[thirdparty/git.git] / git-svnimport.perl

#!/usr/bin/perl -w

# This tool is copyright (c) 2005, Matthias Urlichs.
# It is released under the Gnu Public License, version 2.
#
# The basic idea is to pull and analyze SVN changes.
#
# Checking out the files is done by a single long-running CVS connection
# / server process.
#
# The head revision is on branch "origin" by default.
# You can change that with the '-o' option.

require v5.8.0; # for shell-safe open("-|",LIST)
use strict;
use warnings;
use Getopt::Std;
use File::Spec;
use File::Temp qw(tempfile);
use File::Path qw(mkpath);
use File::Basename qw(basename dirname);
use Time::Local;
use IO::Pipe;
use POSIX qw(strftime dup2);
use IPC::Open2;
use SVN::Core;
use SVN::Ra;

die "Need CVN:COre 1.2.1 or better" if $SVN::Core::VERSION lt "1.2.1";

$SIG{'PIPE'}="IGNORE";
$ENV{'TZ'}="UTC";

our($opt_h,$opt_o,$opt_v,$opt_u,$opt_C,$opt_i,$opt_m,$opt_M,$opt_t,$opt_T,$opt_b);

sub usage() {
	print STDERR <<END;
Usage: ${\basename $0}     # fetch/update GIT from CVS
       [-o branch-for-HEAD] [-h] [-v]
       [-C GIT_repository] [-t tagname] [-T trunkname] [-b branchname]
       [-i] [-u] [-s subst] [-m] [-M regex] [SVN_URL]
END
	exit(1);
}

getopts("b:C:hivmM:o:t:T:u") or usage();
usage if $opt_h;

my $tag_name = $opt_t || "tags";
my $trunk_name = $opt_T || "trunk";
my $branch_name = $opt_b || "branches";

@ARGV <= 1 or usage();

$opt_o ||= "origin";
my $git_tree = $opt_C;
$git_tree ||= ".";

my $cvs_tree;
if ($#ARGV == 0) {
	$cvs_tree = $ARGV[0];
} elsif (-f 'CVS/Repository') {
	open my $f, '<', 'CVS/Repository' or 
	    die 'Failed to open CVS/Repository';
	$cvs_tree = <$f>;
	chomp $cvs_tree;
	close $f;
} else {
	usage();
}

our @mergerx = ();
if ($opt_m) {
	@mergerx = ( qr/\W(?:from|of|merge|merging|merged) (\w+)/i );
}
if ($opt_M) {
	push (@mergerx, qr/$opt_M/);
}

select(STDERR); $|=1; select(STDOUT);


package SVNconn;
# Basic SVN connection.
# We're only interested in connecting and downloading, so ...

use File::Spec;
use File::Temp qw(tempfile);
use POSIX qw(strftime dup2);

sub new {
	my($what,$repo) = @_;
	$what=ref($what) if ref($what);

	my $self = {};
	$self->{'buffer'} = "";
	bless($self,$what);

	$repo =~ s#/+$##;
	$self->{'fullrep'} = $repo;
	$self->conn();

	return $self;
}

sub conn {
	my $self = shift;
	my $repo = $self->{'fullrep'};
	my $s = SVN::Ra->new($repo);

	die "SVN connection to $repo: $!\n" unless defined $s;
	$self->{'svn'} = $s;
	$self->{'repo'} = $repo;
	$self->{'maxrev'} = $s->get_latest_revnum();
}

sub file {
	my($self,$path,$rev) = @_;
	my $res;

	my ($fh, $name) = tempfile('gitsvn.XXXXXX', 
		    DIR => File::Spec->tmpdir(), UNLINK => 1);

	print "... $rev $path ...\n" if $opt_v;
	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
	if ($@ and $@ !~ /Attempted to get checksum/) {
	    # retry
	    $self->conn();
		eval { $self->{'svn'}->get_file($path,$rev,$fh); };
	};
	return () if $@ and $@ !~ /Attempted to get checksum/;
	die $@ if $@;
	close ($fh);

	return ($name, $res);
}


package main;

my $svn = SVNconn->new($cvs_tree);


sub pdate($) {
	my($d) = @_;
	$d =~ m#(\d\d\d\d)-(\d\d)-(\d\d)T(\d\d):(\d\d):(\d\d)#
		or die "Unparseable date: $d\n";
	my $y=$1; $y-=1900 if $y>1900;
	return timegm($6||0,$5,$4,$3,$2-1,$y);
}

sub getwd() {
	my $pwd = `pwd`;
	chomp $pwd;
	return $pwd;
}


sub get_headref($$) {
    my $name    = shift;
    my $git_dir = shift; 
    my $sha;
    
    if (open(C,"$git_dir/refs/heads/$name")) {
	chomp($sha = <C>);
	close(C);
	length($sha) == 40
	    or die "Cannot get head id for $name ($sha): $!\n";
    }
    return $sha;
}


-d $git_tree
	or mkdir($git_tree,0777)
	or die "Could not create $git_tree: $!";
chdir($git_tree);

my $orig_branch = "";
my $forward_master = 0;
my %branches;

my $git_dir = $ENV{"GIT_DIR"} || ".git";
$git_dir = getwd()."/".$git_dir unless $git_dir =~ m#^/#;
$ENV{"GIT_DIR"} = $git_dir;
my $orig_git_index;
$orig_git_index = $ENV{GIT_INDEX_FILE} if exists $ENV{GIT_INDEX_FILE};
my ($git_ih, $git_index) = tempfile('gitXXXXXX', SUFFIX => '.idx',
				    DIR => File::Spec->tmpdir());
close ($git_ih);
$ENV{GIT_INDEX_FILE} = $git_index;
my $maxnum = 0;
my $last_rev = "";
my $last_branch;
my $current_rev = 0;
unless(-d $git_dir) {
	system("git-init-db");
	die "Cannot init the GIT db at $git_tree: $?\n" if $?;
	system("git-read-tree");
	die "Cannot init an empty tree: $?\n" if $?;

	$last_branch = $opt_o;
	$orig_branch = "";
} else {
	-f "$git_dir/refs/heads/$opt_o"
		or die "Branch '$opt_o' does not exist.\n".
		       "Either use the correct '-o branch' option,\n".
		       "or import to a new repository.\n";

	-f "$git_dir/svn2git"
		or die "'$git_dir/svn2git' does not exist.\n".
		       "You need that file for incremental imports.\n";
	$last_branch = basename(readlink("$git_dir/HEAD"));
	unless($last_branch) {
		warn "Cannot read the last branch name: $! -- assuming 'master'\n";
		$last_branch = "master";
	}
	$orig_branch = $last_branch;
	$last_rev = get_headref($orig_branch, $git_dir);
	if (-f "$git_dir/SVN2GIT_HEAD") {
		die <<EOM;
SVN2GIT_HEAD exists.
Make sure your working directory corresponds to HEAD and remove SVN2GIT_HEAD.
You may need to run

    git-read-tree -m -u SVN2GIT_HEAD HEAD
EOM
	}
	system('cp', "$git_dir/HEAD", "$git_dir/SVN2GIT_HEAD");

	$forward_master =
	    $opt_o ne 'master' && -f "$git_dir/refs/heads/master" &&
	    system('cmp', '-s', "$git_dir/refs/heads/master", 
				"$git_dir/refs/heads/$opt_o") == 0;

	# populate index
	system('git-read-tree', $last_rev);
	die "read-tree failed: $?\n" if $?;

	# Get the last import timestamps
	open my $B,"<", "$git_dir/svn2git";
	while(<$B>) {
		chomp;
		my($num,$branch,$ref) = split;
		$branches{$branch}{$num} = $ref;
		$branches{$branch}{"LAST"} = $ref;
		$current_rev = $num+1 if $current_rev < $num+1;
	}
	close($B);
}
-d $git_dir
	or die "Could not create git subdir ($git_dir).\n";

open BRANCHES,">>", "$git_dir/svn2git";


## cvsps output:
#---------------------
#PatchSet 314
#Date: 1999/09/18 13:03:59
#Author: wkoch
#Branch: STABLE-BRANCH-1-0
#Ancestor branch: HEAD
#Tag: (none)
#Log:
#    See ChangeLog: Sat Sep 18 13:03:28 CEST 1999  Werner Koch
#Members:
#	README:1.57->1.57.2.1
#	VERSION:1.96->1.96.2.1
#
#---------------------

my $state = 0;

sub get_file($$$) {
	my($rev,$branch,$path) = @_;

	# revert split_path(), below
	my $svnpath;
	$path = "" if $path eq "/"; # this should not happen, but ...
	if($branch eq "/") {
		$svnpath = "/$trunk_name/$path";
	} elsif($branch =~ m#^/#) {
		$svnpath = "/$tag_name$branch/$path";
	} else {
		$svnpath = "/$branch_name/$branch/$path";
	}

	# now get it
	my ($name, $res) = eval { $svn->file($svnpath,$rev); };
	return () unless defined $name;

	open my $F, '-|', "git-hash-object", "-w", $name
		or die "Cannot create object: $!\n";
	my $sha = <$F>;
	chomp $sha;
	close $F;
	my $mode = "0644"; # SV does not seem to store any file modes
	return [$mode, $sha, $path];
}

sub split_path($$) {
	my($rev,$path) = @_;
	my $branch;

	if($path =~ s#^/\Q$tag_name\E/([^/]+)/?##) {
		$branch = "/$1";
	} elsif($path =~ s#^/\Q$trunk_name\E/?##) {
		$branch = "/";
	} elsif($path =~ s#^/\Q$branch_name\E/([^/]+)/?##) {
		$branch = $1;
	} else {
		print STDERR "$rev: Unrecognized path: $path\n";
		return ()
	}
	$path = "/" if $path eq "";
	return ($branch,$path);
}

sub commit {
	my($branch, $changed_paths, $revision, $author, $date, $message) = @_;
	my($author_name,$author_email,$dest);
	my(@old,@new);

	if (not defined $author) {
		$author_name = $author_email = "unknown";
	} elsif ($author =~ /^(.*?)\s+<(.*)>$/) {
		($author_name, $author_email) = ($1, $2);
	} else {
		$author =~ s/^<(.*)>$/$1/;
		$author_name = $author_email = $author;
	}
	$date = pdate($date);

	my $tag;
	my $parent;
	if($branch eq "/") { # trunk
		$parent = $opt_o;
	} elsif($branch =~ m#^/(.+)#) { # tag
		$tag = 1;
		$parent = $1;
	} else { # "normal" branch
		# nothing to do
		$parent = $branch;
	}
	$dest = $parent;

	my $prev = $changed_paths->{"/"};
	if($prev and $prev->[0] eq "A") {
		delete $changed_paths->{"/"};
		my $oldpath = $prev->[1];
		my $rev;
		if(defined $oldpath) {
			my $p;
			($parent,$p) = split_path($revision,$oldpath);
			if($parent eq "/") {
				$parent = $opt_o;
			} else {
				$parent =~ s#^/##; # if it's a tag
			}
		} else {
			$parent = undef;
		}
	}

	my $rev;
	if(defined $parent) {
		open(H,"git-rev-parse --verify $parent |");
		$rev = <H>;
		close(H) or do {
			print STDERR "$revision: cannot find commit '$parent'!\n";
			return;
		};
		chop $rev;
		if(length($rev) != 40) {
			print STDERR "$revision: cannot find commit '$parent'!\n";
			return;
		}
		$rev = $branches{($parent eq $opt_o) ? "/" : $parent}{"LAST"};
		if($revision != 1 and not $rev) {
			print STDERR "$revision: do not know ancestor for '$parent'!\n";
			return;
		}
	} else {
		$rev = undef;
	}

#	if($prev and $prev->[0] eq "A") {
#		if(not $tag) {
#			unless(open(H,"> $git_dir/refs/heads/$branch")) {
#				print STDERR "$revision: Could not create branch $branch: $!\n";
#				$state=11;
#				next;
#			}
#			print H "$rev\n"
#				or die "Could not write branch $branch: $!";
#			close(H)
#				or die "Could not write branch $branch: $!";
#		}
#	}
	if(not defined $rev) {
		unlink($git_index);
	} elsif ($rev ne $last_rev) {
		print "Switching from $last_rev to $rev ($branch)\n" if $opt_v;
		system("git-read-tree", $rev);
		die "read-tree failed for $rev: $?\n" if $?;
		$last_rev = $rev;
	}

	while(my($path,$action) = each %$changed_paths) {
		if ($action->[0] eq "A") {
			my $f = get_file($revision,$branch,$path);
			push(@new,$f) if $f;
		} elsif ($action->[0] eq "D") {
			push(@old,$path);
		} elsif ($action->[0] eq "M") {
			my $f = get_file($revision,$branch,$path);
			push(@new,$f) if $f;
		} elsif ($action->[0] eq "R") {
			# refer to a file/tree in an earlier commit
			push(@old,$path); # remove any old stuff

			# ... and add any new stuff
			my($b,$p) = split_path($revision,$action->[1]);
			open my $F,"-|","git-ls-tree","-r","-z", $branches{$b}{$action->[2]}, $p;
			local $/ = '\0';
			while(<$F>) {
				chomp;
				my($m,$p) = split(/\t/,$_,2);
				my($mode,$type,$sha1) = split(/ /,$m);
				next if $type ne "blob";
				push(@new,[$mode,$sha1,$p]);
			}
		} else {
			die "$revision: unknown action '".$action->[0]."' for $path\n";
		}
	}

	if(@old) {
		open my $F, "-|", "git-ls-files", "-z", @old or die $!;
		@old = ();
		local $/ = '\0';
		while(<$F>) {
			chomp;
			push(@old,$_);
		}
		close($F);

		while(@old) {
			my @o2;
			if(@old > 55) {
				@o2 = splice(@old,0,50);
			} else {
				@o2 = @old;
				@old = ();
			}
			system("git-update-index","--force-remove","--",@o2);
			die "Cannot remove files: $?\n" if $?;
		}
	}
	while(@new) {
		my @n2;
		if(@new > 12) {
			@n2 = splice(@new,0,10);
		} else {
			@n2 = @new;
			@new = ();
		}
		system("git-update-index","--add",
			(map { ('--cacheinfo', @$_) } @n2));
		die "Cannot add files: $?\n" if $?;
	}

	my $pid = open(C,"-|");
	die "Cannot fork: $!" unless defined $pid;
	unless($pid) {
		exec("git-write-tree");
		die "Cannot exec git-write-tree: $!\n";
	}
	chomp(my $tree = <C>);
	length($tree) == 40
		or die "Cannot get tree id ($tree): $!\n";
	close(C)
		or die "Error running git-write-tree: $?\n";
	print "Tree ID $tree\n" if $opt_v;

	my $pr = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	my $pw = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	$pid = fork();
	die "Fork: $!\n" unless defined $pid;
	unless($pid) {
		$pr->writer();
		$pw->reader();
		open(OUT,">&STDOUT");
		dup2($pw->fileno(),0);
		dup2($pr->fileno(),1);
		$pr->close();
		$pw->close();

		my @par = ();
		@par = ("-p",$rev) if defined $rev;

		# loose detection of merges
		# based on the commit msg
		foreach my $rx (@mergerx) {
			if ($message =~ $rx) {
				my $mparent = $1;
				if ($mparent eq 'HEAD') { $mparent = $opt_o };
				if ( -e "$git_dir/refs/heads/$mparent") {
					$mparent = get_headref($mparent, $git_dir);
					push @par, '-p', $mparent;
					print OUT "Merge parent branch: $mparent\n" if $opt_v;
				}
			} 
		}

		exec("env",
			"GIT_AUTHOR_NAME=$author_name",
			"GIT_AUTHOR_EMAIL=$author_email",
			"GIT_AUTHOR_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
			"GIT_COMMITTER_NAME=$author_name",
			"GIT_COMMITTER_EMAIL=$author_email",
			"GIT_COMMITTER_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
			"git-commit-tree", $tree,@par);
		die "Cannot exec git-commit-tree: $!\n";
	}
	$pw->writer();
	$pr->reader();

	$message =~ s/[\s\n]+\z//;

	print $pw "$message\n"
		or die "Error writing to git-commit-tree: $!\n";
	$pw->close();

	print "Committed change $revision:$branch ".strftime("%Y-%m-%d %H:%M:%S",gmtime($date)).")\n" if $opt_v;
	chomp(my $cid = <$pr>);
	length($cid) == 40
		or die "Cannot get commit id ($cid): $!\n";
	print "Commit ID $cid\n" if $opt_v;
	$pr->close();

	waitpid($pid,0);
	die "Error running git-commit-tree: $?\n" if $?;

	if(not defined $dest) {
		print "... no known parent\n" if $opt_v;
	} elsif(not $tag) {
		print "Writing to refs/heads/$dest\n" if $opt_v;
		open(C,">$git_dir/refs/heads/$dest") and 
		print C ("$cid\n") and
		close(C)
			or die "Cannot write branch $dest for update: $!\n";
	}
	$branches{$branch}{"LAST"} = $cid;
	$branches{$branch}{$revision} = $cid;
	$last_rev = $cid;
	print BRANCHES "$revision $branch $cid\n";
	print "DONE: $revision $dest $cid\n" if $opt_v;

	if($tag) {
		my($in, $out) = ('','');
		$last_rev = "-" if %$changed_paths;
		# the tag was 'complex', i.e. did not refer to a "real" revision
		
		$dest =~ tr/_/\./ if $opt_u;

		my $pid = open2($in, $out, 'git-mktag');
		print $out ("object $cid\n".
		    "type commit\n".
		    "tag $dest\n".
		    "tagger $author_name <$author_email>\n") and
		close($out)
		    or die "Cannot create tag object $dest: $!\n";

		my $tagobj = <$in>;
		chomp $tagobj;

		if ( !close($in) or waitpid($pid, 0) != $pid or
				$? != 0 or $tagobj !~ /^[0123456789abcdef]{40}$/ ) {
			die "Cannot create tag object $dest: $!\n";
		}

		open(C,">$git_dir/refs/tags/$dest") and
		print C ("$tagobj\n") and
		close(C)
			or die "Cannot create tag $branch: $!\n";

		print "Created tag '$dest' on '$branch'\n" if $opt_v;
	}
}

my ($changed_paths, $revision, $author, $date, $message, $pool) = @_;
sub _commit_all {
	($changed_paths, $revision, $author, $date, $message, $pool) = @_;
	my %p;
	while(my($path,$action) = each %$changed_paths) {
		$p{$path} = [ $action->action,$action->copyfrom_path, $action->copyfrom_rev ];
	}
	$changed_paths = \%p;
}

sub commit_all {
	my %done;
	my @col;
	my $pref;
	my $branch;

	while(my($path,$action) = each %$changed_paths) {
		($branch,$path) = split_path($revision,$path);
		next if not defined $branch;
		$done{$branch}{$path} = $action;
	}
	while(($branch,$changed_paths) = each %done) {
		commit($branch, $changed_paths, $revision, $author, $date, $message);
	}
}

while(++$current_rev < $svn->{'maxrev'}) {
	$svn->{'svn'}->get_log("/",$current_rev,$current_rev,$current_rev,1,1,\&_commit_all,"");
	commit_all();
}


unlink($git_index);

if (defined $orig_git_index) {
	$ENV{GIT_INDEX_FILE} = $orig_git_index;
} else {
	delete $ENV{GIT_INDEX_FILE};
}

# Now switch back to the branch we were in before all of this happened
if($orig_branch) {
	print "DONE\n" if $opt_v;
	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
		if $forward_master;
	unless ($opt_i) {
		system('git-read-tree', '-m', '-u', 'SVN2GIT_HEAD', 'HEAD');
		die "read-tree failed: $?\n" if $?;
	}
} else {
	$orig_branch = "master";
	print "DONE; creating $orig_branch branch\n" if $opt_v;
	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
		unless -f "$git_dir/refs/heads/master";
	unlink("$git_dir/HEAD");
	symlink("refs/heads/$orig_branch","$git_dir/HEAD");
	unless ($opt_i) {
		system('git checkout');
		die "checkout failed: $?\n" if $?;
	}
}
unlink("$git_dir/SVN2GIT_HEAD");
close(BRANCHES);
Commit	Line	Data
eaf718f3 MU	1	#!/usr/bin/perl -w
	2
	3	# This tool is copyright (c) 2005, Matthias Urlichs.
	4	# It is released under the Gnu Public License, version 2.
	5	#
	6	# The basic idea is to pull and analyze SVN changes.
	7	#
	8	# Checking out the files is done by a single long-running CVS connection
	9	# / server process.
	10	#
	11	# The head revision is on branch "origin" by default.
	12	# You can change that with the '-o' option.
	13
	14	require v5.8.0; # for shell-safe open("-\|",LIST)
	15	use strict;
	16	use warnings;
	17	use Getopt::Std;
	18	use File::Spec;
	19	use File::Temp qw(tempfile);
	20	use File::Path qw(mkpath);
	21	use File::Basename qw(basename dirname);
	22	use Time::Local;
	23	use IO::Pipe;
	24	use POSIX qw(strftime dup2);
	25	use IPC::Open2;
	26	use SVN::Core;
	27	use SVN::Ra;
	28
37dcf6de MU	29	die "Need CVN:COre 1.2.1 or better" if $SVN::Core::VERSION lt "1.2.1";
37dcf6de MU	30
eaf718f3 MU	31	$SIG{'PIPE'}="IGNORE";
	32	$ENV{'TZ'}="UTC";
	33
	34	our($opt_h,$opt_o,$opt_v,$opt_u,$opt_C,$opt_i,$opt_m,$opt_M,$opt_t,$opt_T,$opt_b);
	35
	36	sub usage() {
	37	print STDERR <<END;
	38	Usage: ${\basename $0} # fetch/update GIT from CVS
	39	[-o branch-for-HEAD] [-h] [-v]
	40	[-C GIT_repository] [-t tagname] [-T trunkname] [-b branchname]
	41	[-i] [-u] [-s subst] [-m] [-M regex] [SVN_URL]
	42	END
	43	exit(1);
	44	}
	45
	46	getopts("b:C:hivmM:o:t:T:u") or usage();
	47	usage if $opt_h;
	48
	49	my $tag_name = $opt_t \|\| "tags";
	50	my $trunk_name = $opt_T \|\| "trunk";
	51	my $branch_name = $opt_b \|\| "branches";
	52
	53	@ARGV <= 1 or usage();
	54
	55	$opt_o \|\|= "origin";
	56	my $git_tree = $opt_C;
	57	$git_tree \|\|= ".";
	58
	59	my $cvs_tree;
	60	if ($#ARGV == 0) {
	61	$cvs_tree = $ARGV[0];
	62	} elsif (-f 'CVS/Repository') {
	63	open my $f, '<', 'CVS/Repository' or
	64	die 'Failed to open CVS/Repository';
	65	$cvs_tree = <$f>;
	66	chomp $cvs_tree;
	67	close $f;
	68	} else {
	69	usage();
	70	}
	71
	72	our @mergerx = ();
	73	if ($opt_m) {
	74	@mergerx = ( qr/\W(?:from\|of\|merge\|merging\|merged) (\w+)/i );
	75	}
	76	if ($opt_M) {
	77	push (@mergerx, qr/$opt_M/);
	78	}
	79
	80	select(STDERR); $\|=1; select(STDOUT);
	81
	82
	83	package SVNconn;
	84	# Basic SVN connection.
	85	# We're only interested in connecting and downloading, so ...
	86
	87	use File::Spec;
	88	use File::Temp qw(tempfile);
	89	use POSIX qw(strftime dup2);
	90
	91	sub new {
	92	my($what,$repo) = @_;
	93	$what=ref($what) if ref($what);
	94
95	my $self = {};
96	$self->{'buffer'} = "";
97	bless($self,$what);
98
99	$repo =~ s#/+$##;
100	$self->{'fullrep'} = $repo;
101	$self->conn();
102
eaf718f3 MU	103	return $self;
	104	}
	105
	106	sub conn {
	107	my $self = shift;
	108	my $repo = $self->{'fullrep'};
	109	my $s = SVN::Ra->new($repo);
	110
	111	die "SVN connection to $repo: $!\n" unless defined $s;
	112	$self->{'svn'} = $s;
	113	$self->{'repo'} = $repo;
	114	$self->{'maxrev'} = $s->get_latest_revnum();
	115	}
	116
	117	sub file {
	118	my($self,$path,$rev) = @_;
	119	my $res;
	120
	121	my ($fh, $name) = tempfile('gitsvn.XXXXXX',
	122	DIR => File::Spec->tmpdir(), UNLINK => 1);
	123
2b5e63d1	124	print "... $rev $path ...\n" if $opt_v;
f02b3eba	125	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
8cd4177d	126	if ($@ and $@ !~ /Attempted to get checksum/) {
eaf718f3 MU	127	# retry
eaf718f3 MU	128	$self->conn();
2b5e63d1	129	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
eaf718f3	130	};
8cd4177d	131	return () if $@ and $@ !~ /Attempted to get checksum/;
2b5e63d1	132	die $@ if $@;
eaf718f3 MU	133	close ($fh);
	134
	135	return ($name, $res);
	136	}
	137
	138
	139	package main;
	140
	141	my $svn = SVNconn->new($cvs_tree);
	142
	143
	144	sub pdate($) {
	145	my($d) = @_;
	146	$d =~ m#(\d\d\d\d)-(\d\d)-(\d\d)T(\d\d):(\d\d):(\d\d)#
	147	or die "Unparseable date: $d\n";
	148	my $y=$1; $y-=1900 if $y>1900;
	149	return timegm($6\|\|0,$5,$4,$3,$2-1,$y);
	150	}
	151
eaf718f3 MU	152	sub getwd() {
	153	my $pwd = `pwd`;
	154	chomp $pwd;
	155	return $pwd;
	156	}
	157
	158
	159	sub get_headref($$) {
	160	my $name = shift;
	161	my $git_dir = shift;
	162	my $sha;
	163
	164	if (open(C,"$git_dir/refs/heads/$name")) {
	165	chomp($sha = <C>);
	166	close(C);
	167	length($sha) == 40
	168	or die "Cannot get head id for $name ($sha): $!\n";
	169	}
	170	return $sha;
	171	}
	172
	173
	174	-d $git_tree
	175	or mkdir($git_tree,0777)
	176	or die "Could not create $git_tree: $!";
	177	chdir($git_tree);
	178
	179	my $orig_branch = "";
	180	my $forward_master = 0;
	181	my %branches;
	182
	183	my $git_dir = $ENV{"GIT_DIR"} \|\| ".git";
	184	$git_dir = getwd()."/".$git_dir unless $git_dir =~ m#^/#;
	185	$ENV{"GIT_DIR"} = $git_dir;
	186	my $orig_git_index;
	187	$orig_git_index = $ENV{GIT_INDEX_FILE} if exists $ENV{GIT_INDEX_FILE};
	188	my ($git_ih, $git_index) = tempfile('gitXXXXXX', SUFFIX => '.idx',
	189	DIR => File::Spec->tmpdir());
	190	close ($git_ih);
	191	$ENV{GIT_INDEX_FILE} = $git_index;
	192	my $maxnum = 0;
	193	my $last_rev = "";
	194	my $last_branch;
	195	my $current_rev = 0;
	196	unless(-d $git_dir) {
	197	system("git-init-db");
	198	die "Cannot init the GIT db at $git_tree: $?\n" if $?;
	199	system("git-read-tree");
	200	die "Cannot init an empty tree: $?\n" if $?;
	201
	202	$last_branch = $opt_o;
	203	$orig_branch = "";
	204	} else {
	205	-f "$git_dir/refs/heads/$opt_o"
	206	or die "Branch '$opt_o' does not exist.\n".
	207	"Either use the correct '-o branch' option,\n".
	208	"or import to a new repository.\n";
	209
	210	-f "$git_dir/svn2git"
	211	or die "'$git_dir/svn2git' does not exist.\n".
	212	"You need that file for incremental imports.\n";
	213	$last_branch = basename(readlink("$git_dir/HEAD"));
	214	unless($last_branch) {
	215	warn "Cannot read the last branch name: $! -- assuming 'master'\n";
216	$last_branch = "master";
217	}
218	$orig_branch = $last_branch;
219	$last_rev = get_headref($orig_branch, $git_dir);
220	if (-f "$git_dir/SVN2GIT_HEAD") {
221	die <<EOM;
222	SVN2GIT_HEAD exists.
223	Make sure your working directory corresponds to HEAD and remove SVN2GIT_HEAD.
224	You may need to run
225
226	git-read-tree -m -u SVN2GIT_HEAD HEAD
227	EOM
228	}
229	system('cp', "$git_dir/HEAD", "$git_dir/SVN2GIT_HEAD");
230
231	$forward_master =
232	$opt_o ne 'master' && -f "$git_dir/refs/heads/master" &&
233	system('cmp', '-s', "$git_dir/refs/heads/master",
234	"$git_dir/refs/heads/$opt_o") == 0;
235
236	# populate index
237	system('git-read-tree', $last_rev);
238	die "read-tree failed: $?\n" if $?;
239
240	# Get the last import timestamps
241	open my $B,"<", "$git_dir/svn2git";
242	while(<$B>) {
243	chomp;
244	my($num,$branch,$ref) = split;
245	$branches{$branch}{$num} = $ref;
246	$branches{$branch}{"LAST"} = $ref;
247	$current_rev = $num+1 if $current_rev < $num+1;
248	}
249	close($B);
250	}
251	-d $git_dir
252	or die "Could not create git subdir ($git_dir).\n";
253
254	open BRANCHES,">>", "$git_dir/svn2git";
255
256
257	## cvsps output:
258	#---------------------
259	#PatchSet 314
260	#Date: 1999/09/18 13:03:59
261	#Author: wkoch
262	#Branch: STABLE-BRANCH-1-0
263	#Ancestor branch: HEAD
264	#Tag: (none)
265	#Log:
266	# See ChangeLog: Sat Sep 18 13:03:28 CEST 1999 Werner Koch
267	#Members:
268	# README:1.57->1.57.2.1
269	# VERSION:1.96->1.96.2.1
270	#
271	#---------------------
272
273	my $state = 0;
274
275	sub get_file($$$) {
276	my($rev,$branch,$path) = @_;
277
278	# revert split_path(), below
279	my $svnpath;
280	$path = "" if $path eq "/"; # this should not happen, but ...
281	if($branch eq "/") {
282	$svnpath = "/$trunk_name/$path";
283	} elsif($branch =~ m#^/#) {
284	$svnpath = "/$tag_name$branch/$path";
285	} else {
286	$svnpath = "/$branch_name/$branch/$path";
287	}
288
289	# now get it
2b5e63d1 MU	290	my ($name, $res) = eval { $svn->file($svnpath,$rev); };
2b5e63d1 MU	291	return () unless defined $name;
eaf718f3	292
6d281217	293	open my $F, '-\|', "git-hash-object", "-w", $name
eaf718f3 MU	294	or die "Cannot create object: $!\n";
	295	my $sha = <$F>;
	296	chomp $sha;
	297	close $F;
eaf718f3 MU	298	my $mode = "0644"; # SV does not seem to store any file modes
	299	return [$mode, $sha, $path];
	300	}
	301
	302	sub split_path($$) {
	303	my($rev,$path) = @_;
	304	my $branch;
	305
	306	if($path =~ s#^/\Q$tag_name\E/([^/]+)/?##) {
	307	$branch = "/$1";
	308	} elsif($path =~ s#^/\Q$trunk_name\E/?##) {
	309	$branch = "/";
	310	} elsif($path =~ s#^/\Q$branch_name\E/([^/]+)/?##) {
	311	$branch = $1;
	312	} else {
	313	print STDERR "$rev: Unrecognized path: $path\n";
	314	return ()
	315	}
	316	$path = "/" if $path eq "";
	317	return ($branch,$path);
	318	}
	319
	320	sub commit {
	321	my($branch, $changed_paths, $revision, $author, $date, $message) = @_;
	322	my($author_name,$author_email,$dest);
	323	my(@old,@new);
	324
2b5e63d1 MU	325	if (not defined $author) {
	326	$author_name = $author_email = "unknown";
	327	} elsif ($author =~ /^(.?)\s+<(.)>$/) {
eaf718f3 MU	328	($author_name, $author_email) = ($1, $2);
	329	} else {
	330	$author =~ s/^<(.*)>$/$1/;
	331	$author_name = $author_email = $author;
	332	}
	333	$date = pdate($date);
	334
	335	my $tag;
	336	my $parent;
	337	if($branch eq "/") { # trunk
	338	$parent = $opt_o;
	339	} elsif($branch =~ m#^/(.+)#) { # tag
	340	$tag = 1;
	341	$parent = $1;
	342	} else { # "normal" branch
	343	# nothing to do
	344	$parent = $branch;
	345	}
	346	$dest = $parent;
	347
	348	my $prev = $changed_paths->{"/"};
f0daa628	349	if($prev and $prev->[0] eq "A") {
eaf718f3	350	delete $changed_paths->{"/"};
f0daa628	351	my $oldpath = $prev->[1];
eaf718f3 MU	352	my $rev;
	353	if(defined $oldpath) {
	354	my $p;
	355	($parent,$p) = split_path($revision,$oldpath);
	356	if($parent eq "/") {
	357	$parent = $opt_o;
	358	} else {
	359	$parent =~ s#^/##; # if it's a tag
	360	}
	361	} else {
	362	$parent = undef;
	363	}
	364	}
	365
	366	my $rev;
	367	if(defined $parent) {
	368	open(H,"git-rev-parse --verify $parent \|");
	369	$rev = <H>;
	370	close(H) or do {
	371	print STDERR "$revision: cannot find commit '$parent'!\n";
	372	return;
	373	};
	374	chop $rev;
	375	if(length($rev) != 40) {
	376	print STDERR "$revision: cannot find commit '$parent'!\n";
	377	return;
	378	}
	379	$rev = $branches{($parent eq $opt_o) ? "/" : $parent}{"LAST"};
	380	if($revision != 1 and not $rev) {
	381	print STDERR "$revision: do not know ancestor for '$parent'!\n";
	382	return;
	383	}
	384	} else {
	385	$rev = undef;
	386	}
	387
f0daa628	388	# if($prev and $prev->[0] eq "A") {
eaf718f3 MU	389	# if(not $tag) {
	390	# unless(open(H,"> $git_dir/refs/heads/$branch")) {
	391	# print STDERR "$revision: Could not create branch $branch: $!\n";
	392	# $state=11;
	393	# next;
	394	# }
	395	# print H "$rev\n"
	396	# or die "Could not write branch $branch: $!";
	397	# close(H)
	398	# or die "Could not write branch $branch: $!";
	399	# }
	400	# }
	401	if(not defined $rev) {
	402	unlink($git_index);
	403	} elsif ($rev ne $last_rev) {
	404	print "Switching from $last_rev to $rev ($branch)\n" if $opt_v;
	405	system("git-read-tree", $rev);
	406	die "read-tree failed for $rev: $?\n" if $?;
	407	$last_rev = $rev;
	408	}
	409
	410	while(my($path,$action) = each %$changed_paths) {
f0daa628	411	if ($action->[0] eq "A") {
eaf718f3 MU	412	my $f = get_file($revision,$branch,$path);
eaf718f3 MU	413	push(@new,$f) if $f;
f0daa628	414	} elsif ($action->[0] eq "D") {
eaf718f3	415	push(@old,$path);
f0daa628	416	} elsif ($action->[0] eq "M") {
eaf718f3 MU	417	my $f = get_file($revision,$branch,$path);
eaf718f3 MU	418	push(@new,$f) if $f;
f0daa628	419	} elsif ($action->[0] eq "R") {
eaf718f3 MU	420	# refer to a file/tree in an earlier commit
	421	push(@old,$path); # remove any old stuff
	422
	423	# ... and add any new stuff
f0daa628 MU	424	my($b,$p) = split_path($revision,$action->[1]);
f0daa628 MU	425	open my $F,"-\|","git-ls-tree","-r","-z", $branches{$b}{$action->[2]}, $p;
eaf718f3 MU	426	local $/ = '\0';
	427	while(<$F>) {
	428	chomp;
	429	my($m,$p) = split(/\t/,$_,2);
	430	my($mode,$type,$sha1) = split(/ /,$m);
	431	next if $type ne "blob";
	432	push(@new,[$mode,$sha1,$p]);
	433	}
	434	} else {
f0daa628	435	die "$revision: unknown action '".$action->[0]."' for $path\n";
eaf718f3 MU	436	}
	437	}
	438
	439	if(@old) {
c7ff5f1d	440	open my $F, "-\|", "git-ls-files", "-z", @old or die $!;
eaf718f3 MU	441	@old = ();
eaf718f3 MU	442	local $/ = '\0';
6d281217	443	while(<$F>) {
eaf718f3 MU	444	chomp;
	445	push(@old,$_);
	446	}
6d281217	447	close($F);
eaf718f3 MU	448
	449	while(@old) {
	450	my @o2;
	451	if(@old > 55) {
	452	@o2 = splice(@old,0,50);
	453	} else {
	454	@o2 = @old;
	455	@old = ();
	456	}
	457	system("git-update-index","--force-remove","--",@o2);
	458	die "Cannot remove files: $?\n" if $?;
	459	}
	460	}
	461	while(@new) {
	462	my @n2;
	463	if(@new > 12) {
	464	@n2 = splice(@new,0,10);
	465	} else {
	466	@n2 = @new;
	467	@new = ();
	468	}
	469	system("git-update-index","--add",
	470	(map { ('--cacheinfo', @$_) } @n2));
	471	die "Cannot add files: $?\n" if $?;
	472	}
	473
	474	my $pid = open(C,"-\|");
	475	die "Cannot fork: $!" unless defined $pid;
	476	unless($pid) {
	477	exec("git-write-tree");
	478	die "Cannot exec git-write-tree: $!\n";
	479	}
	480	chomp(my $tree = <C>);
	481	length($tree) == 40
	482	or die "Cannot get tree id ($tree): $!\n";
	483	close(C)
	484	or die "Error running git-write-tree: $?\n";
	485	print "Tree ID $tree\n" if $opt_v;
	486
	487	my $pr = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	488	my $pw = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	489	$pid = fork();
	490	die "Fork: $!\n" unless defined $pid;
	491	unless($pid) {
	492	$pr->writer();
	493	$pw->reader();
	494	open(OUT,">&STDOUT");
	495	dup2($pw->fileno(),0);
	496	dup2($pr->fileno(),1);
	497	$pr->close();
	498	$pw->close();
	499
	500	my @par = ();
	501	@par = ("-p",$rev) if defined $rev;
	502
	503	# loose detection of merges
	504	# based on the commit msg
	505	foreach my $rx (@mergerx) {
	506	if ($message =~ $rx) {
	507	my $mparent = $1;
	508	if ($mparent eq 'HEAD') { $mparent = $opt_o };
	509	if ( -e "$git_dir/refs/heads/$mparent") {
	510	$mparent = get_headref($mparent, $git_dir);
	511	push @par, '-p', $mparent;
512	print OUT "Merge parent branch: $mparent\n" if $opt_v;
513	}
514	}
515	}
516
517	exec("env",
518	"GIT_AUTHOR_NAME=$author_name",
519	"GIT_AUTHOR_EMAIL=$author_email",
520	"GIT_AUTHOR_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
521	"GIT_COMMITTER_NAME=$author_name",
522	"GIT_COMMITTER_EMAIL=$author_email",
523	"GIT_COMMITTER_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
524	"git-commit-tree", $tree,@par);
525	die "Cannot exec git-commit-tree: $!\n";
526	}
527	$pw->writer();
528	$pr->reader();
529
530	$message =~ s/[\s\n]+\z//;
531
532	print $pw "$message\n"
533	or die "Error writing to git-commit-tree: $!\n";
534	$pw->close();
535
536	print "Committed change $revision:$branch ".strftime("%Y-%m-%d %H:%M:%S",gmtime($date)).")\n" if $opt_v;
537	chomp(my $cid = <$pr>);
538	length($cid) == 40
539	or die "Cannot get commit id ($cid): $!\n";
540	print "Commit ID $cid\n" if $opt_v;
541	$pr->close();
542
543	waitpid($pid,0);
544	die "Error running git-commit-tree: $?\n" if $?;
545
f02b3eba MU	546	if(not defined $dest) {
	547	print "... no known parent\n" if $opt_v;
	548	} elsif(not $tag) {
eaf718f3 MU	549	print "Writing to refs/heads/$dest\n" if $opt_v;
	550	open(C,">$git_dir/refs/heads/$dest") and
	551	print C ("$cid\n") and
	552	close(C)
	553	or die "Cannot write branch $dest for update: $!\n";
eaf718f3 MU	554	}
	555	$branches{$branch}{"LAST"} = $cid;
	556	$branches{$branch}{$revision} = $cid;
	557	$last_rev = $cid;
	558	print BRANCHES "$revision $branch $cid\n";
	559	print "DONE: $revision $dest $cid\n" if $opt_v;
	560
	561	if($tag) {
	562	my($in, $out) = ('','');
	563	$last_rev = "-" if %$changed_paths;
	564	# the tag was 'complex', i.e. did not refer to a "real" revision
	565
f02b3eba	566	$dest =~ tr/_/\./ if $opt_u;
eaf718f3 MU	567
	568	my $pid = open2($in, $out, 'git-mktag');
	569	print $out ("object $cid\n".
	570	"type commit\n".
f02b3eba	571	"tag $dest\n".
eaf718f3 MU	572	"tagger $author_name <$author_email>\n") and
eaf718f3 MU	573	close($out)
f02b3eba	574	or die "Cannot create tag object $dest: $!\n";
eaf718f3 MU	575
	576	my $tagobj = <$in>;
	577	chomp $tagobj;
	578
	579	if ( !close($in) or waitpid($pid, 0) != $pid or
	580	$? != 0 or $tagobj !~ /^[0123456789abcdef]{40}$/ ) {
f02b3eba	581	die "Cannot create tag object $dest: $!\n";
eaf718f3	582	}
eaf718f3	583
f02b3eba MU	584	open(C,">$git_dir/refs/tags/$dest") and
f02b3eba MU	585	print C ("$tagobj\n") and
eaf718f3	586	close(C)
f02b3eba	587	or die "Cannot create tag $branch: $!\n";
eaf718f3	588
f02b3eba	589	print "Created tag '$dest' on '$branch'\n" if $opt_v;
eaf718f3 MU	590	}
	591	}
	592
	593	my ($changed_paths, $revision, $author, $date, $message, $pool) = @_;
	594	sub _commit_all {
	595	($changed_paths, $revision, $author, $date, $message, $pool) = @_;
f0daa628 MU	596	my %p;
	597	while(my($path,$action) = each %$changed_paths) {
	598	$p{$path} = [ $action->action,$action->copyfrom_path, $action->copyfrom_rev ];
	599	}
	600	$changed_paths = \%p;
eaf718f3	601	}
f0daa628	602
eaf718f3 MU	603	sub commit_all {
	604	my %done;
	605	my @col;
	606	my $pref;
	607	my $branch;
	608
	609	while(my($path,$action) = each %$changed_paths) {
	610	($branch,$path) = split_path($revision,$path);
	611	next if not defined $branch;
	612	$done{$branch}{$path} = $action;
	613	}
	614	while(($branch,$changed_paths) = each %done) {
	615	commit($branch, $changed_paths, $revision, $author, $date, $message);
	616	}
	617	}
	618
	619	while(++$current_rev < $svn->{'maxrev'}) {
	620	$svn->{'svn'}->get_log("/",$current_rev,$current_rev,$current_rev,1,1,\&_commit_all,"");
	621	commit_all();
	622	}
	623
	624
	625	unlink($git_index);
	626
	627	if (defined $orig_git_index) {
	628	$ENV{GIT_INDEX_FILE} = $orig_git_index;
	629	} else {
	630	delete $ENV{GIT_INDEX_FILE};
	631	}
	632
	633	# Now switch back to the branch we were in before all of this happened
	634	if($orig_branch) {
	635	print "DONE\n" if $opt_v;
	636	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
	637	if $forward_master;
	638	unless ($opt_i) {
	639	system('git-read-tree', '-m', '-u', 'SVN2GIT_HEAD', 'HEAD');
	640	die "read-tree failed: $?\n" if $?;
	641	}
	642	} else {
	643	$orig_branch = "master";
	644	print "DONE; creating $orig_branch branch\n" if $opt_v;
	645	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
	646	unless -f "$git_dir/refs/heads/master";
	647	unlink("$git_dir/HEAD");
	648	symlink("refs/heads/$orig_branch","$git_dir/HEAD");
	649	unless ($opt_i) {
	650	system('git checkout');
	651	die "checkout failed: $?\n" if $?;
	652	}
	653	}
	654	unlink("$git_dir/SVN2GIT_HEAD");
	655	close(BRANCHES);