GeezerVison

The World Wide Web, restored to 800 pixels of glory.

GeezerVison is showing github.com — fix: validate object types when loading binary SHAs (#2254) · gitpython-developers/GitPython@0610404

File tree

Expand file treeCollapse file tree

Open diff view settings

Filter options

.basedpyright
- baseline.json
git
- diff.py
- index
  - base.py
- objects
test

Expand file treeCollapse file tree

Open diff view settings

Collapse file

`‎.basedpyright/baseline.json‎`

Copy file name to clipboardExpand all lines: .basedpyright/baseline.json

-8Lines changed: 0 additions & 8 deletions

Original file line number	Diff line number	Diff line change
`@@ -77,14 +77,6 @@`
`77`	`77`	`"lineCount": 1`
`78`	`78`	`}`
`79`	`79`	`},`
`80`		`- {`
`81`		`- "code": "reportAssignmentType",`
`82`		`- "range": {`
`83`		`- "startColumn": 38,`
`84`		`- "endColumn": 76,`
`85`		`- "lineCount": 1`
`86`		`- }`
`87`		`- },`
`88`	`80`	`{`
`89`	`81`	`"code": "reportArgumentType",`
`90`	`82`	`"range": {`

Collapse file

`‎git/diff.py‎`

Copy file name to clipboardExpand all lines: git/diff.py

+12-3Lines changed: 12 additions & 3 deletions

Original file line number	Diff line number	Diff line change
`@@ -11,6 +11,7 @@`
`11`	`11`
`12`	`12`	`from git.cmd import Git, handle_process_output`
`13`	`13`	`from git.compat import defenc`
	`14`	`+from git.objects.base import IndexObject`
`14`	`15`	`from git.objects.blob import Blob`
`15`	`16`	`from git.objects.util import mode_str_to_int`
`16`	`17`	`from git.util import finalize_process, hex_to_bin`
`@@ -35,7 +36,6 @@`
`35`	`36`	`if TYPE_CHECKING:`
`36`	`37`	`from subprocess import Popen`
`37`	`38`
`38`		`- from git.objects.base import IndexObject`
`39`	`39`	`from git.objects.commit import Commit`
`40`	`40`	`from git.objects.tree import Tree`
`41`	`41`	`from git.repo.base import Repo`
`@@ -378,6 +378,10 @@ class Diff:`
`378`	`378`	`Diffs keep information about the changed blob objects, the file mode, renames,`
`379`	`379`	`deletions and new files.`
`380`	`380`
	`381`	+ For submodule changes, ``a_blob`` and ``b_blob`` are
	`382`	+ :class:`~git.objects.base.IndexObject` instances whose SHAs refer to commits in
	`383`	`+ the submodule repository.`
	`384`	`+`
`381`	`385`	There are a few cases where ``None`` has to be expected as member variable value:
`382`	`386`
`383`	`387`	`New File::`
`@@ -481,17 +485,22 @@ def __init__(`
`481`	`485`	`repo = submodule.module()`
`482`	`486`	`break`
`483`	`487`
	`488`	`+ # Gitlinks reference commits; generic index objects preserve their path and mode.`
`484`	`489`	`self.a_blob: Union["IndexObject", None]`
`485`	`490`	`if a_blob_id is None or a_blob_id == self.NULL_HEX_SHA:`
`486`	`491`	`self.a_blob = None`
`487`	`492`	`else:`
`488`		`- self.a_blob = Blob(repo, hex_to_bin(a_blob_id), mode=self.a_mode, path=self.a_path)`
	`493`	`+ self.a_blob = (IndexObject if self.a_mode == 0o160000 else Blob)(`
	`494`	`+ repo, hex_to_bin(a_blob_id), mode=self.a_mode, path=self.a_path`
	`495`	`+ )`
`489`	`496`
`490`	`497`	`self.b_blob: Union["IndexObject", None]`
`491`	`498`	`if b_blob_id is None or b_blob_id == self.NULL_HEX_SHA:`
`492`	`499`	`self.b_blob = None`
`493`	`500`	`else:`
`494`		`- self.b_blob = Blob(repo, hex_to_bin(b_blob_id), mode=self.b_mode, path=self.b_path)`
	`501`	`+ self.b_blob = (IndexObject if self.b_mode == 0o160000 else Blob)(`
	`502`	`+ repo, hex_to_bin(b_blob_id), mode=self.b_mode, path=self.b_path`
	`503`	`+ )`
`495`	`504`
`496`	`505`	`self.new_file: bool = new_file`
`497`	`506`	`self.deleted_file: bool = deleted_file`

