diff --git a/.travis.yml b/.travis.yml
index 570beaad6..1bed8368c 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,6 +1,10 @@
 # UNUSED, only for reference. If adjustments are needed, please see github actions
 language: python
 python:
+<<<<<<< HEAD
+  - "3.4"
+=======
+>>>>>>> b0f79c58ad919e90261d1e332df79a4ad0bc40de
   - "3.6"
   - "3.7"
   - "3.8"
diff --git a/git/cmd.py b/git/cmd.py
index 4f58b3146..e078e4a18 100644
--- a/git/cmd.py
+++ b/git/cmd.py
@@ -148,7 +148,6 @@ def dashify(string: str) -> str:
 
 
 def slots_to_dict(self, exclude: Sequence[str] = ()) -> Dict[str, Any]:
-    # annotate self.__slots__ as Tuple[str, ...] once 3.5 dropped
     return {s: getattr(self, s) for s in self.__slots__ if s not in exclude}
 
 
@@ -460,7 +459,7 @@ class CatFileContentStream(object):
         If not all data is read to the end of the objects's lifetime, we read the
         rest to assure the underlying stream continues to work"""
 
-        __slots__ = ('_stream', '_nbr', '_size')
+        __slots__: Tuple[str, ...] = ('_stream', '_nbr', '_size')
 
         def __init__(self, size: int, stream: IO[bytes]) -> None:
             self._stream = stream
diff --git a/git/diff.py b/git/diff.py
index a40fc244e..346a2ca7b 100644
--- a/git/diff.py
+++ b/git/diff.py
@@ -16,7 +16,7 @@
 # typing ------------------------------------------------------------------
 
 from typing import Any, Iterator, List, Match, Optional, Tuple, Type, Union, TYPE_CHECKING
-from git.types import PathLike, TBD, Final, Literal
+from git.types import PathLike, TBD, Literal
 
 if TYPE_CHECKING:
     from .objects.tree import Tree
@@ -31,7 +31,7 @@
 __all__ = ('Diffable', 'DiffIndex', 'Diff', 'NULL_TREE')
 
 # Special object to compare against the empty tree in diffs
-NULL_TREE = object()  # type: Final[object]
+NULL_TREE = object()
 
 _octal_byte_re = re.compile(b'\\\\([0-9]{3})')
 
diff --git a/git/index/fun.py b/git/index/fun.py
index 1012f4801..3fded3473 100644
--- a/git/index/fun.py
+++ b/git/index/fun.py
@@ -109,7 +109,7 @@ def run_commit_hook(name: str, index: 'IndexFile', *args: str) -> None:
     # end handle return code
 
 
-def stat_mode_to_index_mode(mode):
+def stat_mode_to_index_mode(mode: int) -> int:
     """Convert the given mode from a stat call to the corresponding index mode
     and return it"""
     if S_ISLNK(mode):   # symlinks
diff --git a/git/objects/base.py b/git/objects/base.py
index 59f0e8368..884f96515 100644
--- a/git/objects/base.py
+++ b/git/objects/base.py
@@ -3,16 +3,34 @@
 #
 # This module is part of GitPython and is released under
 # the BSD License: http://www.opensource.org/licenses/bsd-license.php
+
+from git.exc import WorkTreeRepositoryUnsupported
 from git.util import LazyMixin, join_path_native, stream_copy, bin_to_hex
 
 import gitdb.typ as dbtyp
 import os.path as osp
-from typing import Optional  # noqa: F401 unused import
 
 from .util import get_object_type_by_name
 
 
-_assertion_msg_format = "Created object %r whose python type %r disagrees with the acutal git object type %r"
+# typing ------------------------------------------------------------------
+
+from typing import Any, TYPE_CHECKING, Optional, Union
+
+from git.types import PathLike
+
+if TYPE_CHECKING:
+    from git.repo import Repo
+    from gitdb.base import OStream
+    from .tree import Tree
+    from .blob import Blob
+    from .tag import TagObject
+    from .commit import Commit
+
+# --------------------------------------------------------------------------
+
+
+_assertion_msg_format = "Created object %r whose python type %r disagrees with the acutual git object type %r"
 
 __all__ = ("Object", "IndexObject")
 
@@ -27,7 +45,7 @@ class Object(LazyMixin):
     __slots__ = ("repo", "binsha", "size")
     type = None  # type: Optional[str] # to be set by subclass
 
-    def __init__(self, repo, binsha):
+    def __init__(self, repo: 'Repo', binsha: bytes):
         """Initialize an object by identifying it by its binary sha.
         All keyword arguments will be set on demand if None.
 
