Also occurs when there is nbsp (u+a0) at the end of the line.

<div></div> 

Google is eaten

It seems only block elements do this. Inline elems are unaffected. Only Parsedown Extra is affected.

I can confirm that this is a serious issue. What makes this a particularly nasty bug is that it results in a complete loss of the data as it was originally intended.

<div>1</div><p>2</p>
The p tag (and contents), along with this line are eaten.

I left a small donation hoping that it helps others resolve this.

I'll see what I can do. It might take some time, though. I'll be quite busy over the next couple of weeks.

It might take some time, though.

Copy that :-) I'm a veteran with PHP, but new to Parsedown. Any ideas where I might start debugging this? If so I might be able to submit a PR for this.

Sure,

It's in the blockMarkup methods in Parsedown.php.

Parsedown does supports one line block-level elements, but it looks for a closing tag at the end of the line. It doesn't consider the element closed when it doesn't find a closing tag at the end of the line.

You could give it a shot, but it might take me much less time to fix. Unless it's urgent, it might make more sense to wait.

@erusev writes...

blockMarkup methods in Parsedown.php

Thank you :-) I'll take a look at this in the next day or so and see what I can make of it. Otherwise, I'll wait patiently as suggested.

I noticed an issue with a self closing element which ended in an endless loop. If an element doesn't have any children, the html should be appended directly and not through processText().

I've fixed this in:
storeman@ddd0d0b

@storeman Can you provide an example that I can use to reproduce the issue?

First,I have to say it only occurs in the fix from @acmitch. The block element problem seems to be resolved.

Just process '<iframe />' gave me this issue.
I wanted to notify you about this, in case you wanted to merge the fix from @acmitch.

@storeman Ah, I see, thanks, I appreciate it.

Had more issues today with the solution from @acmitch. I deleted my fork because i could not find the clear edge case where his solution was breaking my html. One of the issues was, textareas were replicated and the dom-structure was altered. For example:

<textarea></textarea>
<label><input type="checkbox"> <a href="#">Some test</a></label>

Became

<textarea><textarea></textarea></textarea>
<label><input type="checkbox"></label> <a href="#">Some test</a>

It was caused by the surrounding html, but those 13000 lines were to hard to walk through.

@storeman thanks for the feedback! Found a few issues related to the code block you provided.

<textarea></textarea>
<label><input type="checkbox"> <a href="#">Some test</a></label>

The first issue, as you mentioned above, was that specific elements were causing endless loops. This has been fixed in my latest commit acmitch@468ecc6

It is important to note that all inline or voided HTML elements must be within the $this->textLevelElements and $this->voidElements arrays located within the parsedown library for my fix to work properly.

Therefore, to fix the textarea issue you must add the textarea value to the $this->voidElements array, like such:

 protected $voidElements = array(
    'area', 'base', 'br', 'col', 'command', 'embed', 'hr', 'img', 'input', 'link', 'meta', 'param', 'source', 'textarea'
 );

Once, that has been added you should be good to go (I've already tested and works like a charm)! Maybe @erusev can add the textarea permanently if he gets a chance. FYI: I plan to do a little more work to this library over the next few months. I can't seem to get the script tags working as well as a few small things. I will likely have to change the saveXML back to saveHTML once my PHP is updated to a newer version. You might even want to make that change if your PHP version is 5.3.6 or greater.

Here is a failing testcase if it would be of help:

public function testStripping()
{
    $expectedMarkup = '<p><strong>Contact Method:</strong> email</p><p>Test</p><p><em>Some italic text.</em></p>';
    $actualMarkup = (new ParsedownExtra())->text($expectedMarkup);

    $this->assertEquals($expectedMarkup, $actualMarkup);
}

Results:

1) ParsedownExtraTest::testStripping
Failed asserting that two strings are equal.
--- Expected
+++ Actual
@@ @@
-'<p><strong>Contact Method:</strong> email</p><p>Test</p><p><em>Some italic text.</em></p>'
+'<p><strong>Contact Method:</strong> email</p>'

@ericbarnes Thanks, Eric, I appreciate it, hopefully, I'll have some time later this month to address the issue.