GNU bug report logs - #17196
multibyte: printf: %s counts bytes instead of characters

Previous Next

Full log

Message #11 received at 17196 <at> debbugs.gnu.org (full text, mbox):

From: Pádraig Brady <P <at> draigBrady.com>
To: Jan Novak <jn <at> turbo.sk>
Cc: 17196 <at> debbugs.gnu.org
Subject: Re: bug#17196: UTF-8 printf string formating  problem
Date: Sun, 06 Apr 2014 19:13:21 +0100

On 04/06/2014 11:15 AM, Pádraig Brady wrote:
> On 04/06/2014 12:17 AM, Jan Novak wrote:
>> Hello,
>>
>> printf string format counts bytes instead of chars, which leads to broken output ...
>> (the same problem occurs with bash built in printf)
>>
>>
>> just try this:
>>
>> $ echo $LANG
>> us_US.UTF-8
>>
>>
>> $ printf "|%3s|\n" "a"
>> |  a|
>>
>> $ printf "|%3s|\n" "á"     (char is a-acute)
>> | á|
>>
>> expected output:
>> |  á|
>>
>> Is there some easy solution ?
>>
>> TIA for the answer
> 
> Yes printf follows the C standard which only considers bytes.
> awk does respect characters in width specifiers though:
> 
>   $ awk 'BEGIN{printf "|%3s|\n", "á"}'
>   |  á|

Jan points out to me the the awk solution is not portable
to mawk 1.3.3 at least. I used GNU Awk 3.1.8 above.

Pádraig.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.