@@ -40,7 +58,7 @@ def __init__(self, repo, binsha):
         assert len(binsha) == 20, "Require 20 byte binary sha, got %r, len = %i" % (binsha, len(binsha))
 
     @classmethod
-    def new(cls, repo, id):  # @ReservedAssignment
+    def new(cls, repo: 'Repo', id):  # @ReservedAssignment
         """
         :return: New Object instance of a type appropriate to the object type behind
             id. The id of the newly created object will be a binsha even though
@@ -53,7 +71,7 @@ def new(cls, repo, id):  # @ReservedAssignment
         return repo.rev_parse(str(id))
 
     @classmethod
-    def new_from_sha(cls, repo, sha1):
+    def new_from_sha(cls, repo: 'Repo', sha1: bytes) -> Union['Commit', 'TagObject', 'Tree', 'Blob']:
         """
         :return: new object instance of a type appropriate to represent the given
             binary sha1
@@ -67,52 +85,52 @@ def new_from_sha(cls, repo, sha1):
         inst.size = oinfo.size
         return inst
 
-    def _set_cache_(self, attr):
+    def _set_cache_(self, attr: str) -> None:
         """Retrieve object information"""
         if attr == "size":
             oinfo = self.repo.odb.info(self.binsha)
-            self.size = oinfo.size
+            self.size = oinfo.size  # type:  int
             # assert oinfo.type == self.type, _assertion_msg_format % (self.binsha, oinfo.type, self.type)
         else:
             super(Object, self)._set_cache_(attr)
 
-    def __eq__(self, other):
+    def __eq__(self, other: Any) -> bool:
         """:return: True if the objects have the same SHA1"""
         if not hasattr(other, 'binsha'):
             return False
         return self.binsha == other.binsha
 
-    def __ne__(self, other):
+    def __ne__(self, other: Any) -> bool:
         """:return: True if the objects do not have the same SHA1 """
         if not hasattr(other, 'binsha'):
             return True
         return self.binsha != other.binsha
 
-    def __hash__(self):
+    def __hash__(self) -> int:
         """:return: Hash of our id allowing objects to be used in dicts and sets"""
         return hash(self.binsha)
 
-    def __str__(self):
+    def __str__(self) -> str:
         """:return: string of our SHA1 as understood by all git commands"""
         return self.hexsha
 
-    def __repr__(self):
+    def __repr__(self) -> str:
         """:return: string with pythonic representation of our object"""
         return '<git.%s "%s">' % (self.__class__.__name__, self.hexsha)
 
     @property
-    def hexsha(self):
+    def hexsha(self) -> str:
         """:return: 40 byte hex version of our 20 byte binary sha"""
         # b2a_hex produces bytes
         return bin_to_hex(self.binsha).decode('ascii')
 
     @property
-    def data_stream(self):
+    def data_stream(self) -> 'OStream':
         """ :return:  File Object compatible stream to the uncompressed raw data of the object
         :note: returned streams must be read in order"""
         return self.repo.odb.stream(self.binsha)
 
-    def stream_data(self, ostream):
+    def stream_data(self, ostream: 'OStream') -> 'Object':
         """Writes our data directly to the given output stream
         :param ostream: File object compatible stream object.
         :return: self"""
@@ -130,7 +148,9 @@ class IndexObject(Object):
     # for compatibility with iterable lists
     _id_attribute_ = 'path'
 
-    def __init__(self, repo, binsha, mode=None, path=None):
+    def __init__(self,
+                 repo: 'Repo', binsha: bytes, mode: Union[None, int] = None, path: Union[None, PathLike] = None
+                 ) -> None:
         """Initialize a newly instanced IndexObject
 
         :param repo: is the Repo we are located in
@@ -150,14 +170,14 @@ def __init__(self, repo, binsha, mode=None, path=None):
         if path is not None:
             self.path = path
 
-    def __hash__(self):
+    def __hash__(self) -> int:
         """
         :return:
             Hash of our path as index items are uniquely identifiable by path, not
             by their data !"""
         return hash(self.path)
 
-    def _set_cache_(self, attr):
+    def _set_cache_(self, attr: str) -> None:
         if attr in IndexObject.__slots__:
             # they cannot be retrieved lateron ( not without searching for them )
             raise AttributeError(
@@ -168,16 +188,19 @@ def _set_cache_(self, attr):
         # END handle slot attribute
 
     @property
-    def name(self):
+    def name(self) -> str:
         """:return: Name portion of the path, effectively being the basename"""
         return osp.basename(self.path)
 
     @property
-    def abspath(self):
+    def abspath(self) -> PathLike:
         """
         :return:
             Absolute path to this index object in the file system ( as opposed to the
             .path field which is a path relative to the git repository ).
 
