bug-textutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

on the behaviour of "sort"


From: Heinz Andernach
Subject: on the behaviour of "sort"
Date: Thu, 22 Feb 2001 13:19:57 -0600

Dear Colleagues,

with surprise I noticed that the "sort" command under Linux RedHat 7 does 
not seem to have the same functionality as it used to have e.g. under SUN 
solaris
or even under Linux RedHat 6.2.  I noticed the problem when trying to sort the
following table 

col  0       1    2  3  4      5  6  7  ...
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
    9004     355 13 46 33.36 -60 07  4.1  1.2  12.84 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     802 13 48 38.84 -60 07 19.6  0.3  11.88 0.31  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     133 13 45 42.82 -60 07  4.4  0.3  14.01 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     490 13 45 44.00 -60 07  6.6  0.3  14.67 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     331 13 46  0.11 -60 07  9.1  0.4  13.78 0.40  0 01 007L  2 
FFTFFFTF 1976.256
    9004     331 13 46  0.04 -60 07  7.5  0.5  12.42 0.37  0 30 02EW  2 
FFTFFFTF 1975.497
    9004    4372 13 46  8.61 -60 07 20.2  0.5  13.67 0.40  0 30 007L  1 
FFTFFFFF 1976.256
    9004    1518 13 45 59.06 -60 07 27.4  0.4  12.76 0.40  0 02 007L  2 
FFTFFFTF 1976.256
    9004    1518 13 45 59.08 -60 07 27.2  0.3  12.21 0.37  0 02 02EW  2 
FFTFFFTF 1975.497
    9004     277 13 49  3.00 -60 07 56.5  1.2  13.72 0.31  0 30 02EW  1 
FFTFFFFF 1975.497
    9004    1722 13 46 33.78 -60 07 45.1  0.3  14.50 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     955 13 45  8.86 -60 07 34.2  0.3  13.52 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
    9004    3358 13 46 33.21 -60 07 52.1  0.3  14.13 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     835 13 47 37.46 -60 08  5.5  1.2  12.67 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     967 13 48 30.21 -60 08 12.7  0.3  12.67 0.31  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     553 13 48 19.11 -60 08 15.8  0.5  10.89 0.34  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     220 13 46 14.58 -60 08  1.6  0.4  14.34 0.40  0 30 007L  2 
FFTFFFTF 1976.256
    9004     220 13 46 14.62 -60 08  1.5  0.3  12.97 0.37  0 00 02EW  2 
FFTFFFTF 1975.497
    9004    2094 13 45 17.26 -60 14 29.6  0.3  13.10 0.37  0 30 02EW  1 
FFTFFFFF 1975.497    
    9004    2156 13 45  1.91 -60 15 21.7  0.3  14.63 0.37  0 30 02EW  1 
FFTFFFFF 1975.497  
    9004    3046 13 45 23.87 -60 35 34.3  0.3  12.84 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     325 13 45  2.42 -60 15 10.3  1.2  13.34 0.40  0 02 007L  2 
FFTFFFTF 1976.256

by columns 2, 3 and 4 (with decreasing priority). The result with sort under 
linux
Redhat 6.2, issuing the command   sort +2    is the following

 *  9004    2094 13 45 17.26 -60 14 29.6  0.3  13.10 0.37  0 30 02EW  1 
FFTFFFFF 1975.497    
 *  9004    2156 13 45  1.91 -60 15 21.7  0.3  14.63 0.37  0 30 02EW  1 
FFTFFFFF 1975.497  
*   9004    3046 13 45 23.87 -60 35 34.3  0.3  12.84 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
*   9004     325 13 45  2.42 -60 15 10.3  1.2  13.34 0.40  0 02 007L  2 
FFTFFFTF 1976.256
    9004     133 13 45 42.82 -60 07  4.4  0.3  14.01 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     490 13 45 44.00 -60 07  6.6  0.3  14.67 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004    1518 13 45 59.06 -60 07 27.4  0.4  12.76 0.40  0 02 007L  2 
FFTFFFTF 1976.256
*   9004    1518 13 45 59.08 -60 07 27.2  0.3  12.21 0.37  0 02 02EW  2 
FFTFFFTF 1975.497
*   9004     955 13 45  8.86 -60 07 34.2  0.3  13.52 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     331 13 46  0.04 -60 07  7.5  0.5  12.42 0.37  0 30 02EW  2 
FFTFFFTF 1975.497
    9004     331 13 46  0.11 -60 07  9.1  0.4  13.78 0.40  0 01 007L  2 
FFTFFFTF 1976.256
    9004     220 13 46 14.58 -60 08  1.6  0.4  14.34 0.40  0 30 007L  2 
FFTFFFTF 1976.256
    9004     220 13 46 14.62 -60 08  1.5  0.3  12.97 0.37  0 00 02EW  2 
FFTFFFTF 1975.497
    9004    3358 13 46 33.21 -60 07 52.1  0.3  14.13 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     355 13 46 33.36 -60 07  4.1  1.2  12.84 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
*   9004    1722 13 46 33.78 -60 07 45.1  0.3  14.50 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
*   9004    4372 13 46  8.61 -60 07 20.2  0.5  13.67 0.40  0 30 007L  1 
FFTFFFFF 1976.256
    9004     835 13 47 37.46 -60 08  5.5  1.2  12.67 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     553 13 48 19.11 -60 08 15.8  0.5  10.89 0.34  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     967 13 48 30.21 -60 08 12.7  0.3  12.67 0.31  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     802 13 48 38.84 -60 07 19.6  0.3  11.88 0.31  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     277 13 49  3.00 -60 07 56.5  1.2  13.72 0.31  0 30 02EW  1 
FFTFFFFF 1975.497

