Question

I am writing a script to read a file and replace some characters which includes unrecognized characters like

"^H^H^H^H^H^H^H^H^H"

Actually this file was generated by redirecting the console output of a telnet session. I want to remove this character. I have to remove “–More—“ from the file.

I gave a try on replacing them using

set fileID [open "bar" r]
set temp [open "temp.txt" w+]
while {[eof $fileID] != 1} {
    gets $fileID lineInfo
    regsub -all "More" $lineInfo "" lineInfo
    regsub -all "--More--" $lineInfo "" lineInfo  #This is not working
    puts $temp $lineInfo
}

I can remove “More”, but i am not able to remove “–More–“. Can anybody explain this ?

I also tried like

regsub -all "^H^H^H^H^H^H^H^H^H" $lineInfo "" lineInfo  #This is also not working

This also not working.

Thanks in advance.

Answer 1

I would recomment to use the :print: character class in your regex to match all “sensible” characters — refer to the re_syntax manual page.

The resulting call to regsub should replace all non-printable characters with empty strings, so we use a negated character class (via ^), and so the proper incantation would look like this:

set lineInfo [regsub -all {[^[:print:]]} $lineInfo ""]

As to removing “–More–“” — I have two hypotheses:

• regsub is confused treating --More-- as a switch as it starts with a dash. This can be easily fixed by passing -- to regsub after all the switches — see the manual page.
• Your “–More–” as it appears in the file is not really a solid string, and it has certain non-printable characters embedded into it. The usual approach to debugging such situations is inspecting the file usina a HEX-dumper like xxd or a HEX-editor/viewer.