fix(surrogateescape): enable on py2, fix tests

Byron · Byron · commit 9e4a4545dd51 · 2016-10-16T15:09:38.000+02:00
diff --git a/git/compat.py b/git/compat.py
@@ -308,6 +308,6 @@ def register_surrogateescape():
 
 
 try:
- "hello".decode(defenc, "surrogateescape")
+ b"100644 \x9f\0aaa".decode(defenc, "surrogateescape")
 except:
 register_surrogateescape()
diff --git a/git/objects/fun.py b/git/objects/fun.py
@@ -2,6 +2,7 @@
 from stat import S_ISDIR
 from git.compat import (
 byte_ord,
+ safe_decode,
 defenc,
 xrange,
 text_type,
@@ -76,7 +77,7 @@ def tree_entries_from_data(data):
 # default encoding for strings in git is utf8
 # Only use the respective unicode object if the byte stream was encoded
 name = data[ns:i]
- name = name.decode(defenc, 'surrogateescape')
+ name = safe_decode(name)
 
 # byte is NULL, get next 20
 i += 1
diff --git a/git/test/test_fun.py b/git/test/test_fun.py
@@ -16,7 +16,9 @@
 from gitdb.util import bin_to_hex
 from gitdb.base import IStream
 from gitdb.typ import str_tree_type
+from git.compat import PY3
 
+from unittest.case import skipIf
 from stat import (
 S_IFDIR,
 S_IFREG,
@@ -256,6 +258,12 @@ def test_tree_traversal_single(self):
 assert entries
 # END for each commit
 
- def test_tree_entries_from_data_with_failing_name_decode(self):
+ @skipIf(PY3, 'odd types returned ... maybe figure it out one day')
+ def test_tree_entries_from_data_with_failing_name_decode_py2(self):
 r = tree_entries_from_data(b'100644 \x9f\0aaa')
- assert r == [(b'aaa', 33188, b'\x9f')], r
+ assert r == [('aaa', 33188, u'\udc9f')], r
+ 
+ @skipIf(not PY3, 'odd types returned ... maybe figure it out one day')
+ def test_tree_entries_from_data_with_failing_name_decode_py3(self):
+ r = tree_entries_from_data(b'100644 \x9f\0aaa')
+ assert r == [(b'aaa', 33188, '\udc9f')], r

-Original file line number
+Diff line change
 fromstatimportS_ISDIR
 fromgit.compatimport (
 byte_ord,
 +safe_decode,
 defenc,
 xrange,
 text_type,
 # default encoding for strings in git is utf8
 # Only use the respective unicode object if the byte stream was encoded
 name=data[ns:i]
 -name=name.decode(defenc, 'surrogateescape')
 +name=safe_decode(name)
 # byte is NULL, get next 20
 i+=1
-Original file line number
+Diff line change
 fromgitdb.utilimportbin_to_hex
 fromgitdb.baseimportIStream
 fromgitdb.typimportstr_tree_type
 +fromgit.compatimportPY3
 +fromunittest.caseimportskipIf
 fromstatimport (
 S_IFDIR,
 S_IFREG,
 assertentries
 # END for each commit
 -deftest_tree_entries_from_data_with_failing_name_decode(self):
 +@skipIf(PY3, 'odd types returned ... maybe figure it out one day')
 +deftest_tree_entries_from_data_with_failing_name_decode_py2(self):
 r=tree_entries_from_data(b'100644 \x9f\0aaa')
 -assertr== [(b'aaa', 33188, b'\x9f')], r
 +assertr== [('aaa', 33188, u'\udc9f')], r
++
 +@skipIf(notPY3, 'odd types returned ... maybe figure it out one day')
 +deftest_tree_entries_from_data_with_failing_name_decode_py3(self):
 +r=tree_entries_from_data(b'100644 \x9f\0aaa')
 +assertr== [(b'aaa', 33188, '\udc9f')], r