Fall back to 'ascii' encoding if sys.getfilesystemencoding() returns

None. Remove encoding and errors argument from pax create methods in
TarInfo, pax always uses UTF-8.
Adapt the documentation and tests to the new string/unicode concept.

2 files changed, 38 insertions, 28 deletions

diff --git a/Lib/tarfile.py b/Lib/tarfile.py
index 2f05618..bf67eab 100644
--- a/Lib/tarfile.py
+++ b/Lib/tarfile.py
@@ -167,7 +167,7 @@ TOEXEC  = 0o001           # execute/search by other
 #---------------------------------------------------------
 ENCODING = sys.getfilesystemencoding()
 if ENCODING is None:
-    ENCODING = sys.getdefaultencoding()
+    ENCODING = "ascii"
 
 #---------------------------------------------------------
 # Some useful functions
@@ -982,7 +982,7 @@ class TarInfo(object):
         elif format == GNU_FORMAT:
             return self.create_gnu_header(info, encoding, errors)
         elif format == PAX_FORMAT:
-            return self.create_pax_header(info, encoding, errors)
+            return self.create_pax_header(info)
         else:
             raise ValueError("invalid format")
 
@@ -1013,7 +1013,7 @@ class TarInfo(object):
 
         return buf + self._create_header(info, GNU_FORMAT, encoding, errors)
 
-    def create_pax_header(self, info, encoding, errors):
+    def create_pax_header(self, info):
         """Return the object as a ustar header block. If it cannot be
            represented this way, prepend a pax extended header sequence
            with supplement information.
@@ -1056,17 +1056,17 @@ class TarInfo(object):
 
         # Create a pax extended header if necessary.
         if pax_headers:
-            buf = self._create_pax_generic_header(pax_headers, XHDTYPE, encoding, errors)
+            buf = self._create_pax_generic_header(pax_headers, XHDTYPE)
         else:
             buf = b""
 
-        return buf + self._create_header(info, USTAR_FORMAT, encoding, errors)
+        return buf + self._create_header(info, USTAR_FORMAT, "ascii", "replace")
 
     @classmethod
-    def create_pax_global_header(cls, pax_headers, encoding, errors):
+    def create_pax_global_header(cls, pax_headers):
         """Return the object as a pax global header block sequence.
         """
-        return cls._create_pax_generic_header(pax_headers, XGLTYPE, encoding, errors)
+        return cls._create_pax_generic_header(pax_headers, XGLTYPE)
 
     def _posix_split_name(self, name):
         """Split a name longer than 100 chars into a prefix
@@ -1139,7 +1139,7 @@ class TarInfo(object):
                 cls._create_payload(name)
 
     @classmethod
-    def _create_pax_generic_header(cls, pax_headers, type, encoding, errors):
+    def _create_pax_generic_header(cls, pax_headers, type):
         """Return a POSIX.1-2001 extended or global header sequence
            that contains a list of keyword, value pairs. The values
            must be strings.
@@ -1166,7 +1166,7 @@ class TarInfo(object):
         info["magic"] = POSIX_MAGIC
 
         # Create pax header + record blocks.
-        return cls._create_header(info, USTAR_FORMAT, encoding, errors) + \
+        return cls._create_header(info, USTAR_FORMAT, "ascii", "replace") + \
                 cls._create_payload(records)
 
     @classmethod
@@ -1566,8 +1566,7 @@ class TarFile(object):
             self._loaded = True
 
             if self.pax_headers:
-                buf = self.tarinfo.create_pax_global_header(
-                        self.pax_headers.copy(), self.encoding, self.errors)
+                buf = self.tarinfo.create_pax_global_header(self.pax_headers.copy())
                 self.fileobj.write(buf)
                 self.offset += len(buf)
 
diff --git a/Lib/test/test_tarfile.py b/Lib/test/test_tarfile.py
index 0585131..913ab60 100644
--- a/Lib/test/test_tarfile.py
+++ b/Lib/test/test_tarfile.py
@@ -780,8 +780,8 @@ class PaxWriteTest(GNUWriteTest):
 