             The returned path will be native to the system and contains '\' on windows. """
-        return join_path_native(self.repo.working_tree_dir, self.path)
+        if self.repo.working_tree_dir is not None:
+            return join_path_native(self.repo.working_tree_dir, self.path)
+        else:
+            raise WorkTreeRepositoryUnsupported("Working_tree_dir was None or empty")
diff --git a/git/objects/blob.py b/git/objects/blob.py
index 897f892bf..017178f05 100644
--- a/git/objects/blob.py
+++ b/git/objects/blob.py
@@ -23,11 +23,11 @@ class Blob(base.IndexObject):
     __slots__ = ()
 
     @property
-    def mime_type(self):
+    def mime_type(self) -> str:
         """
         :return: String describing the mime type of this file (based on the filename)
         :note: Defaults to 'text/plain' in case the actual file type is unknown. """
         guesses = None
         if self.path:
-            guesses = guess_type(self.path)
+            guesses = guess_type(str(self.path))
         return guesses and guesses[0] or self.DEFAULT_MIME_TYPE
diff --git a/git/objects/commit.py b/git/objects/commit.py
index 45e6d772c..26db6e36d 100644
--- a/git/objects/commit.py
+++ b/git/objects/commit.py
@@ -36,6 +36,11 @@
 from io import BytesIO
 import logging
 
+from typing import List, Tuple, Union, TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from git.repo import Repo
+
 log = logging.getLogger('git.objects.commit')
 log.addHandler(logging.NullHandler())
 
@@ -70,7 +75,8 @@ class Commit(base.Object, Iterable, Diffable, Traversable, Serializable):
 
     def __init__(self, repo, binsha, tree=None, author=None, authored_date=None, author_tz_offset=None,
                  committer=None, committed_date=None, committer_tz_offset=None,
-                 message=None, parents=None, encoding=None, gpgsig=None):
+                 message=None, parents: Union[Tuple['Commit', ...], List['Commit'], None] = None,
+                 encoding=None, gpgsig=None):
         """Instantiate a new Commit. All keyword arguments taking None as default will
         be implicitly set on first query.
 
@@ -133,11 +139,11 @@ def __init__(self, repo, binsha, tree=None, author=None, authored_date=None, aut
             self.gpgsig = gpgsig
 
     @classmethod
-    def _get_intermediate_items(cls, commit):
-        return commit.parents
+    def _get_intermediate_items(cls, commit: 'Commit') -> Tuple['Commit', ...]:  # type: ignore  ## cos overriding super
+        return tuple(commit.parents)
 
     @classmethod
-    def _calculate_sha_(cls, repo, commit):
+    def _calculate_sha_(cls, repo: 'Repo', commit: 'Commit') -> bytes:
         '''Calculate the sha of a commit.
 
         :param repo: Repo object the commit should be part of
@@ -430,7 +436,7 @@ def create_from_tree(cls, repo, tree, message, parent_commits=None, head=False,
 
     #{ Serializable Implementation
 
-    def _serialize(self, stream):
+    def _serialize(self, stream: BytesIO) -> 'Commit':
         write = stream.write
         write(("tree %s\n" % self.tree).encode('ascii'))
         for p in self.parents:
@@ -471,7 +477,7 @@ def _serialize(self, stream):
         # END handle encoding
         return self
 
-    def _deserialize(self, stream):
+    def _deserialize(self, stream: BytesIO) -> 'Commit':
         """:param from_rev_list: if true, the stream format is coming from the rev-list command
         Otherwise it is assumed to be a plain data stream from our object"""
         readline = stream.readline
@@ -511,7 +517,7 @@ def _deserialize(self, stream):
         buf = enc.strip()
         while buf:
             if buf[0:10] == b"encoding ":
-                self.encoding = buf[buf.find(' ') + 1:].decode(
+                self.encoding = buf[buf.find(b' ') + 1:].decode(
                     self.encoding, 'ignore')
             elif buf[0:7] == b"gpgsig ":
                 sig = buf[buf.find(b' ') + 1:] + b"\n"
diff --git a/git/objects/submodule/base.py b/git/objects/submodule/base.py
index e3be1a728..b03fa22a5 100644
--- a/git/objects/submodule/base.py
+++ b/git/objects/submodule/base.py
@@ -3,6 +3,7 @@
 import logging
 import os
 import stat
+from typing import List
 from unittest import SkipTest
 import uuid
 
