Fixed odb performance tests

Author: Byron

git/test/performance/__init__.py

@@ -0,0 +1,2 @@
+"""Note: This init file makes the performance tests an integral part of the test suite
+as nose will now pick them up. Previously the init file was intentionally omitted"""

git/test/performance/db/__init__.py

@@ -0,0 +1 @@
+

git/test/performance/db/odb_impl.py

@@ -0,0 +1,69 @@
+"""Performance tests for object store"""
+
+from time import time
+import sys
+import stat
+import copy
+
+from git.test.performance.lib import (
+	TestBigRepoR,
+	GlobalsItemDeletorMetaCls
+	)
+
+class PerfBaseDeletorMetaClass(GlobalsItemDeletorMetaCls):
+	ModuleToDelete = 'TestObjDBPerformanceBase'
+	
+
+class TestObjDBPerformanceBase(TestBigRepoR):
+	__metaclass__ = PerfBaseDeletorMetaClass
+	
+	def test_random_access_test(self):
+		repo = self.rorepo
+	
+		# GET COMMITS
+		st = time()
+		root_commit = repo.commit(self.head_sha_2k)
+		commits = list(root_commit.traverse())
+		nc = len(commits)
+		elapsed = time() - st
+		
+		print >> sys.stderr, "%s: Retrieved %i commits from ObjectStore in %g s ( %f commits / s )" % (type(repo.odb), nc, elapsed, nc / elapsed)
+			
+		# GET TREES
+		# walk all trees of all commits
+		st = time()
+		blobs_per_commit = list()
+		nt = 0
+		for commit in commits:
+			tree = commit.tree
+			blobs = list()
+			for item in tree.traverse():
+				nt += 1
+				if item.type == 'blob':
+					blobs.append(item)
+				# direct access for speed
+			# END while trees are there for walking
+			blobs_per_commit.append(blobs)
+		# END for each commit
+		elapsed = time() - st
+		
+		print >> sys.stderr, "%s: Retrieved %i objects from %i commits in %g s ( %f objects / s )" % (type(repo.odb), nt, len(commits), elapsed, nt / elapsed)
+		
+		# GET BLOBS
+		st = time()
+		nb = 0
+		too_many = 15000
+		data_bytes = 0
+		for blob_list in blobs_per_commit:
+			for blob in blob_list:
+				data_bytes += len(blob.data_stream.read())
+			# END for each blobsha
+			nb += len(blob_list)
+			if nb > too_many:
+				break
+		# END for each bloblist
+		elapsed = time() - st
+		
+		print >> sys.stderr, "%s: Retrieved %i blob (%i KiB) and their data in %g s ( %f blobs / s, %f KiB / s )" % (type(repo.odb), nb, data_bytes/1000, elapsed, nb / elapsed, (data_bytes / 1000) / elapsed)
+		
+		

git/test/performance/db/test_odb_cmd.py

@@ -0,0 +1,6 @@
+from git.db.complex import CmdCompatibilityGitDB
+from odb_impl import TestObjDBPerformanceBase
+
+class TestCmdDB(TestObjDBPerformanceBase):
+	RepoCls = CmdCompatibilityGitDB
+	

git/test/performance/db/test_odb_pure.py

@@ -0,0 +1,6 @@
+from git.db.complex import PureCompatibilityGitDB
+from odb_impl import TestObjDBPerformanceBase
+
+class TestPureDB(TestObjDBPerformanceBase):
+	RepoCls = PureCompatibilityGitDB
+	

git/test/performance/lib.py

@@ -1,18 +1,12 @@
 """Contains library functions"""
 import os
-from git.test.lib import TestBase
+from git.test.lib import (
+						TestBase, 
+						GlobalsItemDeletorMetaCls
+						)
 import shutil
 import tempfile
 
-from git.db import (
-						CmdGitDB,
-						GitDB
-					)
-
-from git import (
-	Repo
-	)
-
 #{ Invvariants
 k_env_git_repo = "GIT_PYTHON_TEST_GIT_REPO_BASE"
 #} END invariants
@@ -38,41 +32,41 @@ class TestBigRepoR(TestBase):
 	
 	* gitrorepo
 	
-	 * Read-Only git repository - actually the repo of git itself
-	 
-    * puregitrorepo
-    
-     * As gitrepo, but uses pure python implementation
+	 * a big read-only git repository
     """
 
 	#{ Invariants
 	head_sha_2k = '235d521da60e4699e5bd59ac658b5b48bd76ddca'
 	head_sha_50 = '32347c375250fd470973a5d76185cac718955fd5'
 	#} END invariants 
 
+	#{ Configuration
+	RepoCls = None
+	#} END configuration
+	
 	@classmethod
 	def setUpAll(cls):
 		super(TestBigRepoR, cls).setUpAll()
-		repo_path = resolve_or_fail(k_env_git_repo)
-		cls.gitrorepo = Repo(repo_path, odbt=CmdGitDB)
-		cls.puregitrorepo = Repo(repo_path, odbt=GitDB)
+		if cls.RepoCls is None:
+			raise AssertionError("Require RepoCls in class %s to be set" % cls)
+		#END assert configuration
+		cls.rorepo = cls.RepoCls(resolve_or_fail(k_env_git_repo))
 
 
 class TestBigRepoRW(TestBigRepoR):
 	"""As above, but provides a big repository that we can write to.
 	
-	Provides ``self.gitrwrepo`` and ``self.puregitrwrepo``"""
+	Provides ``self.rwrepo``"""
 
 	@classmethod
 	def setUpAll(cls):
 		super(TestBigRepoRW, cls).setUpAll()
 		dirname = tempfile.mktemp()
 		os.mkdir(dirname)
-		cls.gitrwrepo = cls.gitrorepo.clone(dirname, shared=True, bare=True, odbt=CmdGitDB)
-		cls.puregitrwrepo = Repo(dirname, odbt=GitDB)
+		cls.rwrepo = cls.gitrorepo.clone(dirname, shared=True, bare=True)
 
 	@classmethod
 	def tearDownAll(cls):
-		shutil.rmtree(cls.gitrwrepo.working_dir)
+		shutil.rmtree(cls.rwrepo.working_dir)
 
 #} END base classes

git/test/performance/test_commit.py

@@ -70,9 +70,9 @@ def test_commit_iteration(self):
 		print >> sys.stderr, "Iterated %i Commits in %s [s] ( %f commits/s )" % (nc, elapsed_time, nc/elapsed_time)
 
 	def test_commit_serialization(self):
-		assert_commit_serialization(self.gitrwrepo, self.head_sha_2k, True)
+		assert_commit_serialization(self.rwrepo, self.head_sha_2k, True)
 
-		rwrepo = self.gitrwrepo
+		rwrepo = self.rwrepo
 		make_object = rwrepo.odb.store
 		# direct serialization - deserialization can be tested afterwards
 		# serialization is probably limited on IO