Testcases for %a

[Andreas Jaeger]

The problem I reported with mathinline seems to be a bug in printf.

Looking into this I noticed that we don't test %a anywhere.  Here're
two small testcases - now the testsuite fails in tfformat with gcc
2.95.2 for me (2.96 CVS current is fine).

Ulrich, is it ok to commit this?

Andreas

2000-05-22  Andreas Jaeger  <aj@suse.de>

	* stdio-common/tfformat.c (sprint_doubles): Add testcase for %a.

============================================================
Index: stdio-common/tfformat.c
--- stdio-common/tfformat.c	1997/08/04 14:22:43	1.4
+++ stdio-common/tfformat.c	2000/05/22 17:13:58
@@ -4009,6 +4009,8 @@
   {__LINE__, 11.25,			"11.2", "%.1f"},
   {__LINE__, 1.75,			"1.8", "%.1f"},
   {__LINE__, 11.75,			"11.8", "%.1f"},
+  {__LINE__, 16,			"0x1.0p+4", "%.1a"},
+  {__LINE__, 16,			"0x1.00000000000000000000p+4", "%.20a"},
 
   {0 }
 

-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.inka.de