Sort packs of lines alphabetically

Question

_{Note: the complementary question is here: How to separately sort lines within multiple “chunks” separated with headers?}

---

So what I did find an answer to, is how to sort lines in a text files alphabetically. But, it is not exactly what I need to do. I have this file with profiles containing 15 different parameters that goes into an instrument at work and is read by the machine to have a list of profiles.

Sadly, the formatting of this file looks like this:

[ProfileB]
param1=z
param2=y
param3=x
[ProfileA]
param1=k
param2=l
param3=

And I want to sort the Profiles alphabetically, but I need them to stay grouped with their parameters. The above example should be sorted like this:

[ProfileA]
param1=k
param2=l
param3=
[ProfileB]
param1=z
param2=y
param3=x

I guess there is something to work either with the fixed number of lines (name+parameters) or with the character "[" as an identifier for the beginning of a group of lines.

But this is beyond my capacity in text manipulation. I have at my disposition either Sublime Text, R, or Linux command console.

Answer 1

This works in my Debian:

sed '1 ! s/^\[/\x00\[/g' | sort -z | tr -d "\0"

To work with file(s) use redirection(s), e.g. { sed … ; } <input.txt >output.txt, where sed … is the whole command.

The procedure is as follows:

1. sed inserts a null character before every [ that is in the beginning of a line, unless the line is the first one. This way null characters separate profiles.
2. sort -z uses these null characters as separators, so it sorts whole profiles, not separate lines.
3. tr deletes null characters.

Answer 2

Here's a small Perl script that does the job:

my %profiles;
my $profile;

while (<>) {
    if (/\[(.+)\]/) {
        $profile = $1;
        next;
    }
    next if !defined $profile;

    chop if /\n$/;
    push @{ $profiles{$profile} }, $_;
}

foreach my $key (sort keys %profiles) {
    print "[$key]\n";
    foreach my $line (sort @{ $profiles{$profile} }) {
        print "$line\n";
    }
}

Save it into a file, sortProfiles.pl for example, and run:

perl sortProfiles.pl <inputFile.txt >outputFile.txt

How it works

1. It reads the input file (while (<>)).
2. For each profile, [profile] in input file, it remembers its name in $profile variable.
3. It saves each line following the profile header inside an array.
4. Then it sorts the keys of %profiles hash.
5. It also sorts the lines inside the array.

In this script, %profiles is a hash. Its keys are profile names, its values are arrays of lines.
Thus @{ $profiles{$profile} } is the array that stores lines for profile name in $profile variable.