image button again, the 
wmd-prompt-background will appear again, but whether callback(null); or 
callback("http://someurlhere.com/photo.jpg"); is used, the background will not 
disappear.

The JavaScript Console shows at this time "Uncaught TypeError: Cannot call 
method 'removeChild' of null" in Markdown.Editor.js line 1684. 

I have attached the trace that the browser's console provides. The browser 
version is Google Chrome 20.0.1132.57 m on Windows XP.

Hi,

I'm in the process of developing a website in french and I was wondering if you 
already thought of the Localization of the different parts of the editor ?

I'm willing to do the job if we can agree on the architecture behind the 
changes.

Thanks for the great work !

A-Z, a-z, 0-9 and _ are in \w

How about publishing an npm module for nodejs usage? I could fork this project 
on github and publish an npm module myself, but it would be great if you 
consider adding one.

Thanks
Subbu

Adds an optional second argument to linkEnteredCallback, allowing alt text to 
be specified instead of using the default.

The bracket-escaping is moved to the bottom of the function so it is applied to 
the alt text.

Don't working from an IFrame.

Example usage - Facebook Canvas application.

What steps will reproduce the problem?
1. write 
[14]:http://blog.thecanadaline.com/wp-content/uploads/2012/03/IMG_62441-225x300.
jpg ()\n\n
2. try to convert
3.

What is the expected output? What do you see instead?
we are expecting to have the link correctly coded inside the page, instead we 
have a broken link and the unconverted string appears at the end of the page as 
text: 
[14]:http://blog.thecanadaline.com/wp-content/uploads/2012/03/IMG_62441-225x300.
jpg ()\n\n

We should be able to furnish empty parameters in brackets without breaking the 
behaviour of the app...

What version of the product are you using? On what operating system?


Please provide any additional information below.
Removing the empty braces fixes the markdown.

What steps will reproduce the problem?
1. Go to demo http://pagedown.googlecode.com/hg/demo/browser/demo.html
2. Add this text to any of the textboxes:
"http://google.com
http://google.com
text between links
http://google.com
"

What is the expected output? What do you see instead?
Expected:
"http://google.com http://google.com text between links http://google.com"
And all three links are clickable(so transformed to links)

Instead:
The second http://google.com is not transformed to a link.


What version of the product are you using? On what operating system?
Current live web demo.

Please provide any additional information below.

Footnotes are important!

When it comes to specs, I recommend conforming to what's becoming a sort of 
standard in markdown libraries:

http://freewisdom.org/projects/python-markdown/Footnotes
http://michelf.com/projects/php-markdown/extra/#footnotes

I hope user can choose and upload image file.

but how can I register my own insertImageDialog to the editor?

Thank you very much!

Chance

Fairly simple.
Looking at the in-browser example for the Markdown.Sanitizer.js...

document.write(converter.makeHtml("<script>alert(42);</script><b>bold"); // 
creates "alert(42);bold"

you can find that there is only one closing bracket before the ;

I noticed that the four JavaScript files don't have a license header. This 
seems to be a common practice amongst JavaScript libraries out there. Would it 
be possible to add the information to the files, perhaps similar to jQuery's 
text?

( https://github.com/jquery/jquery/blob/master/src/intro.js )

Typing either the letter a for a word beginning with a (caps or not) causes 
<em>a</em>.

See attached file for example.

What steps will reproduce the problem?
1. having a sting like August,29th_to:_ August,30th
2. there is no space before the leading "_"
3.

What is the expected output? What do you see instead?
We expect to see the bolded "to:", we have _to:_ instead

What version of the product are you using? On what operating system?


Please provide any additional information below.

What steps will reproduce the problem?
1. Try to convert this a string like  _From: _July 30th, 2012 _to _August 5th, 
2012 

2. There are spaces after the string to be emphasized: _From: _
3.

What is the expected output? What do you see instead?
we would expect this:
<em>From: </em>July 30th, 2012 <em>to </em>August 5th, 2012
instead we have the input again:
_From: _July 30th, 2012 _to _August 5th, 2012 

What version of the product are you using? On what operating system?


Please provide any additional information below.

