svn import: Add direct HTTP access

Some SVN repositories that are accessible through HTTP don't like when I
retrieve files using SVN methods ("internal server error").

Therefore, I added an option to get the contents using (persistent) HTTP
directly. This also reduces round-trip time, from two or three requests
down to one.

Also corrected error handling a bit.

Signed-Off-By: Matthias Urlichs <smurf@smurf.noris.de>

diff --git a/Documentation/git-svnimport.txt b/Documentation/git-svnimport.txt
index 047f8f334600253fefed5a8d6a5bc734d84f16f4..a27a83596c05e519ff36e65e760f6b8041852af7 100644 (file)
--- a/Documentation/git-svnimport.txt
+++ b/Documentation/git-svnimport.txt
@@ -9,10 +9,11 @@ git-svnimport - Import a SVN repository into git
 
 SYNOPSIS
 --------
-'git-svnimport' [ -o <branch-for-HEAD> ] [ -h ] [ -v ]
+'git-svnimport' [ -o <branch-for-HEAD> ] [ -h ] [ -v ] [ -d | -D ]
                        [ -C <GIT_repository> ] [ -i ] [ -u ] [-l limit_nr_changes]
                        [ -b branch_subdir ] [ -t trunk_subdir ] [ -T tag_subdir ]
-                       [ -s start_chg ] [ -m ] [ -M regex ] [ <SVN_repository_URL> ]
+                       [ -s start_chg ] [ -m ] [ -M regex ]
+                       <SVN_repository_URL> [ <path> ]
 
 
 DESCRIPTION
@@ -82,9 +83,32 @@ When importing incementally, you might need to edit the .git/svn2git file.
 -v::
        Verbosity: let 'svnimport' report what it is doing.
 
+-d::
+       Use direct HTTP requests if possible. The "<path>" argument is used
+       only for retrieving the SVN logs; the path to the contents is
+       included in the SVN log.
+
+-D::
+       Use direct HTTP requests if possible. The "<path>" argument is used
+       for retrieving the logs, as well as for the contents.
++
+There's no safe way to automatically find out which of these options to
+use, so you need to try both. Usually, the one that's wrong will die
+with a 40x error pretty quickly.
+
 <SVN_repository_URL>::
        The URL of the SVN module you want to import. For local
        repositories, use "file:///absolute/path".
++
+If you're using the "-d" or "-D" option, this is the URL of the SVN
+repository itself; it usually ends in "/svn".
+
+<SVN_repository_URL>::
+       The URL of the SVN module you want to import. For local
+       repositories, use "file:///absolute/path".
+
+<path>
+       The path to the module you want to check out.
 
 -h::
        Print a short usage message and exit.

diff --git a/git-svnimport.perl b/git-svnimport.perl
index b976841efd276595770a51562ad7ce9b1d51f743..2f89c31e54eac82a4f0fa04c3681a2378dda82d1 100755 (executable)
--- a/git-svnimport.perl
+++ b/git-svnimport.perl
@@ -30,26 +30,26 @@ 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,$opt_b,$opt_s,$opt_l);
+our($opt_h,$opt_o,$opt_v,$opt_u,$opt_C,$opt_i,$opt_m,$opt_M,$opt_t,$opt_T,$opt_b,$opt_s,$opt_l,$opt_d,$opt_D);
 
 sub usage() {
        print STDERR <<END;
 Usage: ${\basename $0}     # fetch/update GIT from CVS
        [-o branch-for-HEAD] [-h] [-v] [-l max_num_changes]
        [-C GIT_repository] [-t tagname] [-T trunkname] [-b branchname]
-       [-i] [-u] [-s start_chg] [-m] [-M regex] [SVN_URL]
+       [-d|-D] [-i] [-u] [-s start_chg] [-m] [-M regex] [SVN_URL]
 END
        exit(1);
 }
 
-getopts("b:C:hil:mM:o:s:t:T:uv") or usage();
+getopts("b:C:dDhil:mM:o:s:t:T:uv") 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();
+@ARGV == 1 or @ARGV == 2 or usage();
 
 $opt_o ||= "origin";
 $opt_s ||= 1;
@@ -58,6 +58,7 @@ my $git_tree = $opt_C;
 $git_tree ||= ".";
 
 my $svn_url = $ARGV[0];
+my $svn_dir = $ARGV[1];
 
 our @mergerx = ();
 if ($opt_m) {
@@ -106,30 +107,44 @@ sub conn {
 
 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 $@;
+       if($@) {
+               return undef if $@ =~ /Attempted to get checksum/;
+               die $@;
+       }
        close ($fh);
 
-       return ($name, $res);
+       return $name;
 }
 
-
 package main;
+use URI;
 
-my $svn = SVNconn->new($svn_url);
+my $svn = $svn_url;
+$svn .= "/$svn_dir" if defined $svn_dir;
+$svn = SVNconn->new($svn);
 
+my $lwp_ua;
+if($opt_d or $opt_D) {
+       $svn_url = URI->new($svn_url)->canonical;
+       if($opt_D) {
+               $svn_dir =~ s#/*$#/#;
+       } else {
+               $svn_dir = "";
+       }
+       if ($svn_url->scheme eq "http") {
+               use LWP::UserAgent;
+               $lwp_ua = LWP::UserAgent->new(keep_alive => 1, requests_redirectable => []);
+       } else {
+               print STDERR "Warning: not HTTP; turning off direct file access\n";
+               $opt_d=0;
+       }
+}
 
 sub pdate($) {
        my($d) = @_;
@@ -258,8 +273,30 @@ sub get_file($$$) {
        }
 
        # now get it
-       my ($name, $res) = eval { $svn->file($svnpath,$rev); };
-       return () unless defined $name;
+       my $name;
+       if($opt_d) {
+               my($req,$res);
+
+               # /svn/!svn/bc/2/django/trunk/django-docs/build.py
+               my $url=$svn_url->clone();
+               $url->path($url->path."/!svn/bc/$rev/$svn_dir$svnpath");
+               print "Fetching $url...\n" if $opt_v;
+               $req = HTTP::Request->new(GET => $url);
+               $res = $lwp_ua->request($req);
+               if ($res->is_success) {
+                       my $fh;
+                       ($fh, $name) = tempfile('gitsvn.XXXXXX', 
+                       DIR => File::Spec->tmpdir(), UNLINK => 1);
+                       print $fh $res->content;
+                       close($fh) or die "Could not write $name: $!\n";
+               } else {
+                       return undef if $res->code == 301; # directory?
+                       die $res->status_line." at $url\n";
+               }
+       } else {
+               $name = $svn->file($svnpath,$rev);
+               return undef unless defined $name;
+       }
 
        open my $F, '-|', "git-hash-object", "-w", $name
                or die "Cannot create object: $!\n";