[pypy-commit] pypy union-side-effects: Explicitly forbid comparisons between str and unicode

[rlamy]

Author: Ronan Lamy <ronan.lamy at gmail.com>
Branch: union-side-effects
Changeset: r86873:e1f64f1f25cc
Date: 2016-09-05 00:16 +0100
http://bitbucket.org/pypy/pypy/changeset/e1f64f1f25cc/

Log:	Explicitly forbid comparisons between str and unicode

diff --git a/rpython/annotator/binaryop.py b/rpython/annotator/binaryop.py
--- a/rpython/annotator/binaryop.py
+++ b/rpython/annotator/binaryop.py
@@ -641,6 +641,20 @@
             result.const = str1.const + str2.const
         return result
 
+for cmp_op in [op.lt, op.le, op.eq, op.ne, op.gt, op.ge]:
+    @cmp_op.register(SomeUnicodeString, SomeString)
+    @cmp_op.register(SomeUnicodeString, SomeChar)
+    @cmp_op.register(SomeString, SomeUnicodeString)
+    @cmp_op.register(SomeChar, SomeUnicodeString)
+    @cmp_op.register(SomeUnicodeCodePoint, SomeString)
+    @cmp_op.register(SomeUnicodeCodePoint, SomeChar)
+    @cmp_op.register(SomeString, SomeUnicodeCodePoint)
+    @cmp_op.register(SomeChar, SomeUnicodeCodePoint)
+    def cmp_str_unicode(annotator, v1, v2):
+        raise AnnotatorError(
+            "Comparing byte strings with unicode strings is not RPython")
+
+
 class __extend__(pairtype(SomeInteger, SomeList)):
 
     def mul((int1, lst2)):
diff --git a/rpython/rlib/runicode.py b/rpython/rlib/runicode.py
--- a/rpython/rlib/runicode.py
+++ b/rpython/rlib/runicode.py
@@ -1367,7 +1367,7 @@
             oc = ord(ch)
 
             # Escape quotes
-            if quotes and (oc == quote or ch == '\\'):
+            if quotes and (oc == quote or ch == u'\\'):
                 result.append(STR('\\'))
                 result.append(CHR(oc))
                 pos += 1
@@ -1390,13 +1390,13 @@
                 pos -= 1
 
             # Map special whitespace to '\t', \n', '\r'
-            if ch == '\t':
+            if ch == u'\t':
                 result.append(STR('\\t'))
-            elif ch == '\n':
+            elif ch == u'\n':
                 result.append(STR('\\n'))
-            elif ch == '\r':
+            elif ch == u'\r':
                 result.append(STR('\\r'))
-            elif ch == '\\':
+            elif ch == u'\\':
                 result.append(STR('\\\\'))
 
             # Map non-printable or non-ascii to '\xhh' or '\uhhhh'