Collapse file

`‎git/index/base.py‎`

Copy file name to clipboardExpand all lines: git/index/base.py

+4-2Lines changed: 4 additions & 2 deletions

Original file line number	Diff line number	Diff line change
`@@ -310,7 +310,7 @@ def merge_tree(`
`310`	`310`	`return self`
`311`	`311`
`312`	`312`	`@classmethod`
`313`		`- def new(cls, repo: "Repo", *tree_sha: Union[str, Tree]) -> "IndexFile":`
	`313`	`+ def new(cls, repo: "Repo", *tree_sha: Union[str, bytes, Tree]) -> "IndexFile":`
`314`	`314`	`"""Merge the given treeish revisions into a new index which is returned.`
`315`	`315`
`316`	`316`	This method behaves like ``git-read-tree --aggressive`` when doing the merge.
`@@ -326,7 +326,9 @@ def new(cls, repo: "Repo", *tree_sha: Union[str, Tree]) -> "IndexFile":`
`326`	`326`	`If you intend to write such a merged Index, supply an alternate`
`327`	`327`	``file_path`` to its :meth:`write` method.
`328`	`328`	`"""`
`329`		`- tree_sha_bytes: List[bytes] = [to_bin_sha(str(t)) for t in tree_sha]`
	`329`	`+ tree_sha_bytes: List[bytes] = [`
	`330`	`+ to_bin_sha(t if isinstance(t, bytes) else str(t).encode("ascii")) for t in tree_sha`
	`331`	`+ ]`
`330`	`332`	`base_entries = aggressive_tree_merge(repo.odb, tree_sha_bytes)`
`331`	`333`
`332`	`334`	`inst = cls(repo)`

Collapse file

`‎git/objects/base.py‎`

Copy file name to clipboardExpand all lines: git/objects/base.py

+10-3Lines changed: 10 additions & 3 deletions

Original file line number	Diff line number	Diff line change
`@@ -12,7 +12,7 @@`
`12`	`12`	`from git.exc import WorkTreeRepositoryUnsupported`
`13`	`13`	`from git.util import LazyMixin, bin_to_hex, join_path_native, stream_copy`
`14`	`14`
`15`		`-from .util import get_object_type_by_name`
	`15`	`+from .util import _assert_object_type, get_object_type_by_name`
`16`	`16`
`17`	`17`	`# typing ------------------------------------------------------------------`
`18`	`18`
`@@ -107,6 +107,10 @@ def __init__(self, repo: "Repo", binsha: bytes) -> None:`
`107`	`107`
`108`	`108`	`:param binsha:`
`109`	`109`	`20 byte SHA1`
	`110`	`+`
	`111`	`+ :note:`
	`112`	+ Object data is loaded lazily. If `binsha` refers to a different object
	`113`	+ type, accessing that data raises :exc:`ValueError`.
`110`	`114`	`"""`
`111`	`115`	`super().__init__()`
`112`	`116`	`self.repo = repo`
`@@ -155,6 +159,7 @@ def _set_cache_(self, attr: str) -> None:`
`155`	`159`	`"""Retrieve object information."""`
`156`	`160`	`if attr == "size":`
`157`	`161`	`oinfo = self.repo.odb.info(self.binsha)`
	`162`	`+ _assert_object_type(oinfo, self.type)`
`158`	`163`	`self.size = oinfo.size # type: int`
`159`	`164`	`else:`
`160`	`165`	`super()._set_cache_(attr)`
`@@ -198,7 +203,9 @@ def data_stream(self) -> "OStream":`
`198`	`203`	`:note:`
`199`	`204`	`Returned streams must be read in order.`
`200`	`205`	`"""`
`201`		`- return self.repo.odb.stream(self.binsha)`
	`206`	`+ ostream = self.repo.odb.stream(self.binsha)`
	`207`	`+ _assert_object_type(ostream, self.type)`
	`208`	`+ return ostream`
`202`	`209`
`203`	`210`	`def stream_data(self, ostream: SupportsWrite[bytes]) -> "Object":`
`204`	`211`	`"""Write our data directly to the given output stream.`
`@@ -209,7 +216,7 @@ def stream_data(self, ostream: SupportsWrite[bytes]) -> "Object":`
`209`	`216`	`:return:`
`210`	`217`	`self`
`211`	`218`	`"""`
`212`		`- istream = self.repo.odb.stream(self.binsha)`
	`219`	`+ istream = self.data_stream`
`213`	`220`	`stream_copy(istream, ostream)`
`214`	`221`	`return self`
`215`	`222`

