07-02-2008
Record count problem using sed command
Hi,
I have a script which removes 2 header records and 1 trailer record in a list of files. The commands doing the actions are
sed '1,2d' $file > tempfile1.dat
sed '$d' < tempfile1.dat > $output.txt
Its working fine for all records except a file having size=1445509814 and number of records=13152342. The output has only 5297671 records instead of being just 3 less than the original count.
Is there any limitation on the number of records that the sed command can process?
Some details-
>uname -a
SunOS xx 5.10 SunOS_Development sun4u sparc SUNW,Sun-Fire
>df -k .
Filesystem kbytes used avail capacity Mounted on
xx 67091946 64377506 2043521 97% xx
>df -k /tmp
Filesystem kbytes used avail capacity Mounted on
swap 17337536 1504840 15832696 9% xx
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
How do I get the record count in an EBCDIC file on a Linux Box. :confused: (1 Reply)
Discussion started by: oracle8
1 Replies
2. Shell Programming and Scripting
Hi,
In my Scripts,i need to read record from a file and after reading I need to delete that record,as I need to run the script in a loop for no of records in file.( I am using While-do command for that).But my problem is while deleting record using command"`cat ${TEST_FILE}|sed '$d' >... (5 Replies)
Discussion started by: Roopanwita
5 Replies
3. Shell Programming and Scripting
i have a file named file_names.dat where there are several files along with their path.
exp:
/data1/dir1/CTA_ACD_20071208.dat
/data1/dir1/CTA_DFG_20071208.dat
i want to write a script which will create a csv file with the name of the file and record count of that file
the output file... (4 Replies)
Discussion started by: dr46014
4 Replies
4. UNIX for Dummies Questions & Answers
Hi All,
I'm trying to count how many line in my *.txt file.
My *.txt file has 1937 lines.
The problem is : when I use wc -l command
the result is 1936.
Again, I did some test. I create a new file with 100 lines with text editor (Notepad ++, Ultra edit).
And when I count it again with wc... (2 Replies)
Discussion started by: kunimi
2 Replies
5. Shell Programming and Scripting
Hello:
I wrote a sed statement that is inserting 3 variables at the beginning of each record in a comma-delimited file:
for FILE in *gnrc_lkup.csv
do
c=`echo $FILE | cut -c1-3`
d=`grep $c $RTLIST | cut -c4-6`
e=`grep $c $RTLIST | cut -c7`
f=`grep $c $RTLIST | cut -c8`
sed -e... (5 Replies)
Discussion started by: bheeke
5 Replies
6. Shell Programming and Scripting
i use unix command to take the record count for a file1
awk 'END{print NR}' filename
i already have a file2 which conatin the count like
...
..
rec_cnt=100
..
..
I want to replace the record in the file2 using the record i take from file1.
suggest me some simple ways of doing it... (2 Replies)
Discussion started by: er_zeeshan05
2 Replies
7. Shell Programming and Scripting
Hi all
How to verify the number of records in case of delimited file, If the file has records. And then if it is not equal to mentioned no. of records, email is triggered and file is moved to bad directory path.
Example
-----
input file = a.txt
bad directory path : /usr/bin/bad (6 Replies)
Discussion started by: balaji23_d
6 Replies
8. UNIX for Dummies Questions & Answers
Hi
I am having a 'grep' headache
Here is the contents of my file:
(PBZ,CP,(((ME,PBZ,BtM),ON),((ME,((PBZ,DG),(CW9,PG11))),CW9,TS2,RT1)))
I would like to count out how many times 'PBZ' occurs and then place that number in the line above
3... (8 Replies)
Discussion started by: cavanac2
8 Replies
9. Shell Programming and Scripting
Hi ,
I am having a text file with several records., it has a header record and trailer record. The header record has the number of rows (records) found in the text file and time-stamp. The footer record has the total number of records ( along with the header and trailer.,
Suppose:
wc -l... (4 Replies)
Discussion started by: cratercrabs
4 Replies
10. Shell Programming and Scripting
Hi,
I have a .dat file with contents like the below:
Input file
============SEQ NO-1: COLUMN1==========
9835619
7152815
============SEQ NO-2: COLUMN2 ==========
7615348
7015548
9373086
============SEQ NO-3: COLUMN3===========
9373086
Expected Output: (I just... (1 Reply)
Discussion started by: MS06
1 Replies
fixnt(1) Debian fixnt(1)
NAME
fixnt - Filter for the Windows NT postscript printer driver.
SYNOPSIS
fixnt < BADFILE.ps > GOODFILE.ps
DESCRIPTION
The Windows NT postscript driver has a tendency to make broken postscript files, that are incompatible with psutils. fixnt is a filter
that fixes these problems, allowing the use of psnup(1).
The filter takes the broken postscript file on stdin, and outputs a fixed postscript file on stdout. It has no other form for invocation
and takes no options on the command-line.
OPTIONS
fixnt takes no options.
BUGS
fixnt does not check for NTPSOct94. For a workaround, use a sed(1) command to replace 'NTPSOct94' with 'NTPSOct95', like so:
sed 's/NTPSOct94/NTPSOct95/g'
This is particularly important for Windows NT 3.5 users.
AUTHOR
fixnt was written by Holger Bauer <Holger.Bauer@topmail.de>, Michael Rath <rath@itsm.uni-stuttgart.de>, and Akim Demaille
<demaille@inf.enst.fr>.
REPORTING BUGS
Report bugs to the Authors, but avoid sending large postscript files.
Patches are always welcome; send to <bauer@itsm.uni-stuttgart.de>.
SEE ALSO
psnup(1), sed(1)
a2ps February 2003 fixnt(1)