renamed find_all to list_all, changed commit to use iterable interface in preparation for command changes

Author: Byron

CHANGES

@@ -2,9 +2,8 @@
 CHANGES
 =======
 
-0.1.X
+0.2
 =====
-( Future Release )
 General
 -------
 * file mode in Tree, Blob and Diff objects now is an int compatible to definintiions 
@@ -19,7 +18,16 @@ General
 * from_string and list_from_string methods are now private and were renamed to 
   _from_string  and _list_from_string respectively. As part of the private API, they 
   may change without prior notice.
-
+* Renamed all find_all methods to list_items - this method is part of the Iterable interface
+  that also provides a more efficients and more responsive iter_items method
+
+Item Iteration
+--------------
+* Previously one would return and process multiple items as list only which can 
+  hurt performance and memory consumption and reduce response times. 
+  iter_items method provide an iterator that will return items on demand as parsed 
+  from a stream. This way any amount of objects can be handled.
+  
 objects Package
 ----------------
 * blob, tree, tag and commit module have been moved to new objects package. This should 

lib/git/objects/commit.py

@@ -6,14 +6,14 @@
 
 import re
 import time
-
+from git.utils import Iterable
 from git.actor import Actor
-from tree import Tree
 import git.diff as diff
 import git.stats as stats
+from tree import Tree
 import base
 
-class Commit(base.Object):
+class Commit(base.Object, Iterable):
 	"""
 	Wraps a git Commit object.
 	
@@ -81,7 +81,7 @@ def _set_cache_(self, attr):
 		We set all values at once.
 		"""
 		if attr in self.__slots__:
-			temp = Commit.find_all(self.repo, self.id, max_count=1)[0]
+			temp = Commit.list_items(self.repo, self.id, max_count=1)[0]
 			self.parents = temp.parents
 			self.tree = temp.tree
 			self.author = temp.author
@@ -120,15 +120,15 @@ def count(cls, repo, ref, path=''):
 		return len(repo.git.rev_list(ref, '--', path).strip().splitlines())
 
 	@classmethod
-	def find_all(cls, repo, ref, path='', **kwargs):
+	def iter_items(cls, repo, ref, path='', **kwargs):
 		"""
 		Find all commits matching the given criteria.
 
 		``repo``
 			is the Repo
 
 		``ref``
-			is the ref from which to begin (SHA1 or name)
+			is the ref from which to begin (SHA1, Head or name)
 
 		``path``
 			is an optinal path, if set only Commits that include the path 
@@ -146,49 +146,56 @@ def find_all(cls, repo, ref, path='', **kwargs):
 		options.update(kwargs)
 
 		output = repo.git.rev_list(ref, '--', path, **options)
-		return cls._list_from_string(repo, output)
+		return cls._iter_from_stream(repo, iter(output.splitlines(False)))
 
 	@classmethod
-	def _list_from_string(cls, repo, text):
+	def _iter_from_stream(cls, repo, stream):
 		"""
 		Parse out commit information into a list of Commit objects
 
 		``repo``
 			is the Repo
 
-		``text``
-			is the text output from the git-rev-list command (raw format)
+		``stream``
+			output stream from the git-rev-list command (raw format)
 
 		Returns
-			git.Commit[]
+			iterator returning Commit objects
 		"""
-		lines =text.splitlines(False)
-		commits = []
-
-		while lines:
-			id = lines.pop(0).split()[1]
-			tree = lines.pop(0).split()[1]
+		for line in stream:
+			id = line.split()[1]
+			assert line.split()[0] == "commit"
+			tree = stream.next().split()[1]
 
 			parents = []
-			while lines and lines[0].startswith('parent'):
-				parents.append(lines.pop(0).split()[-1])
-			# END while there are parent lines
-			author, authored_date = cls._actor(lines.pop(0))
-			committer, committed_date = cls._actor(lines.pop(0))
+			next_line = None
+			for parent_line in stream:
+				if not parent_line.startswith('parent'):
+					next_line = parent_line
+					break
+				# END abort reading parents
+				parents.append(parent_line.split()[-1])
+			# END for each parent line
+			
+			author, authored_date = cls._actor(next_line)
+			committer, committed_date = cls._actor(stream.next())
 
-			# free line
-			lines.pop(0)
+			# empty line
+			stream.next()
 
 			message_lines = []
-			while lines and not lines[0].startswith('commit'):
-				message_lines.append(lines.pop(0).strip())
+			next_line = None
+			for msg_line in stream:
+				if not msg_line.startswith('    '):
+					break
+				# END abort message reading 
+				message_lines.append(msg_line.strip())
 			# END while there are message lines
-			message = '\n'.join(message_lines[:-1])	# last line is empty
-
-			commits.append(Commit(repo, id=id, parents=parents, tree=tree, author=author, authored_date=authored_date,
-								  committer=committer, committed_date=committed_date, message=message))
-		# END while lines
-		return commits
+			message = '\n'.join(message_lines)
+			
+			yield Commit(repo, id=id, parents=parents, tree=tree, author=author, authored_date=authored_date,
+						  committer=committer, committed_date=committed_date, message=message)
+		# END for each line in stream
 
 	@classmethod
 	def diff(cls, repo, a, b=None, paths=None):

lib/git/refs.py

@@ -73,7 +73,7 @@ def name(self):
 		return '/'.join(tokens[2:])
 
 	@classmethod
-	def find_all(cls, repo, common_path = "refs", **kwargs):
+	def list_items(cls, repo, common_path = "refs", **kwargs):
 		"""
 		Find all refs in the repository
 