Collapse file

`‎git/objects/commit.py‎`

Copy file name to clipboardExpand all lines: git/objects/commit.py

+1-1Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -242,7 +242,7 @@ def replace(self, **kwargs: Any) -> "Commit":`
`242`	`242`	`def _set_cache_(self, attr: str) -> None:`
`243`	`243`	`if attr in Commit.__slots__:`
`244`	`244`	`# Read the data in a chunk, its faster - then provide a file wrapper.`
`245`		`- _binsha, _typename, self.size, stream = self.repo.odb.stream(self.binsha)`
	`245`	`+ _binsha, _typename, self.size, stream = self.data_stream`
`246`	`246`	`self._deserialize(BytesIO(stream.read()))`
`247`	`247`	`else:`
`248`	`248`	`super()._set_cache_(attr)`

Collapse file

`‎git/objects/fun.py‎`

Copy file name to clipboardExpand all lines: git/objects/fun.py

+8-2Lines changed: 8 additions & 2 deletions

Original file line number	Diff line number	Diff line change
`@@ -14,6 +14,8 @@`
`14`	`14`
`15`	`15`	`from git.compat import safe_decode, defenc`
`16`	`16`
	`17`	`+from .util import _assert_object_type`
	`18`	`+`
`17`	`19`	`# typing ----------------------------------------------`
`18`	`20`
`19`	`21`	`from typing import (`
`@@ -200,8 +202,10 @@ def traverse_trees_recursive(`
`200`	`202`	`if tree_sha is None:`
`201`	`203`	`data: List[EntryTupOrNone] = []`
`202`	`204`	`else:`
	`205`	`+ stream = odb.stream(tree_sha)`
	`206`	`+ _assert_object_type(stream, "tree")`
`203`	`207`	`# Make new list for typing as list invariant.`
`204`		`- data = list(tree_entries_from_data(odb.stream(tree_sha).read()))`
	`208`	`+ data = list(tree_entries_from_data(stream.read()))`
`205`	`209`	`# END handle muted trees`
`206`	`210`	`trees_data.append(data)`
`207`	`211`	`# END for each sha to get data for`
`@@ -268,7 +272,9 @@ def traverse_tree_recursive(odb: "ObjectDBR", tree_sha: bytes, path_prefix: str)`
`268`	`272`	`Prefix to prepend to the front of all returned paths.`
`269`	`273`	`"""`
`270`	`274`	`entries = []`
`271`		`- data = tree_entries_from_data(odb.stream(tree_sha).read())`
	`275`	`+ stream = odb.stream(tree_sha)`
	`276`	`+ _assert_object_type(stream, "tree")`
	`277`	`+ data = tree_entries_from_data(stream.read())`
`272`	`278`
`273`	`279`	`# Unpacking/packing is faster than accessing individual items.`
`274`	`280`	`for sha, mode, name in data:`

Collapse file

`‎git/objects/tag.py‎`

Copy file name to clipboardExpand all lines: git/objects/tag.py

+1-1Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -105,7 +105,7 @@ def __init__(`
`105`	`105`	`def _set_cache_(self, attr: str) -> None:`
`106`	`106`	`"""Cache all our attributes at once."""`
`107`	`107`	`if attr in TagObject.__slots__:`
`108`		`- ostream = self.repo.odb.stream(self.binsha)`
	`108`	`+ ostream = self.data_stream`
`109`	`109`	`lines: List[str] = ostream.read().decode(defenc, "replace").splitlines()`
`110`	`110`
`111`	`111`	`_obj, hexsha = lines[0].split(" ")`

Collapse file

`‎git/objects/tree.py‎`

Copy file name to clipboardExpand all lines: git/objects/tree.py

+1-1Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -208,7 +208,7 @@ def _get_intermediate_items(`
`208`	`208`	`def _set_cache_(self, attr: str) -> None:`
`209`	`209`	`if attr == "_cache":`
`210`	`210`	`# Set the data when we need it.`
`211`		`- ostream = self.repo.odb.stream(self.binsha)`
	`211`	`+ ostream = self.data_stream`
`212`	`212`	`self._cache: List[TreeCacheTup] = tree_entries_from_data(ostream.read())`
`213`	`213`	`else:`
`214`	`214`	`super()._set_cache_(attr)`

