http://d.puremagic.com/issues/show_bug.cgi?id=5209

           Summary: posix/sys/select.d: FD_ISSET function should return
                    bool
           Product: D
           Version: D2
          Platform: x86_64
        OS/Version: Linux
            Status: NEW
          Severity: major
          Priority: P2
         Component: druntime
        AssignedTo: sean@invisibleduck.org
        ReportedBy: ibuclaw@ubuntu.com


--- Comment #0 from Iain Buclaw <ibuclaw@ubuntu.com> 2010-11-13 06:17:34 PST ---
This is how FD_ISSET is defined on Linux

#define __FD_ISSET(d, set) \
  ((__FDS_BITS (set)[__FDELT (d)] & __FDMASK (d)) != 0)


So druntime should do the same (patch supplied). This is also a blocker for 64bit, where you would get an implicit cast from long to int message.

Regards

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------

http://d.puremagic.com/issues/show_bug.cgi?id=5209



--- Comment #1 from Iain Buclaw <ibuclaw@ubuntu.com> 2010-11-13 06:18:41 PST ---
Created an attachment (id=809)
patch for 5209

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------

http://d.puremagic.com/issues/show_bug.cgi?id=5209


Iain Buclaw <ibuclaw@ubuntu.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|major                       |critical


--- Comment #2 from Iain Buclaw <ibuclaw@ubuntu.com> 2010-12-21 13:33:04 PST ---
Bumping priority:

It appears *ONE* of the functions has been changed to a boolean. But what's utterly worse is that:


extern (D) bool FD_ISSET( int fd, fd_set* fdset )
{
    return (fdset.fds_bits[__FDELT( fd )] & __FDMASK( fd )) == 0;
}

should be "!= 0"

Regards

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------

http://d.puremagic.com/issues/show_bug.cgi?id=5209


Brad Roberts <braddr@puremagic.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |braddr@puremagic.com


--- Comment #3 from Brad Roberts <braddr@puremagic.com> 2010-12-21 13:43:47 PST ---
Would you do me a favor and write up a block of unit tests for those functions?
 I'll apply both together.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------

http://d.puremagic.com/issues/show_bug.cgi?id=5209



--- Comment #4 from Iain Buclaw <ibuclaw@ubuntu.com> 2010-12-21 16:33:12 PST ---
Created an attachment (id=852)
unittest for sys.select

Well the most obvious thing it affects is std.socket;


import std.socket;

unittest
{
    SocketSet sset = new SocketSet(24);

    // All socket descriptors should not be set before using them.
    for (socket_t i = 0; i < sset.max; i++)
    {
        assert(! sset.isSet(i));
    }
}


Is the attached file OK? :~)

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------

http://d.puremagic.com/issues/show_bug.cgi?id=5209


Brad Roberts <braddr@puremagic.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
         AssignedTo|sean@invisibleduck.org      |braddr@puremagic.com


--- Comment #5 from Brad Roberts <braddr@puremagic.com> 2010-12-21 16:54:21 PST ---
Checked in both the fix and the unittests in r453 and r454.

Thanks for the patches.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------