@@ -134,10 +135,11 @@ def _set_cache_(self, attr):
             super(Submodule, self)._set_cache_(attr)
         # END handle attribute name
 
-    def _get_intermediate_items(self, item):
+    @classmethod
+    def _get_intermediate_items(cls, item: 'Submodule') -> List['Submodule']:  # type: ignore
         """:return: all the submodules of our module repository"""
         try:
-            return type(self).list_items(item.module())
+            return cls.list_items(item.module())
         except InvalidGitRepositoryError:
             return []
         # END handle intermediate items
diff --git a/git/objects/tag.py b/git/objects/tag.py
index b9bc6c248..cb6efbe9b 100644
--- a/git/objects/tag.py
+++ b/git/objects/tag.py
@@ -9,6 +9,15 @@
 from ..util import hex_to_bin
 from ..compat import defenc
 
+from typing import List, TYPE_CHECKING, Union
+
+if TYPE_CHECKING:
+    from git.repo import Repo
+    from git.util import Actor
+    from .commit import Commit
+    from .blob import Blob
+    from .tree import Tree
+
 __all__ = ("TagObject", )
 
 
@@ -18,8 +27,14 @@ class TagObject(base.Object):
     type = "tag"
     __slots__ = ("object", "tag", "tagger", "tagged_date", "tagger_tz_offset", "message")
 
-    def __init__(self, repo, binsha, object=None, tag=None,  # @ReservedAssignment
-                 tagger=None, tagged_date=None, tagger_tz_offset=None, message=None):
+    def __init__(self, repo: 'Repo', binsha: bytes,
+                 object: Union[None, base.Object] = None,
+                 tag: Union[None, str] = None,
+                 tagger: Union[None, 'Actor'] = None,
+                 tagged_date: Union[int, None] = None,
+                 tagger_tz_offset: Union[int, None] = None,
+                 message: Union[str, None] = None
+                 ) -> None:   # @ReservedAssignment
         """Initialize a tag object with additional data
 
         :param repo: repository this object is located in
@@ -34,7 +49,7 @@ def __init__(self, repo, binsha, object=None, tag=None,  # @ReservedAssignment
             authored_date is in, in a format similar to time.altzone"""
         super(TagObject, self).__init__(repo, binsha)
         if object is not None:
-            self.object = object
+            self.object = object    # type: Union['Commit', 'Blob', 'Tree', 'TagObject']
         if tag is not None:
             self.tag = tag
         if tagger is not None:
@@ -46,16 +61,17 @@ def __init__(self, repo, binsha, object=None, tag=None,  # @ReservedAssignment
         if message is not None:
             self.message = message
 
-    def _set_cache_(self, attr):
+    def _set_cache_(self, attr: str) -> None:
         """Cache all our attributes at once"""
         if attr in TagObject.__slots__:
             ostream = self.repo.odb.stream(self.binsha)
-            lines = ostream.read().decode(defenc, 'replace').splitlines()
+            lines = ostream.read().decode(defenc, 'replace').splitlines()  # type: List[str]
 
             _obj, hexsha = lines[0].split(" ")
             _type_token, type_name = lines[1].split(" ")
+            object_type = get_object_type_by_name(type_name.encode('ascii'))
             self.object = \
-                get_object_type_by_name(type_name.encode('ascii'))(self.repo, hex_to_bin(hexsha))
+                object_type(self.repo, hex_to_bin(hexsha))
 
             self.tag = lines[2][4:]  # tag <tag name>
 
diff --git a/git/objects/tree.py b/git/objects/tree.py
index 68e98329b..29b2a6846 100644
--- a/git/objects/tree.py
+++ b/git/objects/tree.py
@@ -17,6 +17,17 @@
     tree_to_stream
 )
 
+
+# typing -------------------------------------------------
+
+from typing import Iterable, Iterator, Tuple, Union, cast, TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from io import BytesIO
+
+#--------------------------------------------------------
+
+
 cmp = lambda a, b: (a > b) - (a < b)
 
