Initial interface including some of the implementation of the RefLog. TestCase scetched out for now

tests: Added tests to verify that objects don't have a dict. Previously, due to a missing __slots__ member in Serializable, most objects would indeed have a dict, although the opposite was intended

8 files changed, 255 insertions, 48 deletions

diff --git a/objects/util.py b/objects/util.py
index a9e1143c..dfaaf7c4 100644
--- a/objects/util.py
+++ b/objects/util.py
@@ -71,7 +71,7 @@ def get_user_id():
 
 def utctz_to_altz(utctz):
 	"""we convert utctz to the timezone in seconds, it is the format time.altzone
-	returns. Git stores it as UTC timezon which has the opposite sign as well, 
+	returns. Git stores it as UTC timezone which has the opposite sign as well, 
 	which explains the -1 * ( that was made explicit here )
 	:param utctz: git utc timezone string, i.e. +0200"""
 	return -1 * int(float(utctz)/100*3600)
@@ -195,53 +195,55 @@ def parse_actor_and_date(line):
 #{ Classes 
 
 class Actor(object):
-    """Actors hold information about a person acting on the repository. They 
-    can be committers and authors or anything with a name and an email as 
-    mentioned in the git log entries."""
-    # precompiled regex
-    name_only_regex = re.compile( r'<(.+)>' )
-    name_email_regex = re.compile( r'(.*) <(.+?)>' ) 
-    
-    def __init__(self, name, email):
-        self.name = name
-        self.email = email
+	"""Actors hold information about a person acting on the repository. They 
+	can be committers and authors or anything with a name and an email as 
+	mentioned in the git log entries."""
+	# precompiled regex
+	name_only_regex = re.compile( r'<(.+)>' )
+	name_email_regex = re.compile( r'(.*) <(.+?)>' )
+	
+	__slots__ = ('name', 'email')
+	
+	def __init__(self, name, email):
+		self.name = name
+		self.email = email
 
-    def __eq__(self, other):
-        return self.name == other.name and self.email == other.email
-        
-    def __ne__(self, other):
-        return not (self == other)
-        
-    def __hash__(self):
-        return hash((self.name, self.email))
+	def __eq__(self, other):
+		return self.name == other.name and self.email == other.email
+		
+	def __ne__(self, other):
+		return not (self == other)
+		
+	def __hash__(self):
+		return hash((self.name, self.email))
 
-    def __str__(self):
-        return self.name
+	def __str__(self):
+		return self.name
 
-    def __repr__(self):
-        return '<git.Actor "%s <%s>">' % (self.name, self.email)
+	def __repr__(self):
+		return '<git.Actor "%s <%s>">' % (self.name, self.email)
 
-    @classmethod
-    def _from_string(cls, string):
-        """Create an Actor from a string.
+	@classmethod
+	def _from_string(cls, string):
+		"""Create an Actor from a string.
 		:param string: is the string, which is expected to be in regular git format
 
 				John Doe <jdoe@example.com>
 				
 		:return: Actor """
-        m = cls.name_email_regex.search(string)
-        if m:
-            name, email = m.groups()
-            return Actor(name, email)
-        else:
-            m = cls.name_only_regex.search(string)
-            if m:
-                return Actor(m.group(1), None)
-            else:
-                # assume best and use the whole string as name
-                return Actor(string, None)
-            # END special case name
-        # END handle name/email matching
+		m = cls.name_email_regex.search(string)
+		if m:
+			name, email = m.groups()
+			return Actor(name, email)
+		else:
+			m = cls.name_only_regex.search(string)
+			if m:
+				return Actor(m.group(1), None)
+			else:
+				# assume best and use the whole string as name
+				return Actor(string, None)
+			# END special case name
+		# END handle name/email matching
 	
 	
 class ProcessStreamAdapter(object):
@@ -359,6 +361,7 @@ class Traversable(object):
 
 class Serializable(object):
 	"""Defines methods to serialize and deserialize objects from and into a data stream"""
+	__slots__ = tuple()
 	
