gitolite/src/gl-compile-conf

#!/usr/bin/perl

use strict;
use warnings;
use Data::Dumper;
$Data::Dumper::Indent = 1;
$Data::Dumper::Sortkeys = 1;

# === add-auth-keys ===

# part of the gitolite (GL) suite

# (1) - "compiles" ~/.ssh/authorized_keys from the list of pub-keys
# (2) - also "compiles" the user-friendly GL conf file into something easier
#       to parse.  We're doing this because both the gl-auth-command and the
#       (gl-)update hook need this, and it seems easier to do this than
#       replicate the parsing code in both those places.  As a bonus, it's
#       probably more efficient.
# (3) - finally does what I have resisted doing all along -- handle gitweb and
#       git-daemon access.  It won't *setup* gitweb/daemon for you -- you have
#       to that yourself.  What this does is make sure that "repo.git"
#       contains the file "git-daemon-export-ok" (for daemon case) and the
#       line "repo.git" exists in the "projects.list" file (for gitweb case).

# how run:      manual, by GL admin
# when:
#     - anytime a pubkey is added/deleted
#     - anytime gitolite.conf is changed
# input:
#     - GL_CONF (default: ~/.gitolite/conf/gitolite.conf)
#     - GL_KEYDIR (default: ~/.gitolite/keydir)
# output:
#     - ~/.ssh/authorized_keys (dictated by sshd)
#     - GL_CONF_COMPILED (default: ~/.gitolite/conf/gitolite.conf-compiled.pm)
# security:
#     - touches a very critical system file that manages the restrictions on
#       incoming users.  Be sure to audit AUTH_COMMAND and AUTH_OPTIONS (see
#       below) on any change to this script
#     - no security checks within program.  The GL admin runs this manually

# warnings:
#     - if the "start" line exists, but the "end" line does not, you lose the
#       rest of the existing authkey file.  In general, "don't do that (TM)",
#       but we do have a "vim -d" popping up so you can see the changes being
#       made, just in case...

# ----------------------------------------------------------------------------
#       common definitions
# ----------------------------------------------------------------------------

# setup quiet mode if asked; please do not use this when running manually
open STDOUT, ">", "/dev/null" if (@ARGV and shift eq '-q');

# these are set by the "rc" file
our ($GL_ADMINDIR, $GL_CONF, $GL_KEYDIR, $GL_CONF_COMPILED, $REPO_BASE, $REPO_UMASK, $PROJECTS_LIST, $GIT_PATH, $GL_WILDREPOS, $GL_GITCONFIG_KEYS, $GL_PACKAGE_HOOKS);
# and these are set by gitolite.pm
our ($REPONAME_PATT, $REPOPATT_PATT, $USERNAME_PATT, $AUTH_COMMAND, $AUTH_OPTIONS, $ABRT, $WARN);

# the common setup module is in the same directory as this running program is
my $bindir = $0;
$bindir =~ s/\/[^\/]+$//;
$bindir = "$ENV{PWD}/$bindir" unless $bindir =~ /^\//;
require "$bindir/gitolite.pm";

# ask where the rc file is, get it, and "do" it
&where_is_rc();
die "$ABRT parse $ENV{GL_RC} failed: " . ($! or $@) unless do $ENV{GL_RC};

# add a custom path for git binaries, if specified
$ENV{PATH} .= ":$GIT_PATH" if $GIT_PATH;

# ----------------------------------------------------------------------------
#       definitions specific to this program
# ----------------------------------------------------------------------------

# command and options for authorized_keys
$AUTH_COMMAND="$bindir/gl-auth-command";
$AUTH_OPTIONS="no-port-forwarding,no-X11-forwarding,no-agent-forwarding";
    # note, for most users there's also a "no-pty" added to this, see later

# groups can now represent user groups or repo groups.

# $groups{group}{member} = "master" (or name of fragment file in which the
# group is defined).
our %groups = ();

# %repos has two functions.

# $repos{repo}{R|W}{user} = 1 if user has R (or W) permissions for at least
# one branch in repo.  This is used by the "level 1 check" (see faq).  There's
# also the new "C" (create a repo) permission now

# $repos{repo}{user} is a list of {ref, perms} pairs.  This is used by the
# level 2 check.  In order to allow "exclude" rules, the order of rules now
# matters, so what used to be entirely "hash of hash of hash" now has a list
# in between :)
my %repos = ();

# <sigh>...  having been forced to use a list as described above, we lose some
# efficiency due to the possibility of the same {ref, perms} pair showing up
# multiple times for the same repo+user.  So...
my %rurp_seen = ();

# catch usernames<->pubkeys mismatches; search for "lint" below
my %user_list = ();

# repo configurations
my %repo_config = ();

# gitweb descriptions and owners; plain text, keyed by "$repo.git"
my %desc = ();
my %owner = ();

# set the umask before creating any files
umask($REPO_UMASK);

# ----------------------------------------------------------------------------
#       subroutines
# ----------------------------------------------------------------------------

sub expand_list
{
    my @list = @_;
    my @new_list = ();

    for my $item (@list)
    {
        if ($item =~ /^@/)      # nested group
        {
            die "$ABRT undefined group $item\n" unless $groups{$item};
            # add those names to the list
            push @new_list, sort keys %{ $groups{$item} };
        }
        else
        {
            push @new_list, $item;
        }
    }

    return @new_list;
}

# ----------------------------------------------------------------------------
#       "compile" GL conf
# ----------------------------------------------------------------------------

sub parse_conf_file
{
    my ($conffile, $fragment) = @_;
    # the second arg, $fragment, is passed in as "master" when parsing the
    # main config, and the fragment name when parsing a fragment.  In the
    # latter case, the parser uses that information to ignore (and warn about)
    # any repos in the fragment that are not members of the "repo group" of
    # the same name.
    my %ignored = ();

    my $conf_fh = wrap_open( "<", $conffile );

    # the syntax is fairly simple, so we parse it inline

    my @repos;
    while (<$conf_fh>)
    {
        # kill comments, but take care of "#" inside *simple* strings
        s/^((".*?"|[^#"])*)#.*/$1/;
        # normalise whitespace; keeps later regexes very simple
        s/=/ = /;
        s/\s+/ /g;
        s/^ //;
        s/ $//;
        # and blank lines
        next unless /\S/;

        # user or repo groups
        if (/^(@\S+) = (.*)/)
        {
            # store the members of each group as hash key.  Keep track of when
            # the group was *first* created by using $fragment as the *value*
            do { $groups{$1}{$_} ||= $fragment } for ( expand_list( split(' ', $2) ) );
            die "$ABRT bad group $1\n" unless $1 =~ $REPONAME_PATT;
        }
        # repo(s)
        elsif (/^repo (.*)/)
        {
            # grab the list and expand any @stuff in it
            @repos = split ' ', $1;
            unless (@repos == 1 and $repos[0] eq '@all') {
                @repos = expand_list ( @repos );
                do  { die "$ABRT bad reponame $_\n" unless ($GL_WILDREPOS ? $_ =~ $REPOPATT_PATT : $_ =~ $REPONAME_PATT) } for @repos;
            }
            s/\bCREAT[EO]R\b/\$creater/g for @repos;
        }
        # actual permission line
        elsif (/^(-|C|D|R|RW|RW\+) (.* )?= (.+)/)
        {
            my $perms = $1;
            my @refs; @refs = split(' ', $2) if $2;
            @refs = expand_list ( @refs );
            my @users = split ' ', $3;
            die "wildrepos disabled, cant use 'C' in config\n" if $perms eq 'C' and not $GL_WILDREPOS;

            # if no ref is given, this PERM applies to all refs
            @refs = qw(refs/.*) unless @refs;
            # deprecation warning
            map { warn "WARNING: old syntax 'PATH/' found; please use new syntax 'NAME/'\n" if s(^PATH/)(NAME/) } @refs;
            # fully qualify refs that dont start with "refs/" or "NAME/";
            # prefix them with "refs/heads/"
            @refs = map { m(^(refs|NAME)/) or s(^)(refs/heads/); $_ } @refs;
            @refs = map { s(/USER/)(/\$gl_user/); $_ } @refs;

            # expand the user list, unless it is just "@all"
            @users = expand_list ( @users )
                unless (@users == 1 and $users[0] eq '@all');
            do  { die "$ABRT bad username $_\n" unless $_ =~ $USERNAME_PATT } for @users;

            s/\bCREAT[EO]R\b/~\$creater/g for @users;
            s/\bREADERS\b/\$readers/g for @users;
            s/\bWRITERS\b/\$writers/g for @users;

            # ok, we can finally populate the %repos hash
            for my $repo (@repos)       # each repo in the current stanza
            {
                # if we're processing a delegated config file (not the master
                # config), we need to prevent attempts by that admin to obtain
                # rights on stuff outside his domain

                # trying to set access for $repo (='foo')...
                if (
                    # processing the master config, not a fragment
                    ( $fragment eq 'master' ) or
                    # fragment is also called 'foo' (you're allowed to have a
                    # fragment that is only concerned with one repo)
                    ( $fragment eq $repo ) or
                    # fragment is called "bar" and "@bar = foo" has been
                    # defined in the master config
                    ( ($groups{"\@$fragment"}{$repo} || '') eq 'master' )
                ) {
                    # all these are fine
                } else {
                    # this is a little more complex

                    # fragment is called "bar", one or more "@bar = regex"
                    # have been specified in master, and "foo" matches some
                    # such "regex"
                    my @matched = grep { $repo =~ /^$_$/ }
                                  grep { $groups{"\@$fragment"}{$_} eq 'master' }
                                  sort keys %{ $groups{"\@$fragment"} };
                    if (@matched < 1) {
                        $ignored{$fragment}{$repo} = 1;
                        next;
                    }
                }
                for my $user (@users)
                {
                    $user_list{$user}++;    # only to catch lint, see later

                    # for 1st level check (see faq/tips doc)
                    $repos{$repo}{C}{$user} = 1, next if $perms eq 'C';
                    $repos{$repo}{R}{$user} = 1 if $perms =~ /R/;
                    $repos{$repo}{W}{$user} = 1 if $perms =~ /W|D/;

                    # if the user specified even a single 'D' anywhere, make
                    # that fact easy to find; this changes the meaning of RW+
                    # to no longer permit deletes (see update hook)
                    $repos{$repo}{DELETE_IS_D} = 1 if $perms eq 'D';

                    # for 2nd level check, store each "ref, perms" pair in order
                    for my $ref (@refs)
                    {
                        # checking NAME based restrictions is expensive for
                        # the update hook (see the changes to src/hooks/update
                        # in this commit for why) so we would *very* much like
                        # to avoid doing it for the large majority of repos
                        # that do *not* use NAME limits.  Setting a flag that
                        # can be checked right away will help us do that
                        $repos{$repo}{NAME_LIMITS} = 1 if $ref =~ /^NAME\//;
                        push @{ $repos{$repo}{$user} }, { $ref => $perms }
                            unless $rurp_seen{$repo}{$user}{$ref}{$perms}++;
                    }
                }
            }
        }
        # configuration
        elsif (/^config (.+) = ?(.*)/)
        {
            my ($key, $value) = ($1, $2);
            my @validkeys = split (' ', ($GL_GITCONFIG_KEYS || ''));
            my @matched = grep { $key =~ /^$_$/ } @validkeys;
            die "$ABRT git config $key not allowed\n" if (@matched < 1);
            for my $repo (@repos)       # each repo in the current stanza
            {
                $repo_config{$repo}{$key} = $value;
            }
        }
        # include
        elsif (/^include "(.+)"/)
        {
            my $file = $1;
            $file = "$GL_ADMINDIR/conf/$file" unless $file =~ /^\//;
            die "$WARN $fragment attempting to include configuration\n" if $fragment ne 'master';
            die "$ABRT included file not found: '$file'\n" unless -f $file;

            parse_conf_file($file, $fragment);
        }
        # very simple syntax for the gitweb description of repo; one of:
        #       reponame = "some description string"
        #       reponame "owner name" = "some description string"
        elsif (/^(\S+)(?: "(.*?)")? = "(.*)"$/)
        {
            my ($repo, $owner, $desc) = ($1, $2, $3);
            die "$ABRT bad repo name $repo\n" unless $repo =~ $REPONAME_PATT;
            die "$WARN $fragment attempting to set description for $repo\n" if
                $fragment ne 'master' and $fragment ne $repo and ($groups{"\@$fragment"}{$repo} || '') ne 'master';
            $desc{"$repo.git"} = $desc;
            $owner{"$repo.git"} = $owner || '';
        }
        else
        {
            die "$ABRT can't make head or tail of '$_'\n";
        }
    }
    for my $ig (sort keys %ignored)
    {
        warn "\n\t\t***** WARNING *****\n" .
            "\t$ig.conf attempting to set access for " .
            join (", ", sort keys %{ $ignored{$ig} }) . "\n";
    }
}

# parse the main config file
parse_conf_file($GL_CONF, 'master');

# parse any delegated fragments
wrap_chdir($GL_ADMINDIR);
for my $fragment_file (glob("conf/fragments/*.conf"))
{
    # we already check (elsewhere) that a fragment called "foo" will not try
    # to specify access control for a repo whose name is not "foo" or is not
    # part of a group called "foo" created by master

    # meanwhile, I found a possible attack where the admin for group B creates
    # a "convenience" group of (a subset of) his users, and then the admin for
    # repo group A (alphabetically before B) adds himself to that same group
    # in his own fragment.

    # as a result, admin_A now has access to group B repos :(

    # so now we lock the groups hash to the value it had after parsing
    # "master", and localise any changes to it by this fragment so that they
    # don't propagate to the next fragment.  Thus, each fragment now has only
    # those groups that are defined in "master" and itself

    local %groups = %groups;

    my $fragment = $fragment_file;
    $fragment =~ s/^conf\/fragments\/(.*).conf$/$1/;
    parse_conf_file($fragment_file, $fragment);
}

my $compiled_fh = wrap_open( ">", $GL_CONF_COMPILED );
my $dumped_data = Data::Dumper->Dump([\%repos], [qw(*repos)]);
# the dump uses single quotes, but we convert any strings containing $creater,
# $readers, $writers, to double quoted strings.  A wee bit sneaky, but not too
# much...
$dumped_data =~ s/'(?=[^']*\$(?:creater|readers|writers|gl_user))~?(.*?)'/"$1"/g;
print $compiled_fh $dumped_data;
close $compiled_fh or die "$ABRT close compiled-conf failed: $!\n";

# ----------------------------------------------------------------------------
#       any new repos to be created?
# ----------------------------------------------------------------------------

# modern gits allow cloning from an empty repo, so we just create it.  Gitosis
# did not have that luxury, so it was forced to detect the first push and
# create it then

# but it turns out not everyone has "modern" gits :)
my $git_version = `git --version`;
my ($gv_maj, $gv_min, $gv_patchrel) = ($git_version =~ m/git version (\d+)\.(\d+)\.(\d+)/);
die "$ABRT I can't understand $git_version\n" unless ($gv_maj >= 1);
$git_version = $gv_maj*10000 + $gv_min*100 + $gv_patchrel;  # now it's "normalised"

