Compare 2 files in Unix

Dev Shed Forums Sponsor:

August 9th, 2011, 07:55 AM

dilip_d21

Registered User

Join Date: Aug 2011

Posts: 2

Time spent in forums: 49 m 3 sec
Reputation Power: 0

Compare 2 files in Unix

Hi Linux gurus,

Help need to compare 2 files .

File 1 :

DB_NAME FIRST_ACTIVE_LOG DBPARTITIONNUM
-------- -------------------- --------------
BP1 279231 0
BP1 12735 1
BP1 12734 2
BP1 12735 3
BP1 12616 4
BP1 12730 5
BP1 12810 6
BP1 12602 7
BP1 12669 8
BP1 12690 9
BP1 12631 10
BP1 9850 11
BP1 9842 12
BP1 9912 13
BP1 9792 14
BP1 6185 15
BP1 6194 16
BP1 6164 17
BP1 6154 18
BP1 6164 19
BP1 6147 20

File 2 :

Rollforward Status

Input database alias = bp1
Number of nodes have returned status = 21

Node number Rollforward Next log Log files processed Last committed transaction
status to be read
----------- -------------------------- ------------------- ------------------------- --------------------------
0 DB working S0278273.LOG S0278240.LOG-S0278272.LOG 2011-08-03-05.00.20.000000 UTC
1 DB working S0011751.LOG S0011717.LOG-S0011750.LOG 2011-08-02-23.11.53.000000 UTC
2 DB working S0011753.LOG S0011719.LOG-S0011752.LOG 2011-08-02-23.11.56.000000 UTC
3 DB working S0011731.LOG S0011696.LOG-S0011730.LOG 2011-08-02-23.11.50.000000 UTC
4 DB working S0011645.LOG S0011611.LOG-S0011644.LOG 2011-08-02-23.11.57.000000 UTC
5 DB working S0011747.LOG S0011713.LOG-S0011746.LOG 2011-08-02-23.11.51.000000 UTC
6 DB working S0011814.LOG S0011779.LOG-S0011813.LOG 2011-08-02-23.11.50.000000 UTC
7 DB working S0011629.LOG S0011595.LOG-S0011628.LOG 2011-08-02-23.11.55.000000 UTC
8 DB working S0011692.LOG S0011658.LOG-S0011691.LOG 2011-08-02-23.11.56.000000 UTC
9 DB working S0011704.LOG S0011670.LOG-S0011703.LOG 2011-08-02-23.11.51.000000 UTC
10 DB working S0011650.LOG S0011616.LOG-S0011649.LOG 2011-08-02-23.11.55.000000 UTC
11 DB working S0009569.LOG S0009560.LOG-S0009568.LOG 2011-08-02-23.11.23.000000 UTC
12 DB working S0009560.LOG S0009552.LOG-S0009559.LOG 2011-08-02-23.10.23.000000 UTC
13 DB working S0009630.LOG S0009621.LOG-S0009629.LOG 2011-08-02-23.11.19.000000 UTC
14 DB working S0009511.LOG S0009502.LOG-S0009510.LOG 2011-08-02-23.11.23.000000 UTC
15 DB working S0005904.LOG S0005895.LOG-S0005903.LOG 2011-08-02-23.11.19.000000 UTC
16 DB working S0005913.LOG S0005904.LOG-S0005912.LOG 2011-08-02-23.11.19.000000 UTC
17 DB working S0005884.LOG S0005875.LOG-S0005883.LOG 2011-08-02-23.11.21.000000 UTC
18 DB working S0005874.LOG S0005865.LOG-S0005873.LOG 2011-08-02-23.11.23.000000 UTC
19 DB working S0005884.LOG S0005875.LOG-S0005883.LOG 2011-08-02-23.11.20.000000 UTC
20 DB working S0005867.LOG S0005858.LOG-S0005866.LOG 2011-08-02-23.11.22.000000 UTC

I need to compare file 1 (DBPARTITIONNUM) column with file 2 ( Node number ) column as a first step and then

Match the FIRST_ACTIVE_LOG column of file 1 with Next log to be read column in file 2 (integer value alone ) and then print the result ..

Result should be like this

If
FIRST_ACTIVE_LOG == Next log to be read

then

Log Gap is zero for Node $DBPARTITIONNUM

example:

Log Gap is zero for Node 0

else

Log gap is (FIRST_ACTIVE_LOG minus Next log to be read ) for node $DBPARTITIONNUM

Example

Log gap is 100 for node 0

Please help

Thanks in advance

Dilip

August 9th, 2011, 02:31 PM

SimonJM

Contributing User

Join Date: Mar 2006

Posts: 2,108

Time spent in forums: 1 Month 1 Week 1 Day 4 h 41 m
Reputation Power: 1485

Ok, so each file has, in effect, 21 lines (in this example) of actual data.
You want to match data based on column 3 (of 3) from the first file to column 1 (of 7) in the second file (columns based on white space delimiter).
So, taking the first rows of data:
BP1 279231 0
would match:
0 DB working S0278273.LOG S0278240.LOG-S0278272.LOG 2011-08-03-05.00.20.000000 UTC
based on the values in red.
Having matched those rows you wish to check column 2 of the first file against the numeric part of the file name in column 4 of the second file:
BP1 279231 0
against:
0 DB working S0278273.LOG S0278240.LOG-S0278272.LOG 2011-08-03-05.00.20.000000 UTC
So you'd be comparing 279231 against 278272.
Is that correct?

__________________
The moon on the one hand, the dawn on the other:
The moon is my sister, the dawn is my brother.
The moon on my left and the dawn on my right.
My brother, good morning: my sister, good night.
-- Hilaire Belloc

August 10th, 2011, 11:51 AM

dilip_d21

Registered User

Join Date: Aug 2011

Posts: 2

Time spent in forums: 49 m 3 sec
Reputation Power: 0

Thanks Simon for the reply , i am absolutely looking for the same

August 10th, 2011, 05:06 PM

SimonJM

Contributing User

Join Date: Mar 2006

Posts: 2,108

Time spent in forums: 1 Month 1 Week 1 Day 4 h 41 m
Reputation Power: 1485

Ok, there are, of course, this is *nix we are talking about, lots of ways of doing this. The method I'd pick (and will have a go at showing when I get a moment) will be to, in effect, merge the two files, maybe just 'extracting' the required columns and placing that data in a temporary file and processing that in some way or another - either in awk or in a loop in a script. We shall see ...

August 11th, 2011, 07:34 AM

SimonJM

Contributing User

Join Date: Mar 2006

Posts: 2,108

Time spent in forums: 1 Month 1 Week 1 Day 4 h 41 m
Reputation Power: 1485

Ok, here we go - do a man of the various commands used (sed, awk, join) for details:

Code:

#!/bin/bash
#
# Extract only the data we want from each file
sed '1,/---/d' f1.txt | awk '{print $3, $2}' > base_f1.txt
sed '1,/---/d' f2.txt | awk '{print $1, $4}' | tr -d "[:alpha:]\." > base_f2.txt

# Now merge the two files to allow easier processing
#  The default settings will work fine for us
join base_f1.txt base_f2.txt > base_data.txt

# And now process the data
awk '{gap=$2-$3; printf("Log Gap is %d for Node: %d\n",gap,$1)}' base_data.txt

Should add that my f1.txt and f2.txt are your two files as shown above!

Last edited by SimonJM : August 11th, 2011 at 07:42 AM.

Compare 2 files in Unix

Developer Shed Advertisers and Affiliates