Fix partial NaN-parsing problem [was RE: sscanf problem]

Dave Korn dave.korn@artimi.com
Tue Apr 5 17:46:00 GMT 2005


----Original Message----
>From: Dave Korn
>Sent: 04 April 2005 19:35


  OK, nobody's answered yet, so just to incentivize the world at large,
here's a testcase - I hear we like those!  This one drops right into the
newlib/testsuite directory (I invented a new subdir for it, newlib.stdio,
because sscanf is a stdio function), and it fails with current HEAD, and it
passes with the patch I RFC'd yesterday.  HTH!

  NB.  I don't currently have a redhat assign sorted out, but a) the patch
to fix the bug is a trivial <10 liner (discounting the comments), the
testcase is a trivial 2-line modification of Michael Hines' original
testcase from http://cygwin.com/ml/cygwin/2005-04/msg00146.html, and the
.exp file is a trivial 2-line modification of
/src/newlib/testsuite/newlib.stdlib/atexit.exp.


    cheers,
      DaveK
-- 
Can't think of a witty .sigline today....
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nan-parse-bug-testcase.diff
Type: application/octet-stream
Size: 1199 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/newlib/attachments/20050405/a47dc880/attachment.obj>


More information about the Newlib mailing list