# Code Review: gems

**URL:** <https://www.ruby-forum.com/t/code-review-gems/184873>\
**Category:** IronRuby\
**Created:** [March 9, 2010, 7:44pm UTC](https://www.ruby-forum.com/t/code-review-gems/184873 "2010-03-09T19:44:13Z")\
**Posts on this page:** 2\
**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:** [March 9, 2010, 7:44pm UTC](https://www.ruby-forum.com/t/code-review-gems/184873/1 "2010-03-09T19:44:13Z")

</div>

tfpt review “/shelveset:gems;REDMOND\sborde”

Comment :  
When installing gems from a remote server that is using MRI, the gem  
information that is  
sent down to the local machine is incorrect since MRI is not able to  
parse the platform  
string of “universal-.net” until a patch is applied to RubyGems in  
Gem::Platform#initialize,  
similar to the way other platforms are handled in that method.  
IronRuby does include  
this patch. For now, we work around the problem in  
Gem::Specification.\_load

Adds a case for “universal-.net” to Gem::Platform#=~. This may not be  
strictly needed to fix the problem above, but is included for  
completeness

Running word\_spec was leaving behind a Word dialog box asking whether  
to save the Word document or not. Fixed the spec to close Word without  
prompting.

---

<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:** [March 9, 2010, 7:47pm UTC](https://www.ruby-forum.com/t/code-review-gems/184873/2 "2010-03-09T19:47:08Z")

</div>

Looks good. We should probably make the same change in the Excel specs,  
but that isn’t pressing.