Bug 1575135 - Make configure complain on non-unicode strings. r=nalexander

Make it a hard error when the sandbox returns non-unicode strings.
This should help quickly catch any remaining non-unicode string that
are not caught by automation.

Differential Revision: https://phabricator.services.mozilla.com/D42607

--HG--
extra : moz-landing-system : lando

configure.py

@@ -10,6 +10,7 @@ import logging
 import os
 import sys
 import textwrap
+from collections import Iterable
 
 
 base_dir = os.path.abspath(os.path.dirname(__file__))
@@ -27,6 +28,7 @@ from mozbuild.util import (
     system_encoding,
 )
 import mozpack.path as mozpath
+import six
 
 
 def main(argv):
@@ -45,6 +47,27 @@ def main(argv):
     return config_status(config)
 
 
+def check_unicode(obj):
+    '''Recursively check that all strings in the object are unicode strings.'''
+    if isinstance(obj, dict):
+        result = True
+        for k, v in six.iteritems(obj):
+            if not check_unicode(k):
+                print("%s key is not unicode." % k, file=sys.stderr)
+                result = False
+            elif not check_unicode(v):
+                print("%s value is not unicode." % k, file=sys.stderr)
+                result = False
+        return result
+    if isinstance(obj, bytes):
+        return False
+    if isinstance(obj, six.text_type):
+        return True
+    if isinstance(obj, Iterable):
+        return all(check_unicode(o) for o in obj)
+    return True
+
+
 def config_status(config):
     # Sanitize config data to feed config.status
     # Ideally, all the backend and frontend code would handle the booleans, but
@@ -71,6 +94,11 @@ def config_status(config):
     sanitized_config['topobjdir'] = config['TOPOBJDIR']
     sanitized_config['mozconfig'] = config.get('MOZCONFIG')
 
+    if not check_unicode(sanitized_config):
+        print("Configuration should be all unicode.", file=sys.stderr)
+        print("Please file a bug for the above.", file=sys.stderr)
+        sys.exit(1)
+
     # Create config.status. Eventually, we'll want to just do the work it does
     # here, when we're able to skip configure tests/use cached results/not rely
     # on autoconf.

python/mozbuild/mozbuild/test/frontend/test_sandbox.py

@@ -334,17 +334,6 @@ class TestMozbuildSandbox(unittest.TestCase):
         for item in sandbox['CONFIGURE_SUBST_FILES']:
             self.assertIsInstance(item, SourcePath)
 
-    def test_invalid_utf8_substs(self):
-        """Ensure invalid UTF-8 in substs is converted with an error."""
-
-        # This is really mbcs. It's a bunch of invalid UTF-8.
-        config = MockConfig(extra_substs={'BAD_UTF8': b'\x83\x81\x83\x82\x3A'})
-
-        sandbox = MozbuildSandbox(Context(VARIABLES, config))
-
-        self.assertEqual(sandbox['CONFIG']['BAD_UTF8'],
-                         u'\ufffd\ufffd\ufffd\ufffd:')
-
     def test_invalid_exports_set_base(self):
         sandbox = self.sandbox()