The space at the end seem to break the converter functionality

What steps will reproduce the problem?
1. use a tool like Chirpy to package all three Markdown.*.js files together: 
Converter, Sanitizer, and Editor.
2. include the javascript file in a page


What is the expected output? What do you see instead?

I'd expect the page to behave the same way as if I included the javascript 
files individually. Instead, it has a javascript error because Sanitizer ends 
with a function call that doesn't return anything, and Editor begins with a 
parenthesis, which makes the javascript think you're calling a function that 
got returned from the Sanitizer's initialization function.

You can fix this issue by adding a semicolon at the end of the Sanitizer.js 
file, as shown in the included Hg patch file.

This issue was already filed at the old 'wmd-new' repo.

The issue is still valid. I'm copypasting this as I just run into the very same 
problem.

What steps will reproduce the problem?
1. Type a list with each step on its own line

one
two
three

2. Highlight list and click the numbered list button
3. Rage at the resulting list

  1. one two three

What is the expected output? What do you see instead?

I want this:

one
two
three

to look like this:

 1. one
 2. two
 3. three


What version of the product are you using? On what operating system?

unknown version in chrome/ie

Please provide any additional information below.

The type-list/highlight/format-list pattern is common and any normal editor 
allows you to do this without deformatting your list and placing multiple lines 
on a single line.  Stop doing this, please.

What steps will reproduce the problem?
1. Try a URL like http://derp.com?__test=1

What is the expected output? What do you see instead?

A link with the proper not a url. Instead there's an <em> in the url.

When building a bundle using some requireje-based tools each file is loaded in 
a separate closure, with the result that Markdown.Convertor creates a Markdown 
in a different scope than window and Markdown.Sanitizer then fails to find it. 
This can be fixed by modifying the converter to explicitly create 
window.Markdown as well:

diff --git a/src/3rdparty/pagedown/Markdown.Converter.js 
b/src/3rdparty/pagedown/Markdown.Converter.js
index 58fc54a..8c1a909 100755
--- a/src/3rdparty/pagedown/Markdown.Converter.js
+++ b/src/3rdparty/pagedown/Markdown.Converter.js
@@ -3,7 +3,7 @@ var Markdown;
 if (typeof exports === "object" && typeof require === "function") // we're in a CommonJS (e.g. Node.js) module
     Markdown = exports;
 else
-    Markdown = {};
+    Markdown = window.Markdown = {};

 // The following text is included for historical reasons, but should
 // be taken with a pinch of salt; it's not all true anymore.

We are trying to use Markdown.Sanitizer.js in Acre (a server-side javascript 
environement).

However, we find ourselves hacking a window object to use Markdown.Sanitizer.js.

But if Markdown.Sanitizer.js passes "this" as the "window" variable, we can use 
Markdown.Sanitizer.js on the server-side as is:

(function(window) {

...

})(this);

When empty line appears in code block (marked with triple backqoute) it breaks 
to simple paragraphs with "```" after <p> and before </p>.

What steps will reproduce the problem?
1. I start the Browser Demo.html 
2. I click the Hyperlink icon
3. I insert 'http://example.com/ "optional title"' as shown in example

What is the expected output? What do you see instead?

Foutdetails webpagina

Gebruikersagent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; 
Trident/4.0; .NET CLR 1.1.4322; OfficeLiveConnector.1.4; OfficeLivePatch.1.3; 
.NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; .NET4.0C)
Tijdstempel: Tue, 14 Aug 2012 21:41:20 UTC


Bericht: '$' is niet gedefinieerd
Regel: 1647
Teken: 17
Code: 0
URI: file:///C:/pl/pagedown/Markdown.Editor.js



What version of the product are you using? On what operating system?

IE8, pagedown Version 3151a581819f39123b0b5fd1ca9e26cebdcaa873

Please provide any additional information below.

I use pagedown also in comment-form it has well keyboard shortcuts support, 
it's very useful to format small code snippets against typing 4 spaces every 
single line. But I have to put button bar on dom also. Of course I can hide it 
with css but that's not on true-way.

There are some other issues I want to post if you don't mind. Actually for now 
there is no any other editor to be used with markdown except markitup — which 
as to me very is very weird and do everything but nothing well.

See attached diff.

