I have a file with thousands of records in it. Each record in the file being of a fixed length of 190 bytes. Example: 0003006315B000000002100601X646000000054717154-12700911700 00Telkwa Trailer Park 0003006317B000000003000601X646000000054717154-12700911700 00Tyhee Trailer Park 0003006613B000000002000601X646000000053242950-13181814600 00Beban Trailer Park 0003006848B000000006000601X646000000054163445-12542196600 00Freeport Trailer Park What I would like to do is for each record position 11 == B and somewhere starting at positon 80 for 49 bytes there is the word "Trailer". I want to replace position 28 for 3 with "xxx" for that record (Example X646 would become Xxxx). I want to do this for the entire file. any help on how to get this to work. I am new to this and I have been doing sed on each individual record and this would take forever. Thanks for any help in advance

awk '\ { if ( substr($0,11,1)=="B" \   &&   match(substr($0,61),"Trailer") )      print substr($0,1,27) "xxx" substr($0,31)   else      print }' filein > fileout

thanks-- if you have a word with "No." how can I get a search with the "." instead of everything with "No" thanks for the help above

For example, in the two lines below, you want only the first line: No. America North America In regular expressions, a dot means any single character. You need to precede the dot with a backslash which will "escape" its special meaning and be taken as an actual dot: grep "No\." myfile awk '/No\./' myfile awk '{if (match($0,"No\."))print}' myfile