Decode binary octet string in a file with perl - perl

Question

I have a file that contains for some of the lines a number that is coded as text -> binary -> octets and I need to decode that to end up with the number.
All the lines where this encoded string is, begins with STRVID:
For example I have in one of the lines:
STRVID: SarI3gXp
If I do this echo "SarI3gXp" | perl -lpe '$_=unpack"B*"' I get the number in binary
0101001101100001011100100100100100110011011001110101100001110000
Now just to decode from binary to octets I do this (assign the previous command to a variable and then convert binary to octets
variable=$(echo "SarI3gXp" | perl -lpe '$_=unpack"B*"') ; printf '%x\n' "$((2#$variable))"
The result is the number but not in the correct order
5361724933675870
To get the previous number in the correct order I have to get for each couple of digits first the second digit and then the first digit to finally have the number I'm looking for. Something like this:
variable=$(echo "SarI3gXp" | perl -lpe '$_=unpack"B*"') ; printf '%x\n' "$((2#$variable))" | gawk 'BEGIN {FS = ""} {print $2 $1 $4 $3 $6 $5 $8 $7 $10 $9 $12 $11 $14 $13 $16 $15}'
And finally I have the number I'm looking for:
3516279433768507
I don't have any clue on how to do this automatically for every line that begins with STRVID: in my file. At the end what I need is the whole file but when a line begins with STRVID: then the decoded value.
When I find this:
STRVID: SarI3gXp
I will have in my file
STRVID: 3516279433768507
Can someone help with this?

Answer 1

First of all, all you need for the conversion is
unpack "h*", "SarI3gXp"
A perl one-liner using -p will execute the provided program for each line, and s///e allows us to modify a string with code as the replacement expression.
perl -pe's/^STRVID:\s*\K\S+/ unpack "h*", $& /e'
See Specifying file to process to Perl one-liner.

Answer 2

Please inspect the following sample demo code snippet for compliance with your problem.
You do not need double conversion when it can be done in one go.
Note: please read pack documentation , unpack utilizes same TEMPLATE
use strict;
use warnings;
use feature 'say';
while( <DATA> ) {
chomp;
/^STRVID: (.+)/
? say 'STRVID: ' . unpack("h*",$1)
: say;
}
__DATA__
It would be nice if you provide proper input data sample
STRVID: SarI3gXp
Perhaps the result of this script complies with your requirements.
To work with real input data file replace
while( <DATA> ) {
with
while( <> ) {
and pass filename as an argument to the script.
Output
It would be nice if you provide proper input data sample
STRVID: 3516279433768507
Perhaps the result of this script complies with your requirements.
To work with real input data file replace
while( <DATA> ) {
with
while( <> ) {
and pass filename as an argument to the script.
./script.pl input_file.dat

Answer 3

you can cross flip the numbers entirely via regex (and without back-references either) :
variable=$(echo "SarI3gXp" | perl -lpe '$_=unpack"B*"') ;
printf '%x\n' "$((2#$variable))" |
mawk -F'^$' 'gsub("..", "_&=&_") + gsub(\
"(^|[0-9]_)(_[0-9]|$)", _)+gsub("=",_)^_'
1 3516279433768507
The idea is to make a duplicate copy on the other side, like this :
_53=53__61=61__72=72__49=49__33=33__67=67__58=58__70=70_
then scrub out the leftovers, since the numbers u now want are anchoring the 2 sides of each equal sign ("=")