 __all__ = ("TreeModifier", "Tree")
@@ -182,8 +193,10 @@ def __init__(self, repo, binsha, mode=tree_id << 12, path=None):
         super(Tree, self).__init__(repo, binsha, mode, path)
 
     @classmethod
-    def _get_intermediate_items(cls, index_object):
+    def _get_intermediate_items(cls, index_object: 'Tree',    # type: ignore
+                                ) -> Tuple['Tree', ...]:
         if index_object.type == "tree":
+            index_object = cast('Tree', index_object)
             return tuple(index_object._iter_convert_to_object(index_object._cache))
         return ()
 
@@ -196,7 +209,8 @@ def _set_cache_(self, attr):
             super(Tree, self)._set_cache_(attr)
         # END handle attribute
 
-    def _iter_convert_to_object(self, iterable):
+    def _iter_convert_to_object(self, iterable: Iterable[Tuple[bytes, int, str]]
+                                ) -> Iterator[Union[Blob, 'Tree', Submodule]]:
         """Iterable yields tuples of (binsha, mode, name), which will be converted
         to the respective object representation"""
         for binsha, mode, name in iterable:
@@ -317,7 +331,7 @@ def __contains__(self, item):
     def __reversed__(self):
         return reversed(self._iter_convert_to_object(self._cache))
 
-    def _serialize(self, stream):
+    def _serialize(self, stream: 'BytesIO') -> 'Tree':
         """Serialize this tree into the stream. Please note that we will assume
         our tree data to be in a sorted state. If this is not the case, serialization
         will not generate a correct tree representation as these are assumed to be sorted
@@ -325,7 +339,7 @@ def _serialize(self, stream):
         tree_to_stream(self._cache, stream.write)
         return self
 
-    def _deserialize(self, stream):
+    def _deserialize(self, stream: 'BytesIO') -> 'Tree':
         self._cache = tree_entries_from_data(stream.read())
         return self
 
diff --git a/git/objects/util.py b/git/objects/util.py
index d15d83c35..087f0166b 100644
--- a/git/objects/util.py
+++ b/git/objects/util.py
@@ -4,19 +4,34 @@
 # This module is part of GitPython and is released under
 # the BSD License: http://www.opensource.org/licenses/bsd-license.php
 """Module for general utility functions"""
+
 from git.util import (
     IterableList,
     Actor
 )
 
 import re
-from collections import deque as Deque
+from collections import deque
 
 from string import digits
 import time
 import calendar
 from datetime import datetime, timedelta, tzinfo
 
+# typing ------------------------------------------------------------
+from typing import (Any, Callable, Deque, Iterator, Sequence, TYPE_CHECKING, Tuple, Type, Union, cast, overload)
+
+if TYPE_CHECKING:
+    from io import BytesIO, StringIO
+    from .submodule.base import Submodule
+    from .commit import Commit
+    from .blob import Blob
+    from .tag import TagObject
+    from .tree import Tree
+    from subprocess import Popen
+
+# --------------------------------------------------------------------
+
 __all__ = ('get_object_type_by_name', 'parse_date', 'parse_actor_and_date',
            'ProcessStreamAdapter', 'Traversable', 'altz_to_utctz_str', 'utctz_to_altz',
            'verify_utctz', 'Actor', 'tzoffset', 'utc')
@@ -26,7 +41,7 @@
 #{ Functions
 
 
-def mode_str_to_int(modestr):
+def mode_str_to_int(modestr: Union[bytes, str]) -> int:
     """
     :param modestr: string like 755 or 644 or 100644 - only the last 6 chars will be used
     :return:
@@ -36,12 +51,14 @@ def mode_str_to_int(modestr):
         for example."""
     mode = 0
     for iteration, char in enumerate(reversed(modestr[-6:])):
+        char = cast(Union[str, int], char)
         mode += int(char) << iteration * 3
     # END for each char
     return mode
 
 
-def get_object_type_by_name(object_type_name):
+def get_object_type_by_name(object_type_name: bytes
+                            ) -> Union[Type['Commit'], Type['TagObject'], Type['Tree'], Type['Blob']]:
     """
     :return: type suitable to handle the given object type name.
         Use the type to create new instances.
@@ -62,10 +79,10 @@ def get_object_type_by_name(object_type_name):
         from . import tree
         return tree.Tree
     else:
-        raise ValueError("Cannot handle unknown object type: %s" % object_type_name)
+        raise ValueError("Cannot handle unknown object type: %s" % object_type_name.decode())
 