Collapse file

`‎git/objects/util.py‎`

Copy file name to clipboardExpand all lines: git/objects/util.py

+11Lines changed: 11 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -54,6 +54,8 @@`
`54`	`54`	`from io import BytesIO, StringIO`
`55`	`55`	`from subprocess import Popen`
`56`	`56`
	`57`	`+ from gitdb.base import OInfo`
	`58`	`+`
`57`	`59`	`from git.types import Protocol, runtime_checkable`
`58`	`60`
`59`	`61`	`from .blob import Blob`
`@@ -88,6 +90,15 @@ class TraverseNT(NamedTuple):`
`88`	`90`	`# { Functions`
`89`	`91`
`90`	`92`
	`93`	`+def _assert_object_type(oinfo: "OInfo", expected_type: Union[str, None]) -> None:`
	`94`	`+ """Reject mismatched object types using only their metadata."""`
	`95`	`+ if expected_type is None or oinfo.type == expected_type.encode("ascii"):`
	`96`	`+ return`
	`97`	`+ raise ValueError(`
	`98`	`+ "Object %s is a %s, not a %s" % (oinfo.hexsha.decode("ascii"), oinfo.type.decode("ascii"), expected_type)`
	`99`	`+ )`
	`100`	`+`
	`101`	`+`
`91`	`102`	`def mode_str_to_int(modestr: Union[bytes, str]) -> int:`
`92`	`103`	`"""Convert mode bits from an octal mode string to an integer mode for git.`
`93`	`104`

Collapse file

`‎test/test_base.py‎`

Copy file name to clipboardExpand all lines: test/test_base.py

+16-1Lines changed: 16 additions & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -4,6 +4,7 @@`
`4`	`4`	`# 3-Clause BSD License: https://opensource.org/license/bsd-3-clause/`
`5`	`5`
`6`	`6`	`import gc`
	`7`	`+from io import BytesIO`
`7`	`8`	`import os`
`8`	`9`	`import os.path as osp`
`9`	`10`	`import sys`
`@@ -13,7 +14,7 @@`
`13`	`14`	`from git import Repo`
`14`	`15`	`from git.objects import Blob, Commit, TagObject, Tree`
`15`	`16`	`import git.objects.base as base`
`16`		`-from git.objects.util import get_object_type_by_name`
	`17`	`+from git.objects.util import _assert_object_type, get_object_type_by_name`
`17`	`18`	`from git.util import HIDE_WINDOWS_FREEZE_ERRORS, hex_to_bin`
`18`	`19`
`19`	`20`	`from test.lib import TestBase as _TestBase, with_rw_and_rw_remote_repo, with_rw_repo`
`@@ -76,6 +77,20 @@ def test_base_object(self):`
`76`	`77`	`# Remove the file this way, instead of with a context manager or "finally",`
`77`	`78`	`# so it is only removed on success, and we can inspect the file on failure.`
`78`	`79`	`os.remove(tmpfile.name)`
	`80`	`+`
	`81`	`+ for wrong_type, attr in ((Blob, "data_stream"), (Tree, "_cache"), (Commit, "tree"), (TagObject, "object")):`
	`82`	`+ if wrong_type is obj_type:`
	`83`	`+ continue`
	`84`	`+ invalid = wrong_type(self.rorepo, binsha)`
	`85`	`+ for invalid_attr in ("size", attr):`
	`86`	`+ with self.assertRaisesRegex(ValueError, f"{hexsha}.{typename}.{wrong_type.type}"):`
	`87`	`+ getattr(invalid, invalid_attr)`
	`88`	`+ with self.assertRaises(ValueError):`
	`89`	`+ invalid.stream_data(BytesIO())`
	`90`	`+ data_stream = item.data_stream`
	`91`	`+ with self.assertRaises(ValueError):`
	`92`	`+ _assert_object_type(data_stream, wrong_type.type)`
	`93`	`+ self.assertEqual(data_stream.read(), data)`
`79`	`94`	`# END for each object type to create`
`80`	`95`
`81`	`96`	`# Each has a unique sha.`

0 commit comments

Comments

0 (0)

You are visitor 01402520 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.