@@ -158,14 +158,14 @@ def commit(self):
 		return self.object
 
 	@classmethod
-	def find_all(cls, repo, common_path = "refs/heads", **kwargs):
+	def list_items(cls, repo, common_path = "refs/heads", **kwargs):
 		"""
 		Returns
 			git.Head[]
 			
-		For more documentation, please refer to git.base.Ref.find_all
+		For more documentation, please refer to git.base.Ref.list_items
 		"""
-		return super(Head,cls).find_all(repo, common_path, **kwargs)
+		return super(Head,cls).list_items(repo, common_path, **kwargs)
 
 	def __repr__(self):
 		return '<git.Head "%s">' % self.name
@@ -181,7 +181,7 @@ class TagRef(Ref):
 	This tag object will always point to a commit object, but may carray additional
 	information in a tag object::
 	
-	 tagref = TagRef.find_all(repo)[0]
+	 tagref = TagRef.list_items(repo)[0]
 	 print tagref.commit.message
 	 if tagref.tag is not None:
 		print tagref.tag.message
@@ -215,14 +215,14 @@ def tag(self):
 		return None
 
 	@classmethod
-	def find_all(cls, repo, common_path = "refs/tags", **kwargs):
+	def list_items(cls, repo, common_path = "refs/tags", **kwargs):
 		"""
 		Returns
 			git.Tag[]
 			
-		For more documentation, please refer to git.base.Ref.find_all
+		For more documentation, please refer to git.base.Ref.list_items
 		"""
-		return super(TagRef,cls).find_all(repo, common_path, **kwargs)
+		return super(TagRef,cls).list_items(repo, common_path, **kwargs)
 
 
 # provide an alias

lib/git/repo.py

@@ -102,7 +102,7 @@ def heads(self):
 		Returns
 			``git.Head[]``
 		"""
-		return Head.find_all(self)
+		return Head.list_items(self)
 
 	# alias heads
 	branches = heads
@@ -115,7 +115,7 @@ def tags(self):
 		Returns
 			``git.Tag[]``
 		"""
-		return Tag.find_all(self)
+		return Tag.list_items(self)
 
 	def blame(self, commit, file):
 		"""
@@ -221,7 +221,7 @@ def commits(self, start='master', path='', max_count=10, skip=0):
 		options = {'max_count': max_count,
 				   'skip': skip}
 
-		return Commit.find_all(self, start, path, **options)
+		return Commit.list_items(self, start, path, **options)
 
 	def commits_between(self, frm, to):
 		"""
@@ -237,7 +237,7 @@ def commits_between(self, frm, to):
 		Returns
 			``git.Commit[]``
 		"""
-		return reversed(Commit.find_all(self, "%s..%s" % (frm, to)))
+		return reversed(Commit.list_items(self, "%s..%s" % (frm, to)))
 
 	def commits_since(self, start='master', path='', since='1970-01-01'):
 		"""
@@ -259,7 +259,7 @@ def commits_since(self, start='master', path='', since='1970-01-01'):
 		"""
 		options = {'since': since}
 
-		return Commit.find_all(self, start, path, **options)
+		return Commit.list_items(self, start, path, **options)
 
 	def commit_count(self, start='master', path=''):
 		"""
@@ -296,7 +296,7 @@ def commit(self, id=None, path = ''):
 			id = self.active_branch
 		options = {'max_count': 1}
 
-		commits = Commit.find_all(self, id, path, **options)
+		commits = Commit.list_items(self, id, path, **options)
 
 		if not commits:
 			raise ValueError, "Invalid identifier %s, or given path '%s' too restrictive" % ( id, path )
@@ -313,7 +313,7 @@ def commit_deltas_from(self, other_repo, ref='master', other_ref='master'):
 		other_repo_refs = other_repo.git.rev_list(other_ref, '--').strip().splitlines()
 
 		diff_refs = list(set(other_repo_refs) - set(repo_refs))
-		return map(lambda ref: Commit.find_all(other_repo, ref, max_count=1)[0], diff_refs)
+		return map(lambda ref: Commit.list_items(other_repo, ref, max_count=1)[0], diff_refs)
 
 	def tree(self, treeish=None):
 		"""
@@ -364,7 +364,8 @@ def log(self, commit='master', path=None, **kwargs):
 		if path:
 			arg.append(path)
 		commits = self.git.log(*arg, **options)
-		return Commit._list_from_string(self, commits)
+		print commits.splitlines(False)
+		return list(Commit._iter_from_stream(self, iter(commits.splitlines())))
 
 	def diff(self, a, b, *paths):
 		"""

lib/git/utils.py

@@ -75,13 +75,13 @@ def list_items(cls, repo, *args, **kwargs):
 		Returns:
 			list(Item,...) list of item instances 
 		"""
-		return list(cls.iter_items, repo, *args, **kwargs)
+		return list(cls.iter_items(repo, *args, **kwargs))
 
 
 	@classmethod
 	def iter_items(cls, repo, *args, **kwargs):
 		"""
-		For more information about the arguments, see find_all
+		For more information about the arguments, see list_items
 		Return: 
 			iterator yielding Items
 		"""

test/git/test_commit.py

@@ -216,7 +216,7 @@ def test_rev_list_bisect_all(self, git):
 									  bisect_all=True)
 		assert_true(git.called)
 
-		commits = Commit._list_from_string(self.repo, revs)
+		commits = Commit._iter_from_stream(self.repo, iter(revs.splitlines(False)))
 		expected_ids = (
 			'cf37099ea8d1d8c7fbf9b6d12d7ec0249d3acb8b',
 			'33ebe7acec14b25c5f84f35a664803fcab2f7781',