JRuby

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: JRuby 1.1.6
• Fix Version/s: JRuby 1.2
• Component/s: Extensions
• Labels:

  None
• Environment:

  Hide

  Ubuntu 8.10
  
  $jruby --version
  jruby 1.1.6 (ruby 1.8.6 patchlevel 114) (2008-12-17 rev 8388) [i386-java]
  
  $ java -version
  java version "1.6.0_0"
  IcedTea6 1.3.1 (6b12-0ubuntu6) Runtime Environment (build 1.6.0_0-b12)
  OpenJDK Server VM (build 1.6.0_0-b12, mixed mode)

  Show

  Ubuntu 8.10 $jruby --version jruby 1.1.6 (ruby 1.8.6 patchlevel 114) (2008-12-17 rev 8388) [i386-java] $ java -version java version "1.6.0_0" IcedTea6 1.3.1 (6b12-0ubuntu6) Runtime Environment (build 1.6.0_0-b12) OpenJDK Server VM (build 1.6.0_0-b12, mixed mode)

http://fisheye.codehaus.org/browse/jruby/trunk/jruby/src/org/jruby/ext/ffi/Util.java

The floatValue method in the above class is restricting the range too much. Float.MIN_VALUE is the smallest positive value possible so all non-positive values will be rejected as well as NaN and +-infinity.

Possible solutions:

• Just cast double to float. This would have the side effect of converting doubles too big or too small (too negative) for a float to +infinity and -infinity respectively, but all values in range should work.
• Change the range to -Float.MAX_VALUE to Float.MAX_VALUE and add special cases for NaN and +-infinity.

• Assignee:

  Unassigned

  Reporter:

  Jay Anderson

• Created:

  06/Jan/09 7:58 PM

  Updated:

  21/Mar/09 1:14 PM

  Resolved:

  09/Jan/09 2:19 AM