Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates and

Victor Stinner · Victor Stinner · commit 13bb71c38f18 · 2010-04-23T21:41:56.000Z
bytes strings for environment keys and values
diff --git a/Lib/os.py b/Lib/os.py
@@ -450,6 +450,8 @@ def setdefault(self, key, value):
 def getenv(key, default=None):
     """Get an environment variable, return None if it doesn't exist.
     The optional second argument can specify an alternate default."""
+    if isinstance(key, bytes):
+        key = key.decode(sys.getfilesystemencoding(), "surrogateescape")
     return environ.get(key, default)
 __all__.append("getenv")
 
diff --git a/Lib/subprocess.py b/Lib/subprocess.py
@@ -1090,7 +1090,10 @@ def _execute_child(self, args, executable, preexec_fn, close_fds,
                         fs_encoding = sys.getfilesystemencoding()
                         def fs_encode(s):
                             """Encode s for use in the env, fs or cmdline."""
-                            return s.encode(fs_encoding, 'surrogateescape')
+                            if isinstance(s, bytes):
+                                return s
+                            else:
+                                return s.encode(fs_encoding, 'surrogateescape')
 
                         # We must avoid complex work that could involve
                         # malloc or free in the child process to avoid
diff --git a/Lib/test/test_subprocess.py b/Lib/test/test_subprocess.py
@@ -782,7 +782,7 @@ def test_terminate(self):
         self.assertStderrEqual(stderr, b'')
         self.assertEqual(p.wait(), -signal.SIGTERM)
 
-    def test_surrogates(self):
+    def test_surrogates_error_message(self):
         def prepare():
             raise ValueError("surrogate:\uDCff")
 
@@ -801,6 +801,28 @@ def prepare():
         else:
             self.fail("Expected ValueError or RuntimeError")
 
+    def test_undecodable_env(self):
+        for key, value in (('test', 'abc\uDCFF'), ('test\uDCFF', '42')):
+            value_repr = repr(value).encode("ascii")
+
+            # test str with surrogates
+            script = "import os; print(repr(os.getenv(%s)))" % repr(key)
+            stdout = subprocess.check_output(
+                [sys.executable, "-c", script],
+                env={key: value})
+            stdout = stdout.rstrip(b'\n\r')
+            self.assertEquals(stdout, value_repr)
+
+            # test bytes
+            key = key.encode("ascii", "surrogateescape")
+            value = value.encode("ascii", "surrogateescape")
+            script = "import os; print(repr(os.getenv(%s)))" % repr(key)
+            stdout = subprocess.check_output(
+                [sys.executable, "-c", script],
+                env={key: value})
+            stdout = stdout.rstrip(b'\n\r')
+            self.assertEquals(stdout, value_repr)
+
 
 @unittest.skipUnless(mswindows, "Windows specific tests")
 class Win32ProcessTestCase(BaseTestCase):
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -333,6 +333,9 @@ C-API
 Library
 -------
 
+- Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates
+  and bytes strings for environment keys and values
+
 - Issue #8467: Pure Python implementation of subprocess encodes the error
   message using surrogatepass error handler to support surrogates in the
   message
diff --git a/Modules/posixmodule.c b/Modules/posixmodule.c