 
-def utctz_to_altz(utctz):
+def utctz_to_altz(utctz: str) -> int:
     """we convert utctz to the timezone in seconds, it is the format time.altzone
     returns. Git stores it as UTC timezone which has the opposite sign as well,
     which explains the -1 * ( that was made explicit here )
@@ -73,7 +90,7 @@ def utctz_to_altz(utctz):
     return -1 * int(float(utctz) / 100 * 3600)
 
 
-def altz_to_utctz_str(altz):
+def altz_to_utctz_str(altz: int) -> str:
     """As above, but inverses the operation, returning a string that can be used
     in commit objects"""
     utci = -1 * int((float(altz) / 3600) * 100)
@@ -83,7 +100,7 @@ def altz_to_utctz_str(altz):
     return prefix + utcs
 
 
-def verify_utctz(offset):
+def verify_utctz(offset: str) -> str:
     """:raise ValueError: if offset is incorrect
     :return: offset"""
     fmt_exc = ValueError("Invalid timezone offset format: %s" % offset)
@@ -101,27 +118,28 @@ def verify_utctz(offset):
 
 
 class tzoffset(tzinfo):
-    def __init__(self, secs_west_of_utc, name=None):
+
+    def __init__(self, secs_west_of_utc: float, name: Union[None, str] = None) -> None:
         self._offset = timedelta(seconds=-secs_west_of_utc)
         self._name = name or 'fixed'
 
-    def __reduce__(self):
+    def __reduce__(self) -> Tuple[Type['tzoffset'], Tuple[float, str]]:
         return tzoffset, (-self._offset.total_seconds(), self._name)
 
-    def utcoffset(self, dt):
+    def utcoffset(self, dt) -> timedelta:
         return self._offset
 
-    def tzname(self, dt):
+    def tzname(self, dt) -> str:
         return self._name
 
-    def dst(self, dt):
+    def dst(self, dt) -> timedelta:
         return ZERO
 
 
 utc = tzoffset(0, 'UTC')
 
 
-def from_timestamp(timestamp, tz_offset):
+def from_timestamp(timestamp, tz_offset: float) -> datetime:
     """Converts a timestamp + tz_offset into an aware datetime instance."""
     utc_dt = datetime.fromtimestamp(timestamp, utc)
     try:
@@ -131,7 +149,7 @@ def from_timestamp(timestamp, tz_offset):
         return utc_dt
 
 
-def parse_date(string_date):
+def parse_date(string_date: str) -> Tuple[int, int]:
     """
     Parse the given date as one of the following
 
@@ -152,18 +170,18 @@ def parse_date(string_date):
     # git time
     try:
         if string_date.count(' ') == 1 and string_date.rfind(':') == -1:
-            timestamp, offset = string_date.split()
+            timestamp, offset_str = string_date.split()
             if timestamp.startswith('@'):
                 timestamp = timestamp[1:]
-            timestamp = int(timestamp)
-            return timestamp, utctz_to_altz(verify_utctz(offset))
+            timestamp_int = int(timestamp)
+            return timestamp_int, utctz_to_altz(verify_utctz(offset_str))
         else:
-            offset = "+0000"                    # local time by default
+            offset_str = "+0000"                    # local time by default
             if string_date[-5] in '-+':
-                offset = verify_utctz(string_date[-5:])
+                offset_str = verify_utctz(string_date[-5:])
                 string_date = string_date[:-6]  # skip space as well
             # END split timezone info
-            offset = utctz_to_altz(offset)
+            offset = utctz_to_altz(offset_str)
 
             # now figure out the date and time portion - split time
             date_formats = []
@@ -218,13 +236,13 @@ def parse_date(string_date):
 _re_only_actor = re.compile(r'^.+? (.*)$')
 
 
-def parse_actor_and_date(line):
+def parse_actor_and_date(line: str) -> Tuple[Actor, int, int]:
     """Parse out the actor (author or committer) info from a line like::
 
         author Tom Preston-Werner <tom@mojombo.com> 1191999972 -0700
 
     :return: [Actor, int_seconds_since_epoch, int_timezone_offset]"""
-    actor, epoch, offset = '', 0, 0
+    actor, epoch, offset = '', '0', '0'
     m = _re_actor_epoch.search(line)
     if m:
         actor, epoch, offset = m.groups()
@@ -247,11 +265,11 @@ class ProcessStreamAdapter(object):
     it if the instance goes out of scope."""
     __slots__ = ("_proc", "_stream")
 
