asked on Error while assigning a character value to a numeric variable
I have a script that uses a numeric variable resA.
#!/bin/ksh
typeset -i resA=1,233
It has to be an integer type variable, however, it's value assigned is character.
So it gives an error:
$> ksh -x run.ksh
+ typeset -i resA=1,22
run.ksh[2]: 1,22: syntax error
How do I change 1,22 into a number 122 in korn shell?
#!/bin/ksh
typeset -i resA=1,233
It has to be an integer type variable, however, it's value assigned is character.
So it gives an error:
$> ksh -x run.ksh
+ typeset -i resA=1,22
run.ksh[2]: 1,22: syntax error
How do I change 1,22 into a number 122 in korn shell?
Unix OSLinux
Last Comment
omarfarid
Have you tried "1.22" instead of "1,22"? Dot instead of comma.
ASKER
I grep the value 1,22 from a log file. So it can't be changed.
ASKER
Can i use sed to remove the comma from 8,454 into 8454 and save it to a file?
This is my requirement in detail:
I have to read a log file for "lines fetched":
4/14/2008 11:15:01 AM: 0067 SQL SELECT "TEAM_CD",
4/14/2008 11:15:01 AM: 0068 "TERR_ID"
4/14/2008 11:15:01 AM: 0069 FROM "OPS$TMS"."DM_TERR"
4/14/2008 11:15:01 AM: 2 fields found: TEAM_CD, TERR_ID, 8,454 lines fetched
I pick 8,454 value, and assign it to a variable resB like this.
resB=`grep '2 fields found: TEAM_CD, TERR_ID,' BCVP_Main_Loader.qvw.log | perl -nle 'print $1 if m/(\d+\,?\d+)\ lines/;'
resB is assigned 8,454.
Then i compare resB with resA
set `sqlplus -s $opstms_conn_string << EOF
set pages 0
WHENEVER SQLERROR CONTINUE
SELECT count(*)
FROM DM_TERR;
exit
EOF`
resA=$1
now as resA is numeric 8454 and resB is character 8,454 I get an error.
The error is because resB is defined as a numeric variable too.
This is my requirement in detail:
I have to read a log file for "lines fetched":
4/14/2008 11:15:01 AM: 0067 SQL SELECT "TEAM_CD",
4/14/2008 11:15:01 AM: 0068 "TERR_ID"
4/14/2008 11:15:01 AM: 0069 FROM "OPS$TMS"."DM_TERR"
4/14/2008 11:15:01 AM: 2 fields found: TEAM_CD, TERR_ID, 8,454 lines fetched
I pick 8,454 value, and assign it to a variable resB like this.
resB=`grep '2 fields found: TEAM_CD, TERR_ID,' BCVP_Main_Loader.qvw.log | perl -nle 'print $1 if m/(\d+\,?\d+)\ lines/;'
resB is assigned 8,454.
Then i compare resB with resA
set `sqlplus -s $opstms_conn_string << EOF
set pages 0
WHENEVER SQLERROR CONTINUE
SELECT count(*)
FROM DM_TERR;
exit
EOF`
resA=$1
now as resA is numeric 8454 and resB is character 8,454 I get an error.
The error is because resB is defined as a numeric variable too.
ASKER
I need to compare resA and resB,
if [ $resA -eq $resB ]; then
...
this can be done only when both resA and resB are numeric.
if [ $resA -eq $resB ]; then
...
this can be done only when both resA and resB are numeric.
ASKER
echo 1,234 | sed 's/,//' gives me 1234
ASKER
This has removed the first comma in the string using sed, how do i remove the second comma.
$> echo 1,085,712 | sed 's/,//'
1085,712
I want an sed command that would remove all commas from the string.
123,456,789,101,112,131,4
$> echo 1,085,712 | sed 's/,//'
1085,712
I want an sed command that would remove all commas from the string.
123,456,789,101,112,131,4
ASKER CERTIFIED SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
Why don't you run it on the whole log file:
sed 'sed s/,//g' < logfile > logfile.tmp
mv logile.tmp logfile
sed 'sed s/,//g' < logfile > logfile.tmp
mv logile.tmp logfile