| #!/usr/bin/perl | |
| # This script takes as STDIN an output from the regFixer.pl | |
| # and reformat the file as if it would have been exported from the registry | |
| # | |
| # Note: the output file is not exactly the one we get when exporting from | |
| # regedit, the tiny difference is that multilines (which were | |
| # single-linized by regFixer.pl) values are not re-multi-linized by | |
| # this script. | |
| # | |
| # Copyright 1999 Sylvain St-Germain | |
| # | |
| ${newkey} = ""; | |
| ${key} = ""; | |
| ${data} = ""; | |
| # I do not validate the integrity of the file, I am assuming that | |
| # the input file comes from the output of regFixer.pl, therefore things | |
| # should be ok, if they are not, investigate and send me an email... | |
| LINE: while(<>) { | |
| chomp; # get rid of new line | |
| next LINE if(/^$/); # This is an empty line | |
| (${key}, ${data}, ${rest})= split /]/,$_ ; # Extract the values from the line | |
| ${key} = "${key}]"; # put the ']' back there... | |
| if (${rest} ne "") # concat we had more than 1 ']' | |
| { # on the line | |
| ${data} = "${data}]${rest}"; | |
| } | |
| if (${key} eq ${newkey}) | |
| { | |
| print "${data}\n"; | |
| } | |
| else | |
| { | |
| ${newkey} = ${key}; # assign it | |
| print "\n"; | |
| print "${key}\n"; | |
| print "${data}\n"; | |
| } | |
| } | |