So I raised a sock and ran a testing bot on it. Here's the start of the testing series.

After some testing, I found out that SE is inserting invisible characters \u200c\u200b every 80 continuous characters (continuous = not separated by space/tab/newline) in the Markdown source. This has confused quite some bots already and generated some work for bot developers to fix their parsing of messages.

Furthermore, the line breaking implementation is wrong........

You see, on my phone it's line-broken wrongly, and my bot showed me what the "rendered" text is:


The source code I used in my bot is rather simple:

@command(str, whole_msg=True)
def whatis(msg, body):
    return '`' + repr(body) + '`'

Can we stop creating line breaks this way? This would better be implemented by CSS.

This has caused some trouble for once more: This message was gracefully broken:


  • @ShadowWizard The same applies to code blocks in comments, so... – iBug is disappointed in SE Mar 13 '18 at 8:01
  • Not anymore. It was fixed a while ago. – Shadow9 Mar 13 '18 at 8:01
  • This is so counterproductive... – iBug is disappointed in SE Mar 13 '18 at 8:17
  • @BhargavRao It's said that Smokey used to do so, but it was later removed for bringing an extra layer of latency. – iBug is disappointed in SE Mar 13 '18 at 8:34
  • Aha, that clarifies my previous message. So it's the new chat layer which screwed up. :p – Bhargav Rao Mar 13 '18 at 8:34

This is fixed. My very last commit to the Stack Overflow codebase before I left two months ago was changing the CSS so it prevents the unwanted overflow, eliminating the need for these characters.

I no longer have permissions to add a tag, but I've flagged the question for someone else who does.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .