od: fix bugs in displaying floating-point values
authorPaul Eggert <eggert@cs.ucla.edu>
Thu, 18 Nov 2010 17:50:49 +0000 (09:50 -0800)
committerPaul Eggert <eggert@cs.ucla.edu>
Thu, 18 Nov 2010 17:51:04 +0000 (09:51 -0800)
commita71c22fdf0bae4a4d89327a78ab168b53a9d020f
treeda51160fd14eba21bc2b6fa3c44ce175636de75c
parentcf90729e09572fa575ce7e1ee5a3cb3f513d2efb
od: fix bugs in displaying floating-point values

* NEWS: Describe patch.
* bootstrap.conf (gnulib_modules): Add ftoastr.
* src/od.c: Include ftoastr.h, not float.h.
(FLT_DIG, DBL_DIG): Remove.  No need to verify LDBL_DIG.
(FMT_BYTES_ALLOCATED): No need to worry about floating point now,
since this format is no longer used for floating point.
(PRINT_FIELDS): New macro, with most of the guts of the old PRINT_TYPE.
(PRINT_TYPE): Rewrite to use PRINT_FIELDS.
(PRINT_FLOATTYPE): New macro.  This uses the new functions from
ftoastr.
(print_float, print_double, print_long_double): Reimplement
using PRINT_FLOATTYPE.
(decode_one_format): Calculate field widths based on ftoastr-supplied
macros.
* tests/Makefile.am (TESTS): Add misc/od-float.
* tests/misc/od-float: New file.
NEWS
bootstrap.conf
src/od.c
tests/Makefile.am
tests/misc/od-float [new file with mode: 0755]