sprintf() question

Hi

The result of Windows is clearly strange.

================ my Linux machine =?good =======================
sessionInfo()
R version 2.9.0 (2009-04-17)
x86_64-pc-linux-gnu

locale:
LC_CTYPE=ja_JP.EUC-JP;LC_NUMERIC=C;LC_TIME=ja_JP.EUC-JP;LC_COLLATE=ja_JP.EUC-JP;
LC_MONETARY=C;LC_MESSAGES=ja_JP.EUC-JP;LC_PAPER=ja_JP.EUC-JP;LC_NAME=C;LC_ADDRES
S=C;LC_TELEPHONE=C;LC_MEASUREMENT=ja_JP.EUC-JP;LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base
sprintf("%a",1:8)
[1] "0x1p+0"   "0x1p+1"   "0x1.8p+1" "0x1p+2"   "0x1.4p+2" "0x1.8p+2" "0x1.cp+2"
[8] "0x1p+3"

================ my Windows machine = OMG ======================
sessionInfo()
R version 2.9.0 (2009-04-17)
i386-pc-mingw32

locale:
LC_COLLATE=Japanese_Japan.932;LC_CTYPE=Japanese_Japan.932;LC_MONETARY=Japanese_Japan.932;LC_NUMERIC=C;LC_TIME=Japanese_Japan.932

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base
sprintf("%a",1:8)
[1] "0x1p+0"            "0x1"               "0x1.8"
"0x1p+4294967294"
[5] "0x1.4p+4294967294" "0x1.8p+4294967294" "0x1.cp+4294967294"
"0x1p+4294967293"

The result improved when I changed handling of uExponent as follows

http://prs.ism.ac.jp/~nakama/working/sprintf_format_a.patch

2009/5/18 Daniel Nordlund <djnordlund at verizon.net>:
-----Original Message-----
From: Ted Harding [mailto:Ted.Harding at manchester.ac.uk]
Sent: Sunday, May 17, 2009 3:32 PM
To: Daniel Nordlund
Cc: r-help at r-project.org
Subject: RE: [R] sprintf() question

On 17-May-09 22:03:19, Daniel Nordlund wrote:
When I type the following, I get results different from what I
expected.

sprintf('%a',3)
[1] "0x1.8"

Shouldn't the result be

[1] "0x1.8p+2"
Well, not "p+2" but "p+1"
  (0x1.8 = 1.1000[2] ; *2 = 11.000[2] = 3[10]) ;
however, I get:

  sprintf('%a',3)
  # [1] "0x1.8p+1"

which is indeed correct.

  R version 2.9.0 (2009-04-17) ## Same as yours
  platform  i486-pc-linux-gnu  ## Different from yours ...

which perhaps suggests that there may be a mis-compilation in the
Windows version.

Ted.

I read through the help ?sprintf and didn't find anything
that changed
my expectation.  What am I misunderstanding?  I am using
R-2.9.0 binary
from CRAN on Windows XP Pro, and my session info is

sessionInfo()
R version 2.9.0 (2009-04-17)
i386-pc-mingw32

locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
States.1252;LC_MONETARY=English_United
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets
methods   base

Thanks for any enlightenment.

Thanks Ted!

Enlightenment is what I asked for, and it is what I got.  I was having a
senior moment I guess.  I was picturing 8 as binary 0100, when obviously it
is binary 1000.  So yes, the required power of 2 is 1, and it is fine with
me that Windows implementation does not display it.  Thanks again.

Dan

Daniel Nordlund
Bothell, WA  USA

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

EI-JI Nakama  <nakama (a) ki.rim.or.jp>
"\u4e2d\u9593\u6804\u6cbb"  <nakama (a) ki.rim.or.jp>
Thank you, incorporated now.

As the help page says, %a is a C99 feature that is not available on 
all platforms.  The *printf in msvcrt.dll is a long way from 
C99-compatible (it dates from ca 10 years ago, but later VC runtimes 
are also incompatible).  Recently MinGW has attempted to supplement 
it, not very comprehensively.

Quite a while ago we switched R for Windows to use the trio emulation 
of *printf, which is intended to be compatible with C99.  This was a 
bug in a bugfix to trio (there have been quite a few bugfixes).

Hi