[Vol-tickets] Re: [Volume Algorithm] #1: MSVSv8 compiler warnings

Volume Algorithm coin-trac at coin-or.org
Sun Feb 11 22:26:43 EST 2007


#1: MSVSv8 compiler warnings
-------------------------+--------------------------------------------------
  Reporter:  mgalati     |       Owner:  somebody
      Type:  defect      |      Status:  closed  
  Priority:  minor       |   Milestone:          
 Component:  component1  |     Version:          
Resolution:  invalid     |    Keywords:  warnings
-------------------------+--------------------------------------------------
Changes (by ladanyi):

  * status:  new => closed
  * resolution:  => invalid

Comment:

 Warning 2 is fixed.

 Warning 4/5 are code internal in MSVC's headers, can't do a thing about
 it.

 Warning 1/2: read http://msdn2.microsoft.com/en-
 us/library/8ef0s5kh(VS.80).aspx
 as far as I can tell (from this page a few other related pages) fopen()
 falls into the "filesystem security" case, i.e., fopen_s provides better
 filesystem security. IMHO, they should have fixed fopen and not introduce
 another function named fopen_s. That just breaks compatibility with other
 platforms. In some other cases (like strcpy) they are right, the original
 function specification is inherently insecure, but I don't think fopen is
 like that. For this reason I'll close the ticket as "invalid".

 Obviously, these warnings could be gotten rid of by defining
 _CRT_NONSTDC_NO_WARNINGS. However, that would get rid of warnings where
 MSVC is right about insecurity, which is not desirable. I'm don't know
 what should be done so that there would be warnings only when warranted...

 --Laci

-- 
Ticket URL: <https://projects.coin-or.org/Vol/ticket/1#comment:1>
Volume Algorithm <http://projects.coin-or.org/Vol>
A subgradient algorithm.



More information about the Vol-tickets mailing list