         tar = tarfile.open(tmpname, "w", format=tarfile.PAX_FORMAT, encoding="iso8859-1")
         t = tarfile.TarInfo()
-        t.name = "\xe4\xf6\xfc"     # non-ASCII
-        t.uid = 8**8        # too large
+        t.name = "\xe4\xf6\xfc" # non-ASCII
+        t.uid = 8**8 # too large
         t.pax_headers = pax_headers
         tar.addfile(t)
         tar.close()
@@ -794,7 +794,6 @@ class PaxWriteTest(GNUWriteTest):
 
 
 class UstarUnicodeTest(unittest.TestCase):
-    # All *UnicodeTests FIXME
 
     format = tarfile.USTAR_FORMAT
 
@@ -814,11 +813,14 @@ class UstarUnicodeTest(unittest.TestCase):
         tar.close()
 
         tar = tarfile.open(tmpname, encoding=encoding)
-        self.assert_(type(tar.getnames()[0]) is not bytes)
         self.assertEqual(tar.getmembers()[0].name, name)
         tar.close()
 
     def test_unicode_filename_error(self):
+        if self.format == tarfile.PAX_FORMAT:
+            # PAX_FORMAT ignores encoding in write mode.
+            return
+
         tar = tarfile.open(tmpname, "w", format=self.format, encoding="ascii", errors="strict")
         tarinfo = tarfile.TarInfo()
 
@@ -839,21 +841,24 @@ class UstarUnicodeTest(unittest.TestCase):
         tar.close()
 
     def test_uname_unicode(self):
-        for name in ("\xe4\xf6\xfc", "\xe4\xf6\xfc"):
-            t = tarfile.TarInfo("foo")
-            t.uname = name
-            t.gname = name
-
-            fobj = io.BytesIO()
-            tar = tarfile.open("foo.tar", mode="w", fileobj=fobj, format=self.format, encoding="iso8859-1")
-            tar.addfile(t)
-            tar.close()
-            fobj.seek(0)
+        t = tarfile.TarInfo("foo")
+        t.uname = "\xe4\xf6\xfc"
+        t.gname = "\xe4\xf6\xfc"
+
+        tar = tarfile.open(tmpname, mode="w", format=self.format, encoding="iso8859-1")
+        tar.addfile(t)
+        tar.close()
 
-            tar = tarfile.open("foo.tar", fileobj=fobj, encoding="iso8859-1")
+        tar = tarfile.open(tmpname, encoding="iso8859-1")
+        t = tar.getmember("foo")
+        self.assertEqual(t.uname, "\xe4\xf6\xfc")
+        self.assertEqual(t.gname, "\xe4\xf6\xfc")
+
+        if self.format != tarfile.PAX_FORMAT:
+            tar = tarfile.open(tmpname, encoding="ascii")
             t = tar.getmember("foo")
-            self.assertEqual(t.uname, "\xe4\xf6\xfc")
-            self.assertEqual(t.gname, "\xe4\xf6\xfc")
+            self.assertEqual(t.uname, "\ufffd\ufffd\ufffd")
+            self.assertEqual(t.gname, "\ufffd\ufffd\ufffd")
 
 
 class GNUUnicodeTest(UstarUnicodeTest):
@@ -861,6 +866,11 @@ class GNUUnicodeTest(UstarUnicodeTest):
     format = tarfile.GNU_FORMAT
 
 
+class PAXUnicodeTest(UstarUnicodeTest):
+
+    format = tarfile.PAX_FORMAT
+
+
 class AppendTest(unittest.TestCase):
     # Test append mode (cp. patch #1652681).
 
@@ -1047,6 +1057,7 @@ def test_main():
         PaxWriteTest,
         UstarUnicodeTest,
         GNUUnicodeTest,
+        PAXUnicodeTest,
         AppendTest,
         LimitsTest,
         MiscTest,