 	def _serialize(self, stream):
 		"""Serialize the data of this object into the given data stream
diff --git a/refs/__init__.py b/refs/__init__.py
index ca5ace02..fc8ce644 100644
--- a/refs/__init__.py
+++ b/refs/__init__.py
@@ -16,5 +16,6 @@ import symbolic
 for item in (HEAD, Head, RemoteReference, TagReference, Reference, SymbolicReference):
 	setattr(symbolic, item.__name__, item)
 del(symbolic)
-# git.objects.Commit -> symbolic
-# git.config.SectionConstraint -> head
+
+
+from log import *
diff --git a/refs/log.py b/refs/log.py
new file mode 100644
index 00000000..f67bea4d
--- /dev/null
+++ b/refs/log.py
@@ -0,0 +1,161 @@
+from head import Head
+from git.util import join_path
+from gitdb.util import (
+						join,
+						file_contents_ro_filepath
+					)
+
+from git.objects.util import (
+								Actor, 
+								parse_actor_and_date,
+								Serializable, 
+								utctz_to_altz,
+								altz_to_utctz_str,
+							)
+
+import os
+
+
+__all__ = ["RefLog", "RefLogEntry"]
+
+
+class RefLogEntry(tuple):
+	"""Named tuple allowing easy access to the revlog data fields"""
+	_fmt = "%s %s %s <%s> %i %s\t%s"
+	__slots__ = tuple()
+	
+	def __repr__(self):
+		"""Representation of ourselves in git reflog format"""
+		act = self.actor
+		time = self.time
+		return self._fmt % (self.oldhexsha, self.newhexsha, act.name, act.email, 
+							time[0], altz_to_utctz_str(time[1]), self.message)
+	
+	@property
+	def oldhexsha(self):
+		"""The hexsha to the commit the ref pointed to before the change""" 
+		return self[0]
+		
+	@property
+	def newhexsha(self):
+		"""The hexsha to the commit the ref now points to, after the change"""
+		return self[1]
+		
+	@property
+	def actor(self):
+		"""Actor instance, providing access"""
+		return self[2]
+		
+	@property
+	def time(self):
+		"""time as tuple:
+		
+		* [0] = int(time)
+		* [1] = int(timezone_offset) in time.altzone format """
+		return self[3]
+		
+	@property
+	def message(self):
+		"""Message describing the operation that acted on the reference"""
+		return self[4]
+	
+	@classmethod
+	def new(self, oldhexsha, newhexsha, actor, time, tz_offset, message):
+		""":return: New instance of a RefLogEntry"""
+		if not isinstance(actor, Actor):
+			raise ValueError("Need actor instance, got %s" % actor)
+		# END check types 
+		return RefLogEntry((oldhexsha, newhexsha, actor, (time, tz_offset), message))
+		
+	@classmethod
+	def from_line(self, line):
+		""":return: New RefLogEntry instance from the given revlog line.
+		:param line: line without trailing newline
+		:raise ValueError: If line could not be parsed"""
+		raise NotImplementedError("todo")
+
+	
+class RefLog(list, Serializable):
+	"""A reflog contains reflog entries, each of which defines a certain state
+	of the head in question. Custom query methods allow to retrieve log entries 
+	by date or by other criteria.
+	
+	Reflog entries are orded, the first added entry is first in the list, the last
+	entry, i.e. the last change of the head or reference, is last in the list."""
+	
+	__slots__ = tuple()
+	
+	#{ Interface
+	
+	@classmethod
+	def from_file(cls, filepath):
+		"""
+		:return: a new RefLog instance containing all entries from the reflog 
+			at the given filepath
+		:param filepath: path to reflog 
+		:raise ValueError: If the file could not be read or was corrupted in some way"""
+		inst = cls()
+		fmap = file_contents_ro_filepath(filepath, stream=False, allow_mmap=True)
+		try:
+			inst._deserialize(fmap)
+		finally:
+			fmap.close()
+		#END handle closing of handle 
+		return inst
+	
+	@classmethod
+	def reflog_path(cls, ref):
+		"""
+		:return: string to absolute path at which the reflog of the given ref 
+			instance would be found. The path is not guaranteed to point to a valid 
+			file though.
+		:param ref: SymbolicReference instance"""
+		return join(ref.repo.git_dir, "logs", ref.path)
+		
+	@classmethod
+	def iter_entries(cls, stream):
+		"""
+		:return: Iterator yielding RefLogEntry instances, one for each line read 
+			sfrom the given stream.
+		:param stream: file-like object containing the revlog in its native format
+			or basestring instance pointing to a file to read"""
+		new_entry = RefLogEntry.from_line
+		if isinstance(stream, basestring):
+			stream = file_contents_ro_filepath(stream)
+		#END handle stream type
+		return (new_entry(line.strip()) for line in stream)
+	
+	def to_file(self, filepath):
+		"""Write the contents of the reflog instance to a file at the given filepath.
+		:param filepath: path to file, parent directories are assumed to exist"""
+		fp = open(filepath, 'wb')
+		try:
+			self._serialize(fp)
+		finally:
+			fp.close()
+		#END handle file streams
+	
+	#} END interface
+	
+	#{ Serializable Interface
+	def _serialize(self, stream):
+		lm1 = len(self) - 1
+		write = stream.write()
+		
+		# write all entries
+		for i, e in self:
+			s = repr(e)
+			if i != lm1:
+				s += "\n"
+			#END handle line separator
+			write(s)
+		#END for each entry
+	
+	def _deserialize(self, stream):
+		new_entry = RefLogEntry.from_line
+		append = self.append
+		# NOTE: should use iter_entries, but this way it will be more direct and faster
+		for line in stream:
+			append(new_entry(line.strip()))
+		#END handle deserializatoin
+	#} END serializable interface
diff --git a/test/test_blob.py b/test/test_blob.py
index 8513328e..661c0501 100644
--- a/test/test_blob.py
+++ b/test/test_blob.py
@@ -9,12 +9,15 @@ from git import *
 from gitdb.util import hex_to_bin
 
