kallithea Changeset - 1fd82c81118d

Changeset - 1fd82c81118d

Parent rev.

Child rev.

[Not reviewed]

default

0 3 0

Mads Kiilerich - 10 years ago 2016-02-15 19:29:26
madski@unity3d.com

vcs: better handling of invalid email addresses: don't consider them email addresses

13da89053853 was in principle right in always returning email adresses as
string ... but unfortunately the function also returned invalid email addresses
that didn't fit into strings.

To fix this, the function is refactored to always use regexp matching of valid
email addresses ... and to be simpler. The behaviour should be the same as
before for all valid email addresses.

3 files changed with 23 insertions and 19 deletions:

kallithea/lib/vcs/utils/__init__.py

kallithea/tests/vcs/test_git.py

kallithea/tests/vcs/test_utils.py

0 comments (0 inline, 0 general)

kallithea/lib/vcs/utils/__init__.py

➞

Show inline comments

 """
 This module provides some useful tools for ``vcs`` like annotate/diff html
 output. It also includes some internal helpers.
 """
 import time
 import datetime
 import re
 def makedate():
     lt = time.localtime()
     if lt[8] == 1 and time.daylight:
         tz = time.altzone
     else:
         tz = time.timezone
     return time.mktime(lt), tz
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
     if isinstance(obj, (basestring)):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
@@ @@ -129,62 +130,65 @@ def safe_str(unicode_, to_encoding=None) @@
     if not to_encoding:
         from kallithea.lib.vcs.conf import settings
         to_encoding = settings.DEFAULT_ENCODINGS
     if not isinstance(to_encoding, (list, tuple)):
         to_encoding = [to_encoding]
     for enc in to_encoding:
         try:
             return unicode_.encode(enc)
         except UnicodeEncodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(unicode_)['encoding']
         if encoding is None:
             raise UnicodeEncodeError()
         return unicode_.encode(encoding)
     except (ImportError, UnicodeEncodeError):
         return unicode_.encode(to_encoding[0], 'replace')
 # Regex taken from http://www.regular-expressions.info/email.html
 email_re = re.compile(
     r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@"""
     r"""(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?""",
     re.IGNORECASE)
 def author_email(author):
     """
     returns email address of given author.
     If any of <,> sign are found, it fallbacks to regex findall()
     and returns first found result or empty string
     Returns email address of given author string.
     If author contains <> brackets, only look inside that.
     If any RFC valid email address is found, return that.
     Else, return empty string.
     Regex taken from http://www.regular-expressions.info/email.html
     """
     if not author:
         return ''
     import re
     r = author.find('>')
     l = author.find('<')
     if l == -1 or r == -1:
         # fallback to regex match of email out of a string
         email_re = re.compile(r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!"""
                               r"""#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z"""
                               r"""0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]"""
                               r"""*[a-z0-9])?""", re.IGNORECASE)
         m = re.findall(email_re, author)
         return safe_str(m[0]) if m else ''
     l = author.find('<') + 1
     if l != 0:
         r = author.find('>', l)
         if r != -1:
             author = author[l:r]
     return safe_str(author[l + 1:r].strip())
     m = email_re.search(author)
     if m is None:
         return ''
     return safe_str(m.group(0))
 def author_name(author):
     """
     get name of author, or else username.
     It'll try to find an email in the author string and just cut it off
     to get the username
     """
     if not author:
         return ''
     if not '@' in author:
         return author
     return author.replace(author_email(author), '').replace('<', '') \
         .replace('>', '').strip()

kallithea/tests/vcs/test_git.py

➞

Show inline comments

@@ @@ -583,60 +583,60 @@ class GitChangesetTest(unittest.TestCase @@
     def test_changeset_author_is_unicode(self):
         for cs in self.repo:
             self.assertEqual(type(cs.author), unicode)
     def test_repo_files_content_is_unicode(self):
         changeset = self.repo.get_changeset()
         for node in changeset.get_node('/'):
             if node.is_file():
                 self.assertEqual(type(node.content), unicode)
     def test_wrong_path(self):
         # There is 'setup.py' in the root dir but not there:
         path = 'foo/bar/setup.py'
         tip = self.repo.get_changeset()
         self.assertRaises(VCSError, tip.get_node, path)
     def test_author_email(self):
         self.assertEqual('marcin@python-blog.com',
           self.repo.get_changeset('c1214f7e79e02fc37156ff215cd71275450cffc3') \
           .author_email)
         self.assertEqual('lukasz.balcerzak@python-center.pl',
           self.repo.get_changeset('ff7ca51e58c505fec0dd2491de52c622bb7a806b') \
           .author_email)
-        self.assertEqual('none@none',
         self.assertEqual('',
           self.repo.get_changeset('8430a588b43b5d6da365400117c89400326e7992') \
           .author_email)
     def test_author_username(self):
         self.assertEqual('Marcin Kuzminski',
           self.repo.get_changeset('c1214f7e79e02fc37156ff215cd71275450cffc3') \
           .author_name)
         self.assertEqual('Lukasz Balcerzak',
           self.repo.get_changeset('ff7ca51e58c505fec0dd2491de52c622bb7a806b') \
           .author_name)
         self.assertEqual('marcink',
+        self.assertEqual('marcink none@none',
           self.repo.get_changeset('8430a588b43b5d6da365400117c89400326e7992') \
           .author_name)
 class GitSpecificTest(unittest.TestCase):
     def test_error_is_raised_for_added_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'
         with self.assertRaises(VCSError):
             changeset.added
     def test_error_is_raised_for_changed_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'
         with self.assertRaises(VCSError):
             changeset.added
     def test_error_is_raised_for_removed_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'

kallithea/tests/vcs/test_utils.py

➞

Show inline comments

@@ @@ -185,49 +185,49 @@ class TestParseDatetime(unittest.TestCas @@
 class TestAuthorExtractors(unittest.TestCase):
     TEST_AUTHORS = [("Username Last'o'Name <username@example.com>",
                     ("Username Last'o'Name", "username@example.com")),
                   ("Username Last'o'Name Spaces < username@example.com >",
                     ("Username Last'o'Name Spaces", "username@example.com")),
                   ("Username Last'o'Name <username.lastname@example.com>",
                     ("Username Last'o'Name", "username.lastname@example.com")),
                   ('mrf RFC_SPEC <username+lastname@example.com>',
                     