passing special character in awk

[Dvaderanakin]

How do awk recognize user input such as "Record (1)"??
I have a script that search an inputed fields by the user. The script search the fields and display what's in between the entered fields.

Sample data:-

Record (1) "ML_REC"
"REC_SEQ_NUMBER" = "000442"
"CALL_ID" = "00000000"
"REC_NUMBER" = "01"
"CAUSE_FOR_OUTPUT" = "04"
"A_IMSI" = "234159006266939F"
"A_IMEI" = ""
"CALLING_NUM" = "11447810816136"
"CALLED_NUM" = "120414207193"
"CELL_1ST" = "05F54000015DCB"
"CELL_LAST" = "05F54000015DCB"
"TIME_CHANNEL" = "122525"
"FAULT_CODE" = ""
"TELE_CODE" = "11"
"BEARER_CODE" = ""
"CHAN_RATE_INFO" = "01"
"TRANS_IND" = ""
"RF_POWER_A" = "04"
"DTMF" = ""
"SS_CODE" = ""
"CUG_INTERLOCK" = ""
"CUG_INDEX" = ""
"CUG_OUT_ACC_USED" = ""
"CUG_OUT_ACC_IND" = ""
"REG_SERV_USED" = ""
"REG_DEP_CO" = ""
"CHARG_CASE" = "0001"
"DATE" = "020723"
"TIME_CHARGE" = "122531"
"DURATION" = "0005"
"B_CAT" = "00"
"CALL_STAT" = "03"
"ABNRML_REL" = "01"
"TARIFF_CLASS" = "01"
"TARIFF_ACT_CODE" = "01020D"
"A_CAT" = "01"
"OUT_ROUTE" = "535346494E2020"
"EXCHANGE_ID" = "4D4F444D53432F4134382F30305830"
"MSC_ID" = "1161415019900F"
"MISC_INFO" = ""
"ORIG_FOR_CHARGE" = "15"
"CAMEL_SERVER_ADDRESS" = "1144385016450F"
"CAMEL_SERVICE_LEVEL" = "0F"
"CAMEL_SERVICE_KEY" = "00000006"
"CAMEL_CALL_REFERENCE" = "11614150199000005C103E7F"
"CAMEL_DESTINATION" = "120414209999"
"CAMEL_MODIFICATION" = "5F"
"INTERNAL_CAUSE_AND_LOC" = "0003"
End of Record (1)


How to make the awk command recognize "Record (1)" and End of Record (1)"???
The output should display whatever in between.

Part of my script:-

echo "Please enter a field name:"
read FILE1
file1=$1
awk '$1 ~ /'"$FILE1"'/' $DIR > tmp
echo "What do you want to do?"
echo "Do you want to display the contents?"
echo "Please enter y/n only"
read ANS
if test "$ANS" = "y"
then
vi tmp
else
echo "You could see your results in ~/tmp"
echo "Searching is done!!"
echo "Please wait for a few second to return to main menu"
sleep 1
fi

 

[Dvaderanakin]

Sorry the script should look like this:-

echo "Please enter 2 fields name:"
read FILE1
read FILE2
file1=$1
awk '$1 ~ /'"$FILE1"'/' $DIR > tmp
awk '$1 ~ /'"$FILE2"'/' $DIR >> tmp
echo "Press a to display the the content between the 2 fields entered."
read ANS
if test "$ANS" = "a"
then
echo "Enter a tmp file name besides tmp:"
read TMP
awk '/'"$FILE1"'/, /'"$FILE2"'/' $DIR > $TMP
vi $TMP
else
echo "You could see your results in ~/tmp"
echo "Searching is done!!"
echo "Please wait for a few second to return to main menu"
sleep 1
fi

 

[marsd]

Sed is probably the easier of the two for this and follows
your script model more concisely.

 

[Dvaderanakin]

Hi marsd,

I'm not familiar with sed commands. Can you please show me how you code this in sed?

Thanks for you help!!

 

[marsd]

example(In line with your script)
read -p "Start pattern: " start
read -p "End pattern: " end
sed -e '/'"$start"'/, /'"$end"'/!d' filename

Good Luck.

 

[Dvaderanakin]

Thanks marsd.
It works but not perfect.

For example:-
if you enter
start:1
end:4 for sample file below

1
2
3
4
5
6
10
111

The output would come out as
1
2
3
4
10
111
It will search all the pattern with "1"
anyway, thanks for your help.

 

[marsd]

That shouldn't happen.
The sed matching should end after the first "4" encountered.
For it to be useful you should probably
use more sophisticated regexps than 1
and 4.

