[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: seq bug? LAST equals 1,000,000
From: |
Bob Proulx |
Subject: |
Re: seq bug? LAST equals 1,000,000 |
Date: |
Thu, 20 Aug 2009 09:53:15 -0600 |
User-agent: |
Mutt/1.5.18 (2008-05-17) |
db_23 wrote:
> When LAST is equal to 1,000,000 (seq 10000000). I got the outcome as
> the following. Numbers after 999,999 appears more than once, but not
> at fixed times.
Thank you for the report. Unfortunately seq uses floating point
numbers internally. This has caused some problems in the resulting
implementation.
> 999997
> 999998
> 999999
> 1e+06
> 1e+06
> 1e+06
I can reproduce your problem on seq 5.97 in an old Debian Etch
machine. However your seq 5.2.1 was released in 2004 and now five
years later there have been many improvements and bug fixes applied
since then. So I think upstream this has already been fixed in later
versions for some time.
In any case this works for me with 6.10. See this output.
$ seq 9999997 10000005
9999997
9999998
9999999
10000000
10000001
10000002
10000003
10000004
10000005
> #seq --version
> seq (coreutils) 5.2.1
$ seq --version
seq (GNU coreutils) 6.10
Bob
- seq bug? LAST equals 1,000,000, db_23, 2009/08/20
- Re: seq bug? LAST equals 1,000,000, Eric Blake, 2009/08/20
- Re: seq bug? LAST equals 1,000,000,
Bob Proulx <=
- Re:Re: seq bug? LAST equals 1,000,000, db_23, 2009/08/21