diff options
| author | Hanno Schlichting <hanno@hannosch.eu> | 2011-07-27 09:44:17 +0000 |
|---|---|---|
| committer | Hanno Schlichting <hanno@hannosch.eu> | 2011-07-27 09:44:17 +0000 |
| commit | ea3ac1206a8cc660bfba7f0911abde96c1baf7ad (patch) | |
| tree | da01aca2e2b32208a90b9ef71e8d404d16136c39 /src | |
| parent | 12c8fe589333d236d82dfd4d652891a984e81d93 (diff) | |
| download | zope-contenttype-ea3ac1206a8cc660bfba7f0911abde96c1baf7ad.tar.gz | |
Properly restore the HTML snippet detection, by looking at the entire string and not just its start.
Diffstat (limited to 'src')
| -rw-r--r-- | src/zope/contenttype/__init__.py | 10 | ||||
| -rw-r--r-- | src/zope/contenttype/tests/testContentTypes.py | 4 |
2 files changed, 8 insertions, 6 deletions
diff --git a/src/zope/contenttype/__init__.py b/src/zope/contenttype/__init__.py index 1b42fd6..c0a26ec 100644 --- a/src/zope/contenttype/__init__.py +++ b/src/zope/contenttype/__init__.py @@ -28,10 +28,9 @@ def text_type(s): """ # at least the maximum length of any tags we look for max_tags = 14 - s = s.strip()[:max_tags] - s2 = s.lower() + s2 = s.strip()[:max_tags].lower() - if len(s) == max_tags: + if len(s2) == max_tags: if s2.startswith('<html>'): return 'text/html' @@ -39,10 +38,11 @@ def text_type(s): return 'text/html' # what about encodings?? - if s.startswith('<?xml'): + if s2.startswith('<?xml'): return 'text/xml' - # we also recognize small snippets of HTML + # we also recognize small snippets of HTML - the closing tag might be + # anywhere, even at the end of if '</' in s: return 'text/html' diff --git a/src/zope/contenttype/tests/testContentTypes.py b/src/zope/contenttype/tests/testContentTypes.py index d82d5bf..a09f278 100644 --- a/src/zope/contenttype/tests/testContentTypes.py +++ b/src/zope/contenttype/tests/testContentTypes.py @@ -72,7 +72,7 @@ class ContentTypesTestCase(unittest.TestCase): self.assertEqual(text_type('<?xml version="1.0"><foo/>'), 'text/xml') self.assertEqual(text_type('<?XML version="1.0"><foo/>'), - 'text/plain') + 'text/xml') self.assertEqual(text_type('foo bar'), 'text/plain') self.assertEqual(text_type('<!DOCTYPE HTML PUBLIC ' @@ -82,6 +82,8 @@ class ContentTypesTestCase(unittest.TestCase): self.assertEqual(text_type('\n\n<!DOCTYPE html>\n'), 'text/html') # we can also parse text snippets self.assertEqual(text_type('<p>Hello</p>'), 'text/html') + longtext = 'abc ' * 100 + self.assertEqual(text_type('<p>%s</p>' % longtext), 'text/html') # See https://bugs.launchpad.net/bugs/487998 self.assertEqual(text_type(' ' * 14 + HTML), 'text/html') |