Ex:
start="^Rec.*"
end="MISC.*"

sed '/'"$start"'/, /'"$end"'/!d' ../scrap.txt
Record (1) "ML_REC"
"REC_SEQ_NUMBER" = "000442"
"CALL_ID" = "00000000"
"REC_NUMBER" = "01"
"CAUSE_FOR_OUTPUT" = "04"
"A_IMSI" = "234159006266939F"
"A_IMEI" = ""
"CALLING_NUM" = "11447810816136"
"CALLED_NUM" = "120414207193"
"CELL_1ST" = "05F54000015DCB"
"CELL_LAST" = "05F54000015DCB"
"TIME_CHANNEL" = "122525"
"FAULT_CODE" = ""
"TELE_CODE" = "11"
"BEARER_CODE" = ""
"CHAN_RATE_INFO" = "01"
"TRANS_IND" = ""
"RF_POWER_A" = "04"
"DTMF" = ""
"SS_CODE" = ""
"CUG_INTERLOCK" = ""
"CUG_INDEX" = ""
"CUG_OUT_ACC_USED" = ""
"CUG_OUT_ACC_IND" = ""
"REG_SERV_USED" = ""
"REG_DEP_CO" = ""
"CHARG_CASE" = "0001"
"DATE" = "020723"
"TIME_CHARGE" = "122531"
"DURATION" = "0005"
"B_CAT" = "00"
"CALL_STAT" = "03"
"ABNRML_REL" = "01"
"TARIFF_CLASS" = "01"
"TARIFF_ACT_CODE" = "01020D"
"A_CAT" = "01"
"OUT_ROUTE" = "535346494E2020"
"EXCHANGE_ID" = "4D4F444D53432F4134382F30305830"
"MSC_ID" = "1161415019900F"
"MISC_INFO" = ""

Good Luck

 

[Dvaderanakin]

Another related question. More specifix search method.
If you search a field in between Record (1) and End of Record (1)......how do you capture or display the Record #??

Let say u search CALL_ID = 00000000 and DURATION = 0005 field and u want to know which Record that actually contain this fields with those specifix value?

 

[marsd]

Try this:
sed -e '/^Rec.*/, /MISC.*/!d;=' scrap.txt
OP:
2
Record (1) "ML_REC"
3
"REC_SEQ_NUMBER" = "000442"
4
"CALL_ID" = "00000000"
5
"REC_NUMBER" = "01"
6
"CAUSE_FOR_OUTPUT" = "04"
7
"A_IMSI" = "234159006266939F"
8
"A_IMEI" = ""
9
"CALLING_NUM" = "11447810816136"
10
"CALLED_NUM" = "120414207193"
11
"CELL_1ST" = "05F54000015DCB"
12
"CELL_LAST" = "05F54000015DCB"
13
"TIME_CHANNEL" = "122525"
14
"FAULT_CODE" = ""
15
"TELE_CODE" = "11"
16
"BEARER_CODE" = ""
17
"CHAN_RATE_INFO" = "01"
18
"TRANS_IND" = ""
19
"RF_POWER_A" = "04"
20
"DTMF" = ""
21
"SS_CODE" = ""
22
"CUG_INTERLOCK" = ""
23
"CUG_INDEX" = ""
24
"CUG_OUT_ACC_USED" = ""
25
"CUG_OUT_ACC_IND" = ""
26
"REG_SERV_USED" = ""
27
"REG_DEP_CO" = ""
28
"CHARG_CASE" = "0001"
29
"DATE" = "020723"
30
"TIME_CHARGE" = "122531"
31
"DURATION" = "0005"
32
"B_CAT" = "00"
33
"CALL_STAT" = "03"
34
"ABNRML_REL" = "01"
35
"TARIFF_CLASS" = "01"
36
"TARIFF_ACT_CODE" = "01020D"
37
"A_CAT" = "01"
38
"OUT_ROUTE" = "535346494E2020"
39
"EXCHANGE_ID" = "4D4F444D53432F4134382F30305830"
40
"MSC_ID" = "1161415019900F"
41
"MISC_INFO" = ""

 

[Dvaderanakin]

I'm not quite sure what you're trying to do here.
Are you telling me that i have to modify the data file?

 

[marsd]

No!
This will work with what you have proposed.
sed -e '/'"$start"'/, /'"$end'"/!d;='

This will print your user defined range and the
line number of each read line.

If you need just the end or beginning lines, awk IS
a better bet.

 

[Dvaderanakin]

I will try it out first.
Thanks marsd.

 

[Dvaderanakin]

Hi marsd,

I got the output.
To make full use of this i need to print the whole data with those numbering.
How do you print the whole content of the data file with sed.
awk uses print $0
Since i'm not familiar with sed, how do you do it. Tried using awk commands but maybe i' doing something wrong.

Thanks again for your help.

 

[Dvaderanakin]

Hi,

Found out already.
Thanks anyway.

sed -n -e '=;p' filename