Shell Programming and Scripting

Hi,

I am using sed command to replace following line in gz file-

Code:

sed -n 's/""COS_12_TM_1" ( 1101110101001001010011110100000010110100010010000000100000000010XX010000000 )"/""COS_12_TM_1" ( 110111010100100101001111MM00000010110100010010000000100000000010XX010000000 )"/g' filename.gz

Code:

$x= ""COS_12_TM_1" ( 1101110101001001010011110100000010110100010010000000100000000010XX010000000 )"
$y=""COS_12_TM_1" ( 110111010100100101001111MM00000010110100010010000000100000000010XX010000000 )"

eg. sed 's/$x/$y/g'
Can anyone tell me correct way to do this.Because it is not working.
Both the strings $x and $y are present in different files($x in list and $y in Masked_list).
Also need help in using nested foreach loop ,Following is the script I am trying to run-

Code:

set x=`cat list`
set y=`cat Masked_list`
foreach org ( $x )
echo "$org" >> list_1
    foreach masked ( $y )
            echo "$masked" >> list_2
        `sed -n 's/$org/$masked/g' lw20_82_55_ac_0x19E1.wgl.gz` >> new_wgl
        set tmp_2=`grep $org new_wgl`
         if ( $tmp_2 == " " ) then
            echo "Flip Flop MAsekd in the Original WGL File"
     end
        endif
end

Thanks in advance!!!

Moderator's Comments:

Please use CODE tags as required by forum rules!

It is not woring because the quoting prevents the expansion of the variables. When you write

Code:

command $VAR something

you expect the string "$VAR" to be replaced by the content of the variable VAR - this is called "expansion" and the quotation in single-quotes was invented to prevent exactly this. (The prevention of variable expansion is the big difference between single- and double-quotes.) Therefore:

Code:

sed 's/'"$x"'/'"$y"'/g'

should work. Notice that this might still not do what you want it to do because the variables content could hold characters with a special meaning to sed (like "*", "/", etc.), but this is a different kind of error.

I hope this helps.

bakunin

And no backticks here

Code:

sed -n 's/'"$org"'/'"$masked"'/g' lw20_82_55_ac_0x19E1.wgl.gz >> new_wgl

tcsh is problematic - use bash (or zsh or ksh) instead! The syntax is different, but it is worth to learn it.
I see .gz and hope it's not a compressed file. That would give you an additional problem, like corrupting the file so cannot be uncompressed.

Thank you guys...It is working now..

---------- Post updated at 02:05 AM ---------- Previous update was at 01:51 AM ----------

One more question:
I want to grep"COS_12_TM_4 pattern from a file look likes :
"COS_12_TM_4" [
"dummy_COS_12_1_TM_4",
"dummy_COS_12_2_TM_4",
"dummy_COS_12_3_TM_4",
"dummy_COS_12_4_TM_4",
"dummy_COS_12_5_TM_4",
"dummy_COS_12_6_TM_4",
"dummy_COS_12_7_TM_4",
"dummy_COS_12_8_TM_4",
"dummy_COS_12_9_TM_4",
"dummy_COS_12_10_TM_4",
"dummy_COS_12_11_TM_4",
"dummy_COS_12_12_TM_4",
"dummy_COS_12_13_TM_4",
"dummy_COS_12_14_TM_4",
"dummy_COS_12_15_TM_4",
"dummy_COS_12_16_TM_4",
"dummy_COS_12_17_TM_4",
"dummy_COS_12_18_TM_4",
"dummy_COS_12_19_TM_4",
"dummy_COS_12_20_TM_4",
"dummy_COS_12_21_TM_4",
"dummy_COS_12_22_TM_4",
"dummy_COS_12_23_TM_4",
"dummy_COS_12_24_TM_4",
"dummy_COS_12_25_TM_4",
"dummy_COS_12_26_TM_4",
"dummy_COS_12_27_TM_4",
"dummy_COS_12_28_TM_4",
"dummy_COS_12_29_TM_4",
"dummy_COS_12_30_TM_4",
"dummy_COS_12_31_TM_4",
"dummy_COS_12_32_TM_4",
"dummy_COS_12_33_TM_4",
"dummy_COS_12_34_TM_4",
"dummy_COS_12_35_TM_4",
"dummy_COS_12_36_TM_4",
"scan_out[5]" ];
I am taking scan_out[5] as the input from the user.
From the above file how to grep "COS_12_TM_4" this string which is corresponding to the input scan_out[5].