@ethomson Apologies for the ping, but is there any chance you could take a look at this? I want to integrate this into a Ruby/RSpec project, which generates a lot of stack traces like:

ActiveRecord::RecordInvalid:
  Validation failed: Name is too long (maximum is 30 characters)
./spec/models/foo_spec.rb:32:in `block (2 levels) in <main>'

Basically all test failures will contain that <main>, and GitHub doesn't handle it well. :(

Actually, it looks like this problem is a bit deeper even. We also have quite verbose test failures, which can include things like diffs between models. This leads to output such as this:

Diff:
-[<Foo:0x00007f1a4d46be00
+[<Foo:0x00007f1a4d46be00
- id: 456,
+ id: 123,
- some_attribute: 'some_expected_value'>
+ some_attribute: 'some_value'>

This renders fine in a triple-backtick code block, but GitHub seems to want to parse it as Markdown inside <pre><code> tags, leading to the whole thing being parsed as a list for some reason. I'm not sure if there's a way to tell GitHub to stop parsing Markdown for a given block, or if this could be converted to use Markdown code blocks instead.

update: I did some digging, and it looks like GitHub-Flavored Markdown won't actually disable the Markdown parser inside <pre> blocks UNLESS they appear at the start of their own line. Right now, this action generates something similar to the following:

<tr><td><pre><code>test
details
here</code></pre></td></tr>

For GitHub to properly parse this, however, it needs to appear as

<tr><td>
<pre><code>test
details
here</code></pre></td></tr>

Thanks for the fix and sorry for the delay!

Also, thanks @BytewaveMLP for the analysis, I've included your suggestion.