From 20ddfb7fcded992bd0e98b9ef8797104fe0900fd Mon Sep 17 00:00:00 2001 From: Harvey Falcic Date: Sat, 24 May 2014 13:22:41 -0400 Subject: [PATCH 1/5] Python 3 'surrogateescape' docs: fix div class for Python code --- Doc/Manual/Python.html | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Doc/Manual/Python.html b/Doc/Manual/Python.html index 45725065d..6dc0ff9ba 100644 --- a/Doc/Manual/Python.html +++ b/Doc/Manual/Python.html @@ -5962,8 +5962,8 @@ When this method is called from Python 3, the return value is the following text string:

-
->>> s = test.non_utf8_c_str()
+
+>>> s = example.non_utf8_c_str()
 >>> s
 'h\udce9llo wörld'
 

@@ -5974,7 +5974,7 @@ bytes are represented as high surrogate characters that can be used to obtain
 the original byte sequence:
 

 
-
+
 >>> b = s.encode('utf-8', errors='surrogateescape')
 >>> b
 b'h\xe9llo w\xc3\xb6rld'
@@ -5985,7 +5985,7 @@ One can then attempt a different encoding, if desired (or simply leave the
 byte string as a raw sequence of bytes for use in binary protocols):
 

 
-
+
 >>> b.decode('latin-1')
 'héllo wörld'
 

@@ -5995,7 +5995,7 @@ Note, however, that text strings containing surrogate characters are rejected
 with the default strict codec error handler. For example:
 

 
-
+
 >>> with open('test', 'w') as f:
 ...     print(s, file=f)
 ...

From 5c5dfc106f00b4125f1ae7a173463afc8cdeccad Mon Sep 17 00:00:00 2001
From: Harvey Falcic 
Date: Sat, 24 May 2014 15:39:53 -0400
Subject: [PATCH 2/5] Python unicode_strings test case: restrict to Python >
 3.0

Also adjust the test method names and content to match the docs.
---
 Examples/test-suite/python/unicode_strings_runme.py | 7 +++++--
 Examples/test-suite/unicode_strings.i               | 8 ++++----
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/Examples/test-suite/python/unicode_strings_runme.py b/Examples/test-suite/python/unicode_strings_runme.py
index 2d26599aa..162f40972 100644
--- a/Examples/test-suite/python/unicode_strings_runme.py
+++ b/Examples/test-suite/python/unicode_strings_runme.py
@@ -1,4 +1,7 @@
+import sys
+
 import unicode_strings
 
-unicode_strings.test_c_str()
-unicode_strings.test_std_string()
+if sys.version_info > (3, 0):
+    unicode_strings.non_utf8_c_str()
+    unicode_strings.non_utf8_std_string()
diff --git a/Examples/test-suite/unicode_strings.i b/Examples/test-suite/unicode_strings.i
index f4a8b8b50..56063c8a4 100644
--- a/Examples/test-suite/unicode_strings.i
+++ b/Examples/test-suite/unicode_strings.i
@@ -4,12 +4,12 @@
 
 %inline %{
 
-const char* test_c_str(void) {
-        return "h\xe9llo";
+const char* non_utf8_c_str(void) {
+        return "h\xe9llo w\xc3\xb6rld";
 }
 
-std::string test_std_string(void) {
-        return std::string("h\xe9llo");
+std::string non_utf8_std_string(void) {
+        return std::string("h\xe9llo w\xc3\xb6rld");
 }
 
 %}

From 2533d0210fb2f3b2e9f441c38bd03b99f4a57a44 Mon Sep 17 00:00:00 2001
From: Harvey Falcic 
Date: Sat, 24 May 2014 16:50:33 -0400
Subject: [PATCH 3/5] unicode_strings test: check return values

---
 Examples/test-suite/python/unicode_strings_runme.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/Examples/test-suite/python/unicode_strings_runme.py b/Examples/test-suite/python/unicode_strings_runme.py
index 162f40972..0e2e1af06 100644
--- a/Examples/test-suite/python/unicode_strings_runme.py
+++ b/Examples/test-suite/python/unicode_strings_runme.py
@@ -2,6 +2,11 @@ import sys
 
 import unicode_strings
 
+# The 'u' string prefix isn't valid in Python 3.0 - 3.2 and is redundant
+# in 3.3+. Since this file is run through 2to3 before testing, though,
+# mark this as a unicode string in 2.x so it'll become a str in 3.x.
+test_string = u'h\udce9llo w\u00f6rld'
+
 if sys.version_info > (3, 0):
-    unicode_strings.non_utf8_c_str()
-    unicode_strings.non_utf8_std_string()
+    assert unicode_strings.non_utf8_c_str() == test_string
+    assert unicode_strings.non_utf8_std_string() == test_string

From ab527b0e4bf592e55b038a6de0fb8fe4540eea92 Mon Sep 17 00:00:00 2001
From: Harvey Falcic 
Date: Sat, 24 May 2014 17:54:53 -0400
Subject: [PATCH 4/5] unicode_strings_runme.py: fix version check

Python 3.0.1 shouldn't pass.
---
 Examples/test-suite/python/unicode_strings_runme.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Examples/test-suite/python/unicode_strings_runme.py b/Examples/test-suite/python/unicode_strings_runme.py
index 0e2e1af06..2110c5ce7 100644
--- a/Examples/test-suite/python/unicode_strings_runme.py
+++ b/Examples/test-suite/python/unicode_strings_runme.py
@@ -7,6 +7,6 @@ import unicode_strings
 # mark this as a unicode string in 2.x so it'll become a str in 3.x.
 test_string = u'h\udce9llo w\u00f6rld'
 
-if sys.version_info > (3, 0):
+if sys.version_info[0:2] >= (3, 1):
     assert unicode_strings.non_utf8_c_str() == test_string
     assert unicode_strings.non_utf8_std_string() == test_string

From 91e93838fc6dbc181181f6ffacb024b6a87cfe63 Mon Sep 17 00:00:00 2001
From: Harvey Falcic 
Date: Sat, 24 May 2014 18:00:04 -0400
Subject: [PATCH 5/5] unicode_strings test: manually check values instead of
 using assert

---
 Examples/test-suite/python/unicode_strings_runme.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/Examples/test-suite/python/unicode_strings_runme.py b/Examples/test-suite/python/unicode_strings_runme.py
index 2110c5ce7..e1fc7adec 100644
--- a/Examples/test-suite/python/unicode_strings_runme.py
+++ b/Examples/test-suite/python/unicode_strings_runme.py
@@ -8,5 +8,7 @@ import unicode_strings
 test_string = u'h\udce9llo w\u00f6rld'
 
 if sys.version_info[0:2] >= (3, 1):
-    assert unicode_strings.non_utf8_c_str() == test_string
-    assert unicode_strings.non_utf8_std_string() == test_string
+    if unicode_strings.non_utf8_c_str() != test_string:
+        raise ValueError('Test comparison mismatch')
+    if unicode_strings.non_utf8_std_string() != test_string:
+        raise ValueError('Test comparison mismatch')