-    def __init__(self, process, stream_name):
+    def __init__(self, process: 'Popen', stream_name: str) -> None:
         self._proc = process
-        self._stream = getattr(process, stream_name)
+        self._stream = getattr(process, stream_name)  # type: StringIO  ## guess
 
-    def __getattr__(self, attr):
+    def __getattr__(self, attr: str) -> Any:
         return getattr(self._stream, attr)
 
 
@@ -260,29 +278,61 @@ class Traversable(object):
     """Simple interface to perform depth-first or breadth-first traversals
     into one direction.
     Subclasses only need to implement one function.
-    Instances of the Subclass must be hashable"""
+    Instances of the Subclass must be hashable
+
+    Defined subclasses = [Commit, Tree, SubModule]
+    """
     __slots__ = ()
 
+    @overload
+    @classmethod
+    def _get_intermediate_items(cls, item: 'Commit') -> Tuple['Commit', ...]:
+        ...
+
+    @overload
+    @classmethod
+    def _get_intermediate_items(cls, item: 'Submodule') -> Tuple['Submodule', ...]:
+        ...
+
+    @overload
+    @classmethod
+    def _get_intermediate_items(cls, item: 'Tree') -> Tuple['Tree', ...]:
+        ...
+
+    @overload
+    @classmethod
+    def _get_intermediate_items(cls, item: 'Traversable') -> Tuple['Traversable', ...]:
+        ...
+
     @classmethod
-    def _get_intermediate_items(cls, item):
+    def _get_intermediate_items(cls, item: 'Traversable'
+                                ) -> Sequence['Traversable']:
         """
         Returns:
-            List of items connected to the given item.
+            Tuple of items connected to the given item.
             Must be implemented in subclass
+
+        class Commit::     (cls, Commit) -> Tuple[Commit, ...]
+        class Submodule::  (cls, Submodule) -> Iterablelist[Submodule]
+        class Tree::       (cls, Tree) -> Tuple[Tree, ...]
         """
         raise NotImplementedError("To be implemented in subclass")
 
-    def list_traverse(self, *args, **kwargs):
+    def list_traverse(self, *args: Any, **kwargs: Any) -> IterableList:
         """
         :return: IterableList with the results of the traversal as produced by
             traverse()"""
-        out = IterableList(self._id_attribute_)
+        out = IterableList(self._id_attribute_)  # type: ignore[attr-defined]  # defined in sublcasses
         out.extend(self.traverse(*args, **kwargs))
         return out
 
-    def traverse(self, predicate=lambda i, d: True,
-                 prune=lambda i, d: False, depth=-1, branch_first=True,
-                 visit_once=True, ignore_self=1, as_edge=False):
+    def traverse(self,
+                 predicate: Callable[[object, int], bool] = lambda i, d: True,
+                 prune: Callable[[object, int], bool] = lambda i, d: False,
+                 depth: int = -1,
+                 branch_first: bool = True,
+                 visit_once: bool = True, ignore_self: int = 1, as_edge: bool = False
+                 ) -> Union[Iterator['Traversable'], Iterator[Tuple['Traversable', 'Traversable']]]:
         """:return: iterator yielding of items found when traversing self
 
         :param predicate: f(i,d) returns False if item i at depth d should not be included in the result
@@ -314,13 +364,16 @@ def traverse(self, predicate=lambda i, d: True,
             destination, i.e. tuple(src, dest) with the edge spanning from
             source to destination"""
         visited = set()
-        stack = Deque()
+        stack = deque()  # type: Deque[Tuple[int, Traversable, Union[Traversable, None]]]
         stack.append((0, self, None))       # self is always depth level 0
 
-        def addToStack(stack, item, branch_first, depth):
+        def addToStack(stack: Deque[Tuple[int, 'Traversable', Union['Traversable', None]]],
+                       item: 'Traversable',
+                       branch_first: bool,
+                       depth) -> None:
             lst = self._get_intermediate_items(item)
             if not lst:
-                return
+                return None
             if branch_first:
                 stack.extendleft((depth, i, item) for i in lst)
             else:
@@ -359,14 +412,14 @@ class Serializable(object):
     """Defines methods to serialize and deserialize objects from and into a data stream"""
     __slots__ = ()
 
