Noah wrote: > Rob Dixon wrote: >> Noah wrote: >>> Rob Dixon wrote: >>>> Noah wrote: >>>> >>>>> let me put toether some bogus output files. >>>> >>>> You could just describe your output. Are you getting nothing at all? Or is >>>> it >>>> showing all lines of both files? Or none from one and all from the other? >>>> Or >>>> just more or fewer lines than you think is correct? >>>> >>>> Remember that non-printable characters (whitespace for instance) may not >>>> match >>>> even though it looks the same: spaces and tab characters for instance. >>> >>> looks like nothing matches. And when I open the files there is a "^M" >>> at the end of each line. what would be the best approach to not pay any >>> attention to the "^M" - search/replace? >>> >>> When I capture the 'vim' man page and remove one line from one of the >>> files - things work fine when there is white space all over the place as >>> long as it matches. >> >> Then I guess you are processing a file that originated on a Windows system? >> Windows text files have a <CR><LF> sequence at the end of each record, >> whereas >> Unix files have just <LF>. <CR> is control-M, which is why vim is showing ^M >> at >> the end of the line. >> >> What you need to do it to remove all whitespace from the end of lines after >> you >> read them (from both the template and the config files) like this. >> >> @config_file_lines = <INPUT>; >> s/\s+$// foreach @config_file_lines; >> >> That will remove and CR or LF characters as well as spaces and tabs (and FFs >> actually) so that there are no line terminators at all and the record >> contents >> can be compared properly. > > okay the necessary line feed at the end of the line is disappearing. > would it be better to something like: > > @config_file_lines = <INPUT>; > s/\s+$/\n/ foreach @config_file_lines;
The newline is only necessary if you perform the substitution only on one set of records. If you also apply it to those from the template file as I said then it will work fine. Rob Rob -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/
