# Code Review: More ActionPack fixes

**URL:** <https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856>\
**Category:** IronRuby\
**Created:** [January 19, 2010, 9:45am UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856 "2010-01-19T09:45:22Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 9:45am UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/1 "2010-01-19T09:45:22Z")

</div>

tfpt review “/shelveset:rails;REDMOND\sborde”  
Comment :  
defined? should clear $! of any exception that may have been thrown by  
const\_missing  
Makes NativeMethods.SetEnvironmentVariable check if the call succeeded  
Adds a critical tag for library\socket\basicsocket\send\_spec as it  
hung on my machine

---

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 6:54pm UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/2 "2010-01-19T18:54:28Z")

</div>

Since the send\_spec tag is working on my machine and snap, I’d like to  
try to figure out why it’s failing for you instead of disabling it.

JD

---

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 7:27pm UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/3 "2010-01-19T19:27:50Z")

</div>

It was marked as critical before the spec update (I sent you an email),  
which means we did look into it when it first started failing. I did  
look into the failure (not knowing that it was previously tagged as  
critical) and know what is causing the problem, and it is a race. It  
will fail on your machine and in Snap given enough time.

We should not be untagging such non-deterministic critical and unstable  
failures. Otherwise we end up spending time investigating the failures  
everytime. Do you know if any other critical and unstable tags were  
removed? Would be good to send the list of such removed tags to the team  
after a RubySpec update so folks can keep an eye out for the failures.

---

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 8:14pm UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/4 "2010-01-19T20:14:04Z")

</div>

Criticals should mean that it causes a hang or a termination. Those  
should be immediately apparent for that reason. This should have been  
(and should be) tagged unstable. When I update the tags, I run every  
critical spec, and if it passes, I remove the tag. Unstables, are  
different, after you pointed out the unstable tags, I ran them in a loop  
to see if they failed. I tagged the ones that did, and left the rest.

So, to recap, please change this to an unstable tag, and in the future,  
I will regress it more than one time. Also, I like your idea of sending  
out a changelist on the major tags, so I will do that too in the future.

JD

---

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 8:17pm UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/5 "2010-01-19T20:17:00Z")

</div>

Looks good.

Tomas

---

<div class="post-metadata">

**Author:** ![Shri\_B](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/898d66/32.png) [@Shri\_B](https://www.ruby-forum.com/u/Shri_B)\
**Post date:** [January 19, 2010, 8:17pm UTC](https://www.ruby-forum.com/t/code-review-more-actionpack-fixes/181856/6 "2010-01-19T20:17:35Z")

</div>

Marking it as unstable sounds good.
