Merge lp:~jelmer/dulwich/httprepo into lp:~vcs-imports/dulwich/trunk

[Jelmer Vernooij]

HTTP Dumb access support

[lp:~jelmer/dulwich/httprepo]

462. By Jelmer Vernooij on 2010-06-08

merge trunk

Unmerged revisions

462. By Jelmer Vernooij on 2010-06-08

merge trunk

461. By Jelmer Vernooij on 2010-02-12

Add basic dumb HTTP repository implementation.

460. By Jelmer Vernooij on 2010-02-12

Remove one-line function, only used in one place.

459. By Jelmer Vernooij on 2010-02-12

Remove put_named_file from public API

458. By Jelmer Vernooij on 2010-02-12

Fix PackData._get_size().

457. By Jelmer Vernooij on 2010-02-12

Delay checking size until it's actually used.

456. By Jelmer Vernooij on 2010-02-12

Add convenience function for parsing info/refs file.

455. By Jelmer Vernooij on 2010-02-12

Extend DictRefsContainer.

454. By Jelmer Vernooij on 2010-02-12

Mark current version as 0.5.0

453. By Jelmer Vernooij on 2010-02-12

merge support for use with in-memory packs.

=== added file 'dulwich/httprepo.py'
--- dulwich/httprepo.py	1970-01-01 00:00:00 +0000
+++ dulwich/httprepo.py	2010-06-08 19:26:25 +0000
@@ -0,0 +1,127 @@
+# Copyright (C) 2010 Jelmer Vernooij <[e-mail]>
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+
+"""(dumb) HTTP repository access."""
+
+import urllib
+
+from dulwich.errors import (
+    NotGitRepository,
+    NoIndexPresent,
+    )
+from dulwich.objects import (
+    ShaFile,
+    )
+from dulwich.object_store import (
+    PackBasedObjectStore,
+    PACKDIR,
+    )
+from dulwich.pack import (
+    PackData,
+    Pack,
+    load_pack_index_file,
+    )
+from dulwich.repo import (
+    BaseRepo,
+    DictRefsContainer,
+    OBJECTDIR,
+    read_info_refs,
+    )
+
+
+class HTTPRepo(BaseRepo):
+
+    def __init__(self, baseurl):
+        self._opener = urllib.URLopener()
+        self.baseurl = baseurl.rstrip("/") + "/"
+        refsf = self.get_named_file(".git/info/refs")
+        if refsf:
+            self.bare = False
+            self._controlurl = urllib.basejoin(self.baseurl, '.git')
+        else:
+            refsf = self.get_named_file("info/refs")
+            if refsf is None:
+                raise NotGitRepository(self.baseurl)
+            self.bare = True
+            self._controlurl = self.baseurl
+        object_store = HTTPObjectStore(self._opener,
+            urllib.basejoin(self.baseurl, OBJECTDIR))
+        refs = read_info_refs(refsf)
+        refs["HEAD"] = self.get_named_file("HEAD").read().rstrip("\n")
+        super(HTTPRepo, self).__init__(object_store, 
+                DictRefsContainer(refs))
+
+    def get_named_file(self, path):
+        """Get a file from the control dir with a specific name.
+
+        Although the filename should be interpreted as a filename relative to
+        the control dir in a disk-baked Repo, the object returned need not be
+        pointing to a file in that location.
+
+        :param path: The path to the file, relative to the control dir.
+        :return: An open file object, or None if the file does not exist.
+        """
+        try:
+            return self._opener.open(urllib.basejoin(self.baseurl, path.lstrip('/')))
+        except IOError:
+            # FIXME: Be more specific than *any* IOError?
+            return None
+
+    def open_index(self):
+        """Open the index for this repository."""
+        raise NoIndexPresent()
+
+    def __repr__(self):
+        return "<HTTPRepo for %r>" % self.baseurl
+
+
+class HTTPObjectStore(PackBasedObjectStore):
+    """Git-style object store that exists on disk."""
+
+    def __init__(self, opener, url):
+        """Open an object store.
+
+        """
+        super(HTTPObjectStore, self).__init__()
+        self._opener = opener
+        self.url = url.rstrip("/") + "/"
+        self.pack_url = urllib.basejoin(self.url, PACKDIR).rstrip("/")+"/"
+
+    def _load_packs(self):
+        ret = []
+        packs_index = urllib.basejoin(self.url, 'info/packs')
+        for line in self._opener.open(packs_index).readlines():
+            line = line.rstrip("\n")
+            if not line:
+                continue
+            (kind, name) = line.split(" ", 1)
+            if kind != "P":
+                continue
+            if name.startswith("pack-") and name.endswith(".pack"):
+                pack_url = urllib.basejoin(self.pack_url, name)
+                pd = PackData(name, self._opener.open(pack_url))
+                idxname = name.replace(".pack", ".idx")
+                index_url = urllib.basejoin(self.pack_url, idxname)
+                idx = load_pack_index_file(idxname, self._opener.open(index_url))
+                ret.append(Pack.from_objects(pd, idx))
+        return ret
+
+    def _split_loose_object(self, sha):
+        return (sha[:2], sha[2:])
+
+    def _get_loose_object(self, sha):
+        path = '%s/%s' % self._split_loose_object(sha)
+        return ShaFile._parse_file(self._opener.open(urllib.basejoin(self.url, path)).read())