From 798169805ad13a01c057cda0b66a18605a59a529 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20Gra=C3=B1a?= <dangra@gmail.com>
Date: Tue, 28 Feb 2012 14:32:55 -0200
Subject: [PATCH] Adapt response encoding detection to pass test cases

---
 scrapy/http/response/text.py       | 34 +++++++++++++-----------------
 scrapy/tests/test_http_response.py |  4 ++--
 scrapy/utils/encoding.py           | 11 ----------
 3 files changed, 17 insertions(+), 32 deletions(-)
 delete mode 100644 scrapy/utils/encoding.py

diff --git a/scrapy/http/response/text.py b/scrapy/http/response/text.py
index 00053f5ec..175ba02b3 100644
--- a/scrapy/http/response/text.py
+++ b/scrapy/http/response/text.py
@@ -6,10 +6,9 @@ See documentation in docs/topics/request-response.rst
 """
 
 from w3lib.encoding import html_to_unicode, resolve_encoding, \
-    html_body_declared_encoding, http_content_type_encoding
+    html_body_declared_encoding, http_content_type_encoding, to_unicode
 from scrapy.http.response import Response
 from scrapy.utils.python import memoizemethod_noargs
-from scrapy.utils.encoding import encoding_exists
 from scrapy.conf import settings
 
 
@@ -48,17 +47,7 @@ class TextResponse(Response):
 
     @property
     def encoding(self):
-        return self._get_encoding(infer=True)
-
-    def _get_encoding(self, infer=False):
-        enc = self._declared_encoding()
-        if enc and not encoding_exists(enc):
-            enc = None
-        if not enc and infer:
-            enc = self._body_inferred_encoding()
-        if not enc:
-            enc = self._DEFAULT_ENCODING
-        return resolve_encoding(enc)
+        return self._declared_encoding() or self._body_inferred_encoding()
 
     def _declared_encoding(self):
         return self._encoding or self._headers_encoding() \
@@ -67,7 +56,7 @@ class TextResponse(Response):
     def body_as_unicode(self):
         """Return body as unicode"""
         if self._cached_ubody is None:
-            self._cached_ubody = self.body.decode(self.encoding, 'scrapy_replace')
+            self._cached_ubody = to_unicode(self.body, self.encoding)
         return self._cached_ubody
 
     @memoizemethod_noargs
@@ -78,14 +67,21 @@ class TextResponse(Response):
     def _body_inferred_encoding(self):
         if self._cached_benc is None:
             content_type = self.headers.get('Content-Type')
-            benc, _ = html_to_unicode(content_type, self.body, default_encoding=self._DEFAULT_ENCODING)
+            benc, ubody = html_to_unicode(content_type, self.body, \
+                    auto_detect_fun=self._auto_detect_fun, \
+                    default_encoding=self._DEFAULT_ENCODING)
             self._cached_benc = benc
-            # XXX: is this needed?
-            # UnicodeDammit is buggy decoding utf-16
-            #if self._cached_ubody is None and benc != 'utf-16':
-            #    self._cached_ubody = dammit.unicode
+            self._cached_ubody = ubody
         return self._cached_benc
 
+    def _auto_detect_fun(self, text):
+        for enc in (self._DEFAULT_ENCODING, 'utf-8', 'cp1252'):
+            try:
+                text.decode(enc)
+            except UnicodeError:
+                continue
+            return resolve_encoding(enc)
+
     @memoizemethod_noargs
     def _body_declared_encoding(self):
         return html_body_declared_encoding(self.body)
diff --git a/scrapy/tests/test_http_response.py b/scrapy/tests/test_http_response.py
index fdf923996..2d3ca5ecf 100644
--- a/scrapy/tests/test_http_response.py
+++ b/scrapy/tests/test_http_response.py
@@ -171,8 +171,8 @@ class TextResponseTest(BaseResponseTest):
         self.assertEqual(r2._headers_encoding(), None)
         self.assertEqual(r2._declared_encoding(), 'utf-8')
         self._assert_response_encoding(r2, 'utf-8')
-        self.assertEqual(r3._headers_encoding(), "iso-8859-1")
-        self.assertEqual(r3._declared_encoding(), "iso-8859-1")
+        self.assertEqual(r3._headers_encoding(), "cp1252")
+        self.assertEqual(r3._declared_encoding(), "cp1252")
         self.assertEqual(r4._headers_encoding(), None)
         self.assertEqual(r5._headers_encoding(), None)
         self._assert_response_encoding(r5, "utf-8")
diff --git a/scrapy/utils/encoding.py b/scrapy/utils/encoding.py
deleted file mode 100644
index 7e33943cd..000000000
--- a/scrapy/utils/encoding.py
+++ /dev/null
@@ -1,11 +0,0 @@
-import codecs
-
-from w3lib.encoding import resolve_encoding
-
-def encoding_exists(encoding):
-    """Returns ``True`` if encoding is valid, otherwise returns ``False``"""
-    try:
-        codecs.lookup(resolve_encoding(encoding))
-    except LookupError:
-        return False
-    return True
-- 
GitLab