I have marked with asterisk the pairs of lines which do NOT show the
correct numeric order in column 4.   Even when using the -b or -n or even the 
-g 
option the result does not change.   However, using the sort command under 
linux RedHat 6.2 the command "sort +2" gives the correct result:

    9004    2156 13 45  1.91 -60 15 21.7  0.3  14.63 0.37  0 30 02EW  1 
FFTFFFFF 1975.497  
    9004     325 13 45  2.42 -60 15 10.3  1.2  13.34 0.40  0 02 007L  2 
FFTFFFTF 1976.256
    9004     955 13 45  8.86 -60 07 34.2  0.3  13.52 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
    9004    2094 13 45 17.26 -60 14 29.6  0.3  13.10 0.37  0 30 02EW  1 
FFTFFFFF 1975.497    
    9004    3046 13 45 23.87 -60 35 34.3  0.3  12.84 0.37  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     133 13 45 42.82 -60 07  4.4  0.3  14.01 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     490 13 45 44.00 -60 07  6.6  0.3  14.67 0.37  0 30 02EW  1 
FFTFFFFF 1975.497
    9004    1518 13 45 59.06 -60 07 27.4  0.4  12.76 0.40  0 02 007L  2 
FFTFFFTF 1976.256
    9004    1518 13 45 59.08 -60 07 27.2  0.3  12.21 0.37  0 02 02EW  2 
FFTFFFTF 1975.497
    9004     331 13 46  0.04 -60 07  7.5  0.5  12.42 0.37  0 30 02EW  2 
FFTFFFTF 1975.497
    9004     331 13 46  0.11 -60 07  9.1  0.4  13.78 0.40  0 01 007L  2 
FFTFFFTF 1976.256
    9004    4372 13 46  8.61 -60 07 20.2  0.5  13.67 0.40  0 30 007L  1 
FFTFFFFF 1976.256
    9004     220 13 46 14.58 -60 08  1.6  0.4  14.34 0.40  0 30 007L  2 
FFTFFFTF 1976.256
    9004     220 13 46 14.62 -60 08  1.5  0.3  12.97 0.37  0 00 02EW  2 
FFTFFFTF 1975.497
    9004    3358 13 46 33.21 -60 07 52.1  0.3  14.13 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     355 13 46 33.36 -60 07  4.1  1.2  12.84 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004    1722 13 46 33.78 -60 07 45.1  0.3  14.50 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     835 13 47 37.46 -60 08  5.5  1.2  12.67 0.34  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     553 13 48 19.11 -60 08 15.8  0.5  10.89 0.34  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     967 13 48 30.21 -60 08 12.7  0.3  12.67 0.31  0 30 02EW  1 
FFTFFFFF 1975.497
    9004     802 13 48 38.84 -60 07 19.6  0.3  11.88 0.31  0 02 02EW  1 
FFTFFFFF 1975.497
    9004     277 13 49  3.00 -60 07 56.5  1.2  13.72 0.31  0 30 02EW  1 
FFTFFFFF 1975.497

Another example:
the list

A2804
A2667
S  89
A3466
A 533
A 514
A3223
A2881
A2814
A 999
A 193

is sorted (in my opinion "properly") by the "old" sort command (under Redhat 
6.2,
and without any options) as follows:

A 193
A 514
A 533
A 999
A2667
A2804
A2814
A2881
A3223
A3466
S  89

However, under Redhat 7 the result of a simple "sort" is the following:

A 193
A2667
A2804
A2814
A2881
A3223
A3466
A 514
A 533
A 999
S  89

Obviously the blanks between the letter and numeric part are interpreted 
differently
in both versions.

For your reference, my man pages say that the "old" sort and the "new" sort are 
authored both by Mike Haertel. I wonder whether there is a genuine intention
for a difference between the two versions or there is some problem with my
installation?

Many thanks for your help


Heinz Andernach     
Depto. de Astronomia, Univ. Guanajuato  tel: +52-4-732-9548 or 732-9607
Apartado Postal 144                     FAX: +52-4-732-0253 
Guanajuato, C.P. 36000, GTO, Mexico     Email: address@hidden



reply via email to

[Prev in Thread] Current Thread [Next in Thread]