Searching for control chars

[grega]

Cross posted from the Perl forum in case anyone has any ideas here.

I have a tab delimited file, 16 fields, 1,000,000 records.

I want to scan the file and report all lines which have control characters in them. When I say control characters, I basically mean all chars except [a-z][A-Z][0-9] and all the other punctuation characters available via a standard keyboard, i.e. (`¬!&quot;£$%^&*-=_+[]{};'#~,./<>?\|) etc. Probably want to ignore carriage returns also.

Hope this makes sense. I'm happy to use any tool available but I'm best with awk, sed or perl. I figure all is needs is the right regular expression.

Any advice much appreciated.

Greg.

 

[tdatgod]

Hi,
if your editor nows how to enter control characters

awk '/[^A-^Z] { printf(&quot;Contorl character at line %d\n&quot;,NR);}' filename

^A would be entered as <Ctrl-v> <ctrl-A> in vi
^Z would be entered as <Ctrl-v> <ctrl-Z> in vi

I guess you build the AWK script with all the unprintable characters ( written in C or Perl )

for ( $x = 1 ; $x < 255 ; $x ++ )
{
if ( ($x < 32 ) || (($x > 127) && ($x<160)))
{
printf(&quot;/%c/ { printf(%c Control Character %o found on line %%d\\n%c,NR); }\n&quot;,
$x,'&quot;',$x,'&quot;');
}
}

Just delete the lines for TAB and Line Feed and <Carriage return> from the script before you run it.