用于屏蔽日志文件中敏感数据的Unix脚本？

Question

I have to write a script which will mask the sensitive data in the log files. I am confused how to implement this? Which option will be best for doing the same:

• Using AWK
• Using SED
• Using SED,AWK
• Using PERL
• Using simple file read and searching logic.

If you have any suggestions then please share.

Input File:
Name  Jack
Add   New York
Phone 333-333-3434

Output File:
Name   Jack
Add    New York
Phone  XXX-XXX-XXXX

I tried this using awk:

cat $HOME_DIR/testdata.dat | awk 'BEGIN{ 
    i=1; 
    FS=" "; 
} 
{ 
    for (i = 1; i < NF; i++) { 
        fld = $(i); 
        if( fld == "PHONE") { 
            printf ("%s$%s", $(i),$(i+1)); 
        } 
        else if( fld == "PIN") { 
            printf ("%s$%s", $(i),$(i+1)); 
        } 
        else if( fld == "DOB") { 
            printf ("%s$%s", $(i),$(i+1)); 
        } else { 
            printf ("%s", $(i)); 
        } 
    } 
    printf ("\n"); 
} 
END{ 
    i=1 
}' > $HOME_DIR/testdataupd.dat

Answer 1

One way using awk . When found words phone , dob or pin at the beginning of the line (ignoring case) substitute in second field all characters but - with X . The print command is executed for every line.

awk '
    BEGIN { 
        IGNORECASE = 1
    }
    $1 ~ /^(phone|dob|pin)$/ {
        gsub( /[^-]/, "X", $2 )
    }
    { print }
' $HOME_DIR/testdata.dat >$HOME_DIR/testdataupd.dat

Answer 2

Here is a 90% answer, does not format the Xs as you asked though.

sed -re 's/(Phone )(([0-9]+)-?)*/\1xxxxx/g'

for more fields

sed -r -e 's/(Phone )(([0-9]+)-?)*/\1xxxxx/g' -e 's/regexp-to-search-for/replacement-pattern/g' …

note: you can replace / with and character you like as long as it is same all 3 times eg s~regexp~rep~g

Answer 3

This might work for you (GNU sed):

sed '/^Phone\|^DOB\|^Pin/!b;h;s/\S*\s*//;s/[^-]/X/g;H;x;s/\(\S*\)\n\(\S*\)/\2/' file

Explanation:

• /^Phone\\|^DOB\\|^Pin/!b only process lines beginning Phone , DOB or Pin (add more here)
• h copy pattern space (PS) to hold space (HS) ie make a copy of the current line.
• s/\\S*\\s*// delete the first first field and following white space.
• s/[^-]/X/g replace all occurrences of - 's with X 's in the remaining field.
• H append a newline and then the PS to the HS.
• x swap HS with PS
• s/\\(\\S*\\)\\n\\(\\S*\\)/\\2/ replace the original second field with the amended one.