Support catpipe get(...,size=True); require git >= 1.5.6

Add a new size argument to get() that requests the object size in
addition to the type.  To support this, require git 1.5.6 (circa 2008)
or newer so that we'll have cat-file --batch.

Remove the _slow_get() fallback since it's no longer needed.

Signed-off-by: Rob Browning <rlb@defaultvalue.org>
Tested-by: Rob Browning <rlb@defaultvalue.org>

diff --git a/lib/bup/git.py b/lib/bup/git.py
index 950a2d4ebb41e0399851d8b71ee49d6e9c301d5b..94016a4c7ec59af671b429af382a04e7b65428b3 100644 (file)
--- a/lib/bup/git.py
+++ b/lib/bup/git.py
@@ -1138,14 +1138,9 @@ class CatPipe:
         self.repo_dir = repo_dir
         wanted = ('1','5','6')
         if ver() < wanted:
-            if not _ver_warned:
-                log('warning: git version < %s; bup will be slow.\n'
-                    % '.'.join(wanted))
-                _ver_warned = 1
-            self.get = self._slow_get
-        else:
-            self.p = self.inprogress = None
-            self.get = self._fast_get
+            log('error: git version must be at least 1.5.6\n')
+            sys.exit(1)
+        self.p = self.inprogress = None
 
     def _abort(self):
         if self.p:
@@ -1163,15 +1158,19 @@ class CatPipe:
                                   bufsize = 4096,
                                   preexec_fn = _gitenv(self.repo_dir))
 
-    def _fast_get(self, id):
+    def get(self, id, size=False):
+        """Yield the object type, and then an iterator over the data referred
+        to by the id ref.  If size is true, yield (obj_type, obj_size)
+        instead of just the type.
+
+        """
         if not self.p or self.p.poll() != None:
             self.restart()
         assert(self.p)
         poll_result = self.p.poll()
         assert(poll_result == None)
         if self.inprogress:
-            log('_fast_get: opening %r while %r is open\n'
-                % (id, self.inprogress))
+            log('get: opening %r while %r is open\n' % (id, self.inprogress))
         assert(not self.inprogress)
         assert(id.find('\n') < 0)
         assert(id.find('\r') < 0)
@@ -1186,12 +1185,15 @@ class CatPipe:
         spl = hdr.split(' ')
         if len(spl) != 3 or len(spl[0]) != 40:
             raise GitError('expected blob, got %r' % spl)
-        (hex, type, size) = spl
-
-        it = _AbortableIter(chunkyreader(self.p.stdout, int(spl[2])),
-                           onabort = self._abort)
+        hex, typ, sz = spl
+        sz = int(sz)
+        it = _AbortableIter(chunkyreader(self.p.stdout, sz),
+                            onabort=self._abort)
         try:
-            yield type
+            if size:
+                yield typ, sz
+            else:
+                yield typ
             for blob in it:
                 yield blob
             readline_result = self.p.stdout.readline()
@@ -1201,20 +1203,6 @@ class CatPipe:
             it.abort()
             raise
 
-    def _slow_get(self, id):
-        assert(id.find('\n') < 0)
-        assert(id.find('\r') < 0)
-        assert(id[0] != '-')
-        type = _git_capture(['git', 'cat-file', '-t', id]).strip()
-        yield type
-
-        p = subprocess.Popen(['git', 'cat-file', type, id],
-                             stdout=subprocess.PIPE,
-                             preexec_fn = _gitenv(self.repo_dir))
-        for blob in chunkyreader(p.stdout):
-            yield blob
-        _git_wait('git cat-file', p)
-
     def _join(self, it):
         type = it.next()
         if type == 'blob':

diff --git a/lib/bup/t/tgit.py b/lib/bup/t/tgit.py
index 0293153b8bf7c211b17639304bd36ccbc59d4419..d12924c07246ae4b052d1dd50e7fe35896b16fc2 100644 (file)
--- a/lib/bup/t/tgit.py
+++ b/lib/bup/t/tgit.py
@@ -414,3 +414,30 @@ def test__git_date_str():
         WVPASSEQ('0 +0000', git._git_date_str(0, 0))
         WVPASSEQ('0 -0130', git._git_date_str(0, -90 * 60))
         WVPASSEQ('0 +0130', git._git_date_str(0, 90 * 60))
+
+
+@wvtest
+def test_cat_pipe():
+    with no_lingering_errors():
+        with test_tempdir('bup-tgit-') as tmpdir:
+            os.environ['BUP_MAIN_EXE'] = bup_exe
+            os.environ['BUP_DIR'] = bupdir = tmpdir + "/bup"
+            src = tmpdir + '/src'
+            mkdirp(src)
+            with open(src + '/1', 'w+') as f:
+                print f, 'something'
+            with open(src + '/2', 'w+') as f:
+                print f, 'something else'
+            git.init_repo(bupdir)
+            exc(bup_exe, 'index', src)
+            exc(bup_exe, 'save', '-n', 'src', '--strip', src)
+            git_type = exo('git', '--git-dir', bupdir,
+                           'cat-file', '-t', 'src').strip()
+            git_size = int(exo('git', '--git-dir', bupdir,
+                               'cat-file', '-s', 'src'))
+            it = git.cp().get('src', size=True)
+            get_type, get_size = it.next()
+            for buf in it.next():
+                pass
+            WVPASSEQ(get_type, git_type)
+            WVPASSEQ(get_size, git_size)