In order to get this library updated on cdnjs, someone needs to submit a pull 
request which follows the rules at https://github.com/cdnjs/cdnjs  A very old 
version of this library is already on cdnjs as 1.0, but I can't find a proper 
version number for pagedown, so can't update that version number to the correct 
value when supplying the latest source.

The fakeButton variable used to handle the shift+enter indent is missing a var 
statement.

Patched at 
https://code.google.com/r/tmslft-pagedown/source/detail?r=785c6f8483248f084d0879
3da9e1cac5e2037ec8

The syntax [link](test.txt) does not work.  Instead, I see "link" (the word), 
but it is not made into a link for me.

Links could be improved to open in a new tab (it currently opens up in the same 
page).

Adding the target="_blank" attribute would fix it.

What steps will reproduce the problem?
1. Open /browser/demo.html
2. Input the following text into the editor:

This is a bug.

    here is a code block, properly monospaced

But if I start a list,

* even if it only has one item

    and then I try to make a code block, it doesn't work.

On the other hand, if I start another normal paragraph first,

    code blocks work again



What is the expected output? What do you see instead?

"    and then I try to make a code block, it doesn't work." should be displayed 
as a code block. Instead it is included as part of the list element.

What version of the product are you using? On what operating system?

changeset:   21:44a4db795617
tag:         tip
user:        balpha
date:        Fri Mar 02 14:55:03 2012 +0100

Fixes an XSS vulnerability reported by Mario Heiderich 
(http://twitter.com/0x6d6172696f).

This change ensures that…

   new Markdown.Converter().makeHtml('**_[Free iPad Here!](javascript:alert(1))_**')

…has the same result as…

   new Markdown.Converter().makeHtml('**_[Free iPad Here!](javascript:alert(1))_**')

> new Markdown.Converter().makeHtml('[<svg/onload=alert(1)//]()')
→ '<p><a href=""><svg/onload=alert(1)//</a></p>'

This displays an alert in Firefox (XSS). Demo: 
data:text/html;charset=utf-8,<p><a href=""><svg/onload=alert('XSS')//</a></p>

Expected output is:

→ '<p><a href="">&lt;svg/onload=alert(1)//</a></p>'

I.e. the `<` should always be escaped.

Supporting asynchronous module definition [1] would allow to use 
CommonJS-compliant asynchronous javascript loaders like requirejs [2] on the 
client side.

This also allows to declare dependencies between modules, and to ensure that 
some module is loaded before an other.

