Hi! I want to know if there is any way to substitute the contextual grep (-C) function of Linux in UNIX. My problem is that I need the 10 lines preceiding my grep search in a log file? Any suggestions? Thanks! kart

The following script search file(s) for a string (ereg specification) and print context lines (before and after). Usage: grep_with_context [ -v before=before_lines ] [ -v after=after_lines ][ -v context=context_lines ] ereg file... before_lines : number of lines to display after line containing ereg after_lines : number of lines to display after line containing ereg context-lines: number of lines to display before and after line containing ereg ereg : searched string as a regular expression file.. : input file(s) Example: grep_with_context -v before=10 'ERROR' input_file ------ grep_with_context ------ #!/usr/bin/awk -f function PrintBeforeContext ( lindex, lfrom, lto) { lfrom = before_index - before; if (lfrom < 0) lfrom = 0 ; lto = before_index - 1 ; for ( lindex = lfrom ; lindex <= lto ; lindex++ ) { print before_context[lindex % before] ; } before_index = 0 ; } BEGIN { context = context + 0; before = before + 0 ; after = after + 0 ; if (context > 0) { before = context; after = context; } before_index = 0 ; before_context[0] = ""; after_index = 0 ; pattern = ARGV[1]; ARGV[1] = ""; if (ARGC <= 2) ARGV[ARGC++] = "-"; } $0 ~ pattern { PrintBeforeContext() ; print $0 ; after_index = after ; next ; } after_index > 0 { print $0 ; after_index-- ; next ; } before > 0 { before_context[before_index % before] = $0 ; before_index++ ; } ------ grep_with_context ------ Jean-Pierre.

more readable : #!/usr/bin/awk -f function PrintBeforeContext ( lindex, lfrom, lto) {    lfrom = before_index - before;    if (lfrom < 0) lfrom = 0 ;    lto = before_index - 1 ;    for ( lindex = lfrom ; lindex <= lto ; lindex++ ) {        print before_context[lindex % before] ;    }    before_index = 0 ; } BEGIN {    context = context + 0;    before = before + 0 ;    after = after + 0 ;    if (context > 0) {        before = context;        after = context;    }    before_index = 0 ;    before_context[0] = "";    after_index = 0 ;    pattern = ARGV[1];    ARGV[1] = "";    if (ARGC <= 2) ARGV[ARGC++] = "-"; } $0 ~ pattern { PrintBeforeContext() ;    print $0 ;    after_index = after ;    next ; } after_index > 0 {    print $0 ;    after_index-- ;    next ; } before > 0 {    before_context[before_index % before] = $0 ;    before_index++ ; } Jean-Pierre.

As I noted in response to this posting in the Linux Forum: how about ex: ex <<! file g/pattern/.,.-10p q ! This corrects the Linux posting which had ".+10p" To get 5 before and 5 after g/pattern/.,.-5,.+5p A problem with this is that if there are not 10 lines before "pattern", ex silently fails.

Typing error : g/pattern/.-10,.p There is some problems with your solution With the following input file : 1 SUCCES 2 SUCCES 3 ERROR 4 SUCCES 5 SUCCES 6 ERROR 7 ERROR 8 SUCCES The result for g/ERROR/.-2,.p is : ^[[24;1H"u.txt" 8 lines, 109 characters 1 SUCCES 2 SUCCES 3 ERROR 4 SUCCES 5 SUCCES 6 ERROR 5 SUCCES 6 ERROR 7 ERROR ^[[J^[[?25h 1) There are not wanted display (escape seq, file name ) 2) Some lines are displayed many times The correct result is : 1 SUCCES 2 SUCCES 3 ERROR 4 SUCCES 5 SUCCES 6 ERROR 7 ERROR Jean-Pierre.