# repo-base needs to be an absolute path for this loop to work right
# so if it was not already absolute, prefix $HOME.
my $repo_base_abs = ( $REPO_BASE =~ m(^/) ? $REPO_BASE : "$ENV{HOME}/$REPO_BASE" );

wrap_chdir("$repo_base_abs");

for my $repo (sort keys %repos) {
    next unless $repo =~ $REPONAME_PATT;
    next if $repo =~ m(^EXTCMD/);   # these are not real repos
    next if $repo eq '@all';
    unless (-d "$repo.git") {
        print STDERR "creating $repo...\n";
        new_repo($repo, "$GL_ADMINDIR/hooks/common");
        # new_repo would have chdir'd us away; come back
        wrap_chdir("$repo_base_abs");
    }

    # when repos are copied over from elsewhere, one had to run easy install
    # once again to make the new (OS-copied) repo contain the proper update
    # hook.  Perhaps we can make this easier now, and eliminate the easy
    # install, with a quick check (and a new, empty, "hook" as a sentinel)
    unless (-l "$repo.git/hooks/gitolite-hooked") {
        ln_sf("$GL_ADMINDIR/hooks/common", "*", "$repo.git/hooks");
        # in case of package install, GL_ADMINDIR is no longer the top cop;
        # override with the package hooks
        ln_sf("$GL_PACKAGE_HOOKS/common", "*", "$repo.git/hooks") if $GL_PACKAGE_HOOKS;
    }
}

warn "\n\t\t***** WARNING *****\n" .
    "\tyour git version is older than 1.6.2\n" .
    "\tgitolite will work but you MUST read the section on\n" .
    "\t\"git version dependency\" in doc/3-faq-tips-etc.mkd\n"
    if $git_version < 10602;     # that's 1.6.2 to you

# ----------------------------------------------------------------------------
#       update repo configurations
# ----------------------------------------------------------------------------

for my $repo (keys %repo_config) {
    wrap_chdir("$repo_base_abs/$repo.git");
    while ( my ($key, $value) = each(%{ $repo_config{$repo} }) ) {
        if ($value) {
            $value =~ s/^"(.*)"$/$1/;
            system("git", "config", $key, $value);
        } else {
            system("git", "config", "--unset-all", $key);
        }
    }
}

# ----------------------------------------------------------------------------
#       handle gitweb and daemon
# ----------------------------------------------------------------------------

# How you specify gitweb and daemon access is quite different from gitosis.  I
# just assume you'll never have any *real* users called "gitweb" or "daemon"
# :-)  These are now "pseduo users" -- giving them "R" access to a repo is all
# you have to do

wrap_chdir("$repo_base_abs");

# daemons first...
for my $repo (sort keys %repos) {
    next unless $repo =~ $REPONAME_PATT;
    my $export_ok = "$repo.git/git-daemon-export-ok";
    if ($repos{$repo}{'R'}{'daemon'}) {
        system("touch $export_ok");
    } else {
        unlink($export_ok);
    }
}

my %projlist = ();
# ...then gitwebs
for my $repo (sort keys %repos) {
    next unless $repo =~ $REPONAME_PATT;
    my $desc_file = "$repo.git/description";
    # note: having a description also counts as enabling gitweb
    if ($repos{$repo}{'R'}{'gitweb'} or $desc{"$repo.git"}) {
        $projlist{"$repo.git"} = 1;
        # add the description file; no messages to user or error checking :)
        $desc{"$repo.git"} and open(DESC, ">", $desc_file) and print DESC $desc{"$repo.git"} . "\n" and close DESC;
        if ($owner{"$repo.git"}) {
            # set the repository owner
            system("git", "--git-dir=$repo.git", "config", "gitweb.owner", $owner{"$repo.git"});
        } else {
            # remove the repository owner setting
            system("git --git-dir=$repo.git config --unset-all gitweb.owner 2>/dev/null");
        }
    } else {
        # delete the description file; no messages to user or error checking :)
        unlink $desc_file;
        # remove the repository owner setting
        system("git --git-dir=$repo.git config --unset-all gitweb.owner 2>/dev/null");
    }

