Shell Programming and Scripting

Hi All,

I was asked to rewrite a script which is working in Solaris to Linux. In the existing script i have following 'tr' command.

Code:

tr -d '\002\003\013\032\001\014\272'  <$INPUT_FILE | tr -cd '[:print:]\n' >$OUTPUT_FILE

Some1 help me to understand this. Here is my understanding,
tr -d option deletes the occurance of '\002\003\013\032\001\014\272' in the input_file and prints with new line character.

Brief abt script, we are extracting our feed file from Aix (mainframe) and loading that in application running in the Linux OS. Since mainframe uses EBCDIC i have to remove some extra character which is not applicable or printable in ascii. So this script is used to remove those non-printable character and load into solaris server successfully. It is my time to convert this into Linux.

thx for help

the numbers in -d ' ' have different meaning.

Code:

 000 NUL  001 SOH  002 STX  003 ETX  004 EOT 005 ENQ  006 ACK  007 BEL   
 010 BS   011 HT†  012 NL   013 VT   014 NP  015 CR   016 SO   017 SI 
 020 DLE  021 DC1  022 DC2  023 DC3  024 DC4 025 NAK  026 SYN  027 ETB
 030 CAN  031 EM   032 SUB  033 ESC  034 FS  035 GS   036 RS   037 US   
 177 DEL

Code:

tr -cd '[:print:]\n'

The second "tr" command deletes every character which is not in the "printing characters" set called ":print:" except a line-feed character "\n".
The first "tr" commands has already removed some characters which are in the "printing characters" set. It is referring to the Octal value of each character (e.g. \003 is the control code Ctrl/C).
The final result contains a subset of the characters which are in the "printing characters" set or line-feed characters.

On my unix computer the explanation of the meaning of ":print:" is in "man 5 regexp". It will probably be a different "man" page in Linux.


The script posted would only make sense if the input file was already in ASCII (not EBCDIC) or perhaps the environment locale has been changed?