-    def _serialize(self, stream):
+    def _serialize(self, stream: 'BytesIO') -> 'Serializable':
         """Serialize the data of this object into the given data stream
         :note: a serialized object would ``_deserialize`` into the same object
         :param stream: a file-like object
         :return: self"""
         raise NotImplementedError("To be implemented in subclass")
 
-    def _deserialize(self, stream):
+    def _deserialize(self, stream: 'BytesIO') -> 'Serializable':
         """Deserialize all information regarding this object from the stream
         :param stream: a file-like object
         :return: self"""
diff --git a/git/repo/base.py b/git/repo/base.py
index 6cc560310..5abd49618 100644
--- a/git/repo/base.py
+++ b/git/repo/base.py
@@ -530,7 +530,7 @@ def iter_trees(self, *args: Any, **kwargs: Any) -> Iterator['Tree']:
         :note: Takes all arguments known to iter_commits method"""
         return (c.tree for c in self.iter_commits(*args, **kwargs))
 
-    def tree(self, rev: Union['Commit', 'Tree', None] = None) -> 'Tree':
+    def tree(self, rev: Union['Commit', 'Tree', str, None] = None) -> 'Tree':
         """The Tree object for the given treeish revision
         Examples::
 
diff --git a/git/types.py b/git/types.py
index 91d35b567..a410cb366 100644
--- a/git/types.py
+++ b/git/types.py
@@ -7,15 +7,12 @@
 from typing import Union, Any
 
 if sys.version_info[:2] >= (3, 8):
-    from typing import Final, Literal  # noqa: F401
+    from typing import Final, Literal, SupportsIndex  # noqa: F401
 else:
-    from typing_extensions import Final, Literal  # noqa: F401
+    from typing_extensions import Final, Literal, SupportsIndex  # noqa: F401
 
 
-if sys.version_info[:2] < (3, 6):
-    # os.PathLike (PEP-519) only got introduced with Python 3.6
-    PathLike = str
-elif sys.version_info[:2] < (3, 9):
+if sys.version_info[:2] < (3, 9):
     # Python >= 3.6, < 3.9
     PathLike = Union[str, os.PathLike]
 elif sys.version_info[:2] >= (3, 9):
diff --git a/git/util.py b/git/util.py
index edbd5f1e7..516c315c1 100644
--- a/git/util.py
+++ b/git/util.py
@@ -29,7 +29,7 @@
 if TYPE_CHECKING:
     from git.remote import Remote
     from git.repo.base import Repo
-from .types import PathLike, TBD, Literal
+from .types import PathLike, TBD, Literal, SupportsIndex
 
 # ---------------------------------------------------------------------
 
@@ -971,7 +971,10 @@ def __getattr__(self, attr: str) -> Any:
         # END for each item
         return list.__getattribute__(self, attr)
 
-    def __getitem__(self, index: Union[int, slice, str]) -> Any:
+    def __getitem__(self, index: Union[SupportsIndex, int, slice, str]) -> Any:
+
+        assert isinstance(index, (int, str, slice)), "Index of IterableList should be an int or str"
+
         if isinstance(index, int):
             return list.__getitem__(self, index)
         elif isinstance(index, slice):
@@ -983,12 +986,13 @@ def __getitem__(self, index: Union[int, slice, str]) -> Any:
                 raise IndexError("No item found with id %r" % (self._prefix + index)) from e
         # END handle getattr
 
-    def __delitem__(self, index: Union[int, str, slice]) -> None:
+    def __delitem__(self, index: Union[SupportsIndex, int, slice, str]) -> Any:
+
+        assert isinstance(index, (int, str)), "Index of IterableList should be an int or str"
 
         delindex = cast(int, index)
         if not isinstance(index, int):
             delindex = -1
-            assert not isinstance(index, slice)
             name = self._prefix + index
             for i, item in enumerate(self):
                 if getattr(item, self._id_attr) == name:
diff --git a/setup.py b/setup.py
index 850d680d4..2845bbecd 100755
--- a/setup.py
+++ b/setup.py
@@ -123,7 +123,6 @@ def build_py_modules(basedir, excludes=[]):
         "Operating System :: MacOS :: MacOS X",
         "Programming Language :: Python",
         "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.5",
         "Programming Language :: Python :: 3.6",
         "Programming Language :: Python :: 3.7",
         "Programming Language :: Python :: 3.8",