This would look like this for Markdown.Converter:

``` Javascript
define(function() {
    // return Converter here
});

Original issue reported on code.google.com by `arnaud.lb` on 4 Aug 2011 at 5:09

If there was a live preview site for pagedown,
I could test out a bug on the current code.
Right now I can only report a bug against the current version in use somewhere, 
which might not be the current version.

I use the editor within a 'modal' dialog within the page (using colorbox). When 
someone closes the dialog, I'd like to be able to cleanup after the editor has 
been shown so that if they re-open the dialog, everything is fresh again. 

This means unbinding any events and removing elements like the wmd-button-bar. 
Would it be possible to add a editor.close() function?

Hi,

I implemented a shared settings object which allows you to tweak the editor at 
creation time without editing the code.

I also added a handleUndo switch which allows you to turn off the undo 
functionality which I needed for my project with shareJS.

The change is at 
http://code.google.com/r/woutmertens-noundo/source/detail?r=52c43623e71b1c912d68
4bd4b513c9fe1cc23c09

What steps will reproduce the problem?
1. Go to demo http://pagedown.googlecode.com/hg/demo/browser/demo.html
2. Add this text to the unsanitized text box:
      "See my [About](about/) page for details."
    Example like: http://daringfireball.net/projects/markdown/syntax#link

What is the expected output? What do you see instead?
Expected the link to be to "about/", but it was to 
http://pagedown.googlecode.com/hg/demo/browser/%2aabout%2a/
Is my markdown syntax right? 

What version of the product are you using? On what operating system?

Current live web demo on current Chrome OSX.

Please provide any additional information below.

i cloned pagedown an ported the AutoNewLines option form MarkDownSharp to 
PageDown.

it could be usefull for other people as well, plz pull (and review) if you got 
the time.

http://code.google.com/r/marcdrexel-pagedown/

Usage of WAI-ARIA and correct HTML would help ensure a slightly better 
experience.

For example using a <button> opposed to a <span> and using roles using as 
"application" and "menu" or "button" to the inserted mark-up shouldn't be that 
tricky to implement.

Am happy to hack around and provide more concrete examples if needed...

What steps will reproduce the problem?
1. var converter = Markdown.getSanitizingConverter();
2. converter.makeHtml( '#h4-header')
3. the text 'h4-header' is returned.

What is the expected output? What do you see instead?
I expected '<h4>h4-header</h4>' to be returned

What version of the product are you using? On what operating system?
Latest version, OS X 10.7


Please provide any additional information below.

The issue is that on line 23 of Markdown-Sanitizer.js a whitelist is declared 
that only allows h tags up to h3.  However, line 747 of the 
Markdown-Converter.js allows h1-h6. Adding h4-h6 to the whitelist fixed the 
issues.

What steps will reproduce the problem?
1. create some markdown with a link to send an email like 
[mail]([email protected])
2. call the converter on the string
string = '[mail](mailto:[email protected])'
converter.makeHtml(string);
3.

What is the expected output? What do you see instead?
The output should be 
<a href="mailto:[email protected]">mail</a>

instead we have
<a href="{absoluteURL}mailto%[email protected]">mail</a>

What version of the product are you using? On what operating system?


Please provide any additional information below.

This has been fixed by tweaking one of the attackLab functions (below), to 
avoid encoding the ":" when preceeded by "mailto".

function encodeProblemUrlChars(url) {
   if (!url)
     return "";
//if we have an "email:" type of link then do not encode the ":"
   else if (/mailto:/.test(url))
     return url;
   else {
     var len = url.length;
     return url.replace(_problemUrlChars, function (match, offset) {
    if (match == "~D") // escape for dollar
          return "%24";
    if (match == ":") { 
      if (offset == len - 1 || /[0-9\/]/.test(url.charAt(offset + 1)))
        return ":"
    } 
    return "%" + match.charCodeAt(0).toString(16);
     });
  }
}

~~text~~

should replace to

<del>text</del>

It would be great if the Markdown.Converter had a method which only created 
links. I'm dealing with feed posts and I don't want the user to be able to 
create headers and all that jazz.

Type this:
--------
Example:

 1. Number One
 2. Then Number Two


--------

on last second blank line press Ctr+U and you'll get this:
--------
Example:

 - Number One
 - Then Number Two
 - List item
--------

Chrome 17, Ubuntu
UPD: pressing Ctrl+U converts every list before to bulleted, and Ctrl+O 
converts them all to numbered

I get:

![![enter image description here][5]][5]

What steps will reproduce the problem?

1. made a page much like the demopage
2. initialized it like the following

<script type="text/javascript">
    function my_dialog(callback){
        alert ("hello");
        var retval = callback("http://xyz.png/x.png");
        return true
    }
    var converter1 = Markdown.getSanitizingConverter();
    var editor1 = new Markdown.Editor(converter1);
    editor1.hooks.set("insertImageDialog",my_dialog)
    editor1.run();
</script> 

3. pressed the image link

What is the expected output? What do you see instead?
![enter image description here][5]

What version of the product are you using? On what operating system?
N/A

Please provide any additional information below.

I can make it work if commenting out the 
                postProcessing();


inside the linkEnteredCallback function

I'd like to be able to set rel=nofollow on rendered links.

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=96569

thanks.

> Markdown.Editor.js:327
> Uncaught SyntaxError: Invalid flags supplied to RegExp constructor 'sfsdf 
sdfs dfsdf sdf sf'

How i got it: 
- type a list of words and create a new blank line at last
- select every line (except the blank one). The last non-blank one should be 
hilighted to the end of line — not end of last word in line.
- press Numbered or Bulleted list via shortcut or toolbar button

Since my English is not as well as I want so, attaching a screenshot :)

Allow configuring which buttons are visible and functional and cleaned up the 
button creation code a bit. Changes are backward compatible with the existing 
custom "help" button configuration.