    # unless there are other gitweb.* keys set, remove the section to keep the
    # config file clean
    my $keys = `git --git-dir=$repo.git config --get-regexp '^gitweb\\.' 2>/dev/null`;
    if (length($keys) == 0) {
        system("git --git-dir=$repo.git config --remove-section gitweb 2>/dev/null");
    }
}

# update the project list
my $projlist_fh = wrap_open( ">", $PROJECTS_LIST);
for my $proj (sort keys %projlist) {
    print $projlist_fh "$proj\n";
}
close $projlist_fh;

# ----------------------------------------------------------------------------
#       "compile" ssh authorized_keys
# ----------------------------------------------------------------------------

my $authkeys_fh = wrap_open( "<", $ENV{HOME} . "/.ssh/authorized_keys",
    "\tFor security reasons, gitolite will not *create* this file if it does\n" .
    "\tnot already exist.  Please see the \"admin\" document for details\n");
my $newkeys_fh = wrap_open( ">", $ENV{HOME} . "/.ssh/new_authkeys" );
# save existing authkeys minus the GL-added stuff
while (<$authkeys_fh>)
{
    print $newkeys_fh $_ unless (/^# gito(sis-)?lite start/../^# gito(sis-)?lite end/);
}

# add our "start" line, each key on its own line (prefixed by command and
# options, in the standard ssh authorized_keys format), then the "end" line.
print $newkeys_fh "# gitolite start\n";
wrap_chdir($GL_KEYDIR);
for my $pubkey (`find . -type f`)
{
    chomp($pubkey); $pubkey =~ s(^\./)();

    # security check (thanks to divVerent for catching this)
    unless ($pubkey =~ $REPONAME_PATT) {
        print STDERR "$pubkey contains some unsavoury characters; ignored...\n";
        next;
    }

    # lint check 1
    unless ($pubkey =~ /\.pub$/)
    {
        print STDERR "WARNING: pubkey files should end with \".pub\", ignoring $pubkey\n";
        next;
    }

    my $user = $pubkey;
    $user =~ s(.*/)();                  # foo/bar/baz.pub -> baz.pub
    $user =~ s/(\@[^.]+)?\.pub$//;      # baz.pub, baz@home.pub -> baz

    # lint check 2
    print STDERR "WARNING: pubkey $pubkey exists but user $user not in config\n"
        unless $user_list{$user};
    $user_list{$user} = 'has pubkey';
    # apparently some pubkeys don't end in a newline...
    my $pubkey_content = `cat $pubkey`;
    $pubkey_content =~ s/\s*$/\n/;
    # don't trust files with multiple lines (i.e., something after a newline)
    if ($pubkey_content =~ /\n./)
    {
        print STDERR "WARNING: a pubkey file can only have one line (key); ignoring $pubkey\n";
        next;
    }
    print $newkeys_fh "command=\"$AUTH_COMMAND $user\",$AUTH_OPTIONS,no-pty ";
    print $newkeys_fh $pubkey_content;
}
# lint check 3; a little more severe than the first two I guess...
for my $user (sort keys %user_list)
{
    next if $user =~ /^(gitweb|daemon|\@all|~\$creater|\$readers|\$writers)$/ or $user_list{$user} eq 'has pubkey';
    print STDERR "$WARN user $user in config, but has no pubkey!\n";
}

print $newkeys_fh "# gitolite end\n";
close $newkeys_fh or die "$ABRT close newkeys failed: $!\n";

# all done; overwrite the file (use cat to avoid perm changes)
system("cat $ENV{HOME}/.ssh/authorized_keys > $ENV{HOME}/.ssh/old_authkeys");
system("cat $ENV{HOME}/.ssh/new_authkeys > $ENV{HOME}/.ssh/authorized_keys")
    and die "couldn't write authkeys file\n";
system("rm  $ENV{HOME}/.ssh/new_authkeys");