[ruby-core:17436] Re: URGENT: Possible fixes for segfaults and vulnerabilities available for review in ruby-talk

Tanaka Akira wrote:
> In article <[e-mail]>,
>   Igal Koshevoy <[e-mail]> writes:
>> [http://redmine.ruby-lang.org/issues/show/199] and uploaded the 
>> following files:
>> 1. wrapper.sh -- Sample commands I'm running to build Ruby and execute 
>> the test suites
>> 2. logs.tar.gz -- The test suite logs for the various programs
>>     
> I read the logs of rubyspec.
>
>
> I think there are bugs in Ruby and RubySpec.
>   
Thank you for the help!

> Ruby bugs:
>   Module#remove_method
>   String#%
>   Iconv
>   
I agree, these seem to be bugs in the current implementation of Ruby and 
these definitely worked earlier.

> RubySpec bugs:
>   BigDecimal
>   Date
>   
I agree. But how should these be fixed? The spec examples say these bugs 
only exist in Ruby versions up to and including p114, but these bugs 
still exist. Should the patchlevel value in these examples be updated to 
p238 or should it be set to a more general value, like :ruby18, to 
indicate that no known version of MRI 1.8 can pass this test?

> RubySpec bugs:
> [...]
>   ERB
>   REXML
>   Singleton#_dump
>   
I think these are bugs in Ruby, not in RubySpec. The code in these 
worked correctly without any guard conditions on my older interpreter, 
therefore if the new version is failing these, something has changed. 
Can you or someone else please review these specs to confirm this?

Just to be clear, REXML has some errors due to the guard conditions 
explained in the "RubySpec bugs" section above. However, some of the 
REXML errors were in examples without guard conditions, therefore I 
think they're Ruby errors.

The above list is all the errors that were flagged for attention in the 
RubySpec logs. Bugs in String#%, Iconv, ERB and REXML bugs are fairly 
important given that these are widely used.

Is there a way to generate code coverage reports for both the C and Ruby 
code being run? This would really provide a better idea of how complete 
RubySpec really is.

I'm delighted that we have RubySpecs to help us. Thank you for the hard 
work: Brian Ford, Federico Builes, Arthur Schreiber, and all the other 
contributors. You're providing a valuable service to the Ruby community. :)

-igal