sed or grep or awk to match very very long lines

more file

param1=" 1,deerfntjefnerjfntrjgntrjnvgrvgrtbvggfrjbntr*rfr4fv*frfftrjgtrignmtignmtyightygjn      2,3,4,5,6,7,8, 
rfcmckmfdkckemdio8u548384omxc,mor0ckofcmineucfhcbdjcnedjcnywedpeodl40fcrcmkedmrikmckffmcrffmrfrifmtrifmrifvysdfn drfr4fdr4fmedmifmitfmifrtfrfrfrfnurfnurnfrunfrufnrufnrufnrufnruf"****

need to match the content of param1 as

sed -n "/$param1/p" file 

but because the line length (very long line) I cant match开发者_如何学运维 the line what’s the best way to match very long lines?

The problem you are facing is that param1 contains special characters which are being interpreted by sed. The asterisk ('*') is used to mean 'zero or more occurrences of the previous character', so when this character is interpreted by sed there is nothing left to match the literal asterisk you are looking for.

The following is a working bash script that should help:

#!/bin/bash

param1=' 1,deerfntjefnerjfntrjgntrjnvgrvgrtbvggfrjbntr\*rfr4fv\*frfftrjgtrignmtignmtyightygjn 2,3,4,5,6,7,8, rfcmckmfdkckemdio8u548384omxc,mor0ckofcmineucfhcbdjcnedjcnywedpeodl40fcrcmkedmrikmckffmcrffmrfrifmtrifmrifvysdfn'

cat <<EOF | sed "s/${param1}/Bubba/g"
 1,deerfntjefnerjfntrjgntrjnvgrvgrtbvggfrjbntr*rfr4fv*frfftrjgtrignmtignmtyightygjn 2,3,4,5,6,7,8, rfcmckmfdkckemdio8u548384omxc,mor0ckofcmineucfhcbdjcnedjcnywedpeodl40fcrcmkedmrikmckffmcrffmrfrifmtrifmrifvysdfn    
EOF

Maybe the problem is that your $param1 contains special characters? This works for me:

A="$(perl -e 'print "a" x 10000')"
echo $A | sed -n "/$A/p"

($A contains 10 000 a characters).

echo $A | grep -F $A

and

echo $A | grep -P $A

also works (second requires grep with built-in PCRE support. If you want pattern matching you should use either this or pcregrep. If you don't, use the fixed grep (grep -F)).

echo $A | grep $A

is too slow.