 class TestBlob(TestBase):
-    
-    def test_mime_type_should_return_mime_type_for_known_types(self):
-        blob = Blob(self.rorepo, **{'binsha': Blob.NULL_BIN_SHA, 'path': 'foo.png'})
-        assert_equal("image/png", blob.mime_type)
+	
+	def test_mime_type_should_return_mime_type_for_known_types(self):
+		blob = Blob(self.rorepo, **{'binsha': Blob.NULL_BIN_SHA, 'path': 'foo.png'})
+		assert_equal("image/png", blob.mime_type)
   
-    def test_mime_type_should_return_text_plain_for_unknown_types(self):
-        blob = Blob(self.rorepo, **{'binsha': Blob.NULL_BIN_SHA,'path': 'something'})
-        assert_equal("text/plain", blob.mime_type)
+	def test_mime_type_should_return_text_plain_for_unknown_types(self):
+		blob = Blob(self.rorepo, **{'binsha': Blob.NULL_BIN_SHA,'path': 'something'})
+		assert_equal("text/plain", blob.mime_type)
   
+	def test_nodict(self):
+		self.failUnlessRaises(AttributeError, setattr, self.rorepo.tree()['AUTHORS'], 'someattr', 2)
+	  
diff --git a/test/test_commit.py b/test/test_commit.py
index a00a8de8..4a8d8b87 100644
--- a/test/test_commit.py
+++ b/test/test_commit.py
@@ -70,6 +70,8 @@ class TestCommit(TestBase):
 	def test_bake(self):
 
 		commit = self.rorepo.commit('2454ae89983a4496a445ce347d7a41c0bb0ea7ae')
+		# commits have no dict
+		self.failUnlessRaises(AttributeError, setattr, commit, 'someattr', 1)
 		commit.author # bake
 
 		assert_equal("Sebastian Thiel", commit.author.name)
diff --git a/test/test_reflog.py b/test/test_reflog.py
new file mode 100644
index 00000000..efcc7f33
--- /dev/null
+++ b/test/test_reflog.py
@@ -0,0 +1,32 @@
+from git.test.lib import *
+from git.objects import IndexObject, Actor
+from git.refs import *
+
+class TestRefLog(TestBase):
+
+	def test_reflogentry(self):
+		nullhexsha = IndexObject.NULL_HEX_SHA
+		hexsha = 'F' * 40
+		actor = Actor('name', 'email')
+		msg = "message"
+		
+		self.failUnlessRaises(ValueError, RefLogEntry.new, nullhexsha, hexsha, 'noactor', 0, 0, "")
+		e = RefLogEntry.new(nullhexsha, hexsha, actor, 0, 1, msg)
+		
+		assert e.oldhexsha == nullhexsha
+		assert e.newhexsha == hexsha
+		assert e.actor == actor
+		assert e.time[0] == 0
+		assert e.time[1] == 1
+		assert e.message == msg
+		
+		# check representation (roughly)
+		assert repr(e).startswith(nullhexsha)
+	
+	def test_base(self):
+		pass
+		# raise on invalid revlog
+		# TODO: Try multiple corrupted ones !
+	
+	
+		# test serialize and deserialize - results must match exactly
diff --git a/test/test_refs.py b/test/test_refs.py
index 700e5fac..ebf1a00d 100644
--- a/test/test_refs.py
+++ b/test/test_refs.py
@@ -33,6 +33,8 @@ class TestRefs(TestBase):
 			if tag.tag is not None:
 				tag_object_refs.append( tag )
 				tagobj = tag.tag
+				# have no dict
+				self.failUnlessRaises(AttributeError, setattr, tagobj, 'someattr', 1)
 				assert isinstance( tagobj, TagObject ) 
 				assert tagobj.tag == tag.name
 				assert isinstance( tagobj.tagger, Actor )
diff --git a/test/test_tree.py b/test/test_tree.py
index 80db2e4b..ec10e962 100644
--- a/test/test_tree.py
+++ b/test/test_tree.py
@@ -23,6 +23,9 @@ class TestTree(TestBase):
 				continue
 			# END skip non-trees
 			tree = item
+			# trees have no dict
+			self.failUnlessRaises(AttributeError, setattr, tree, 'someattr', 1)
+			
 			orig_data = tree.data_stream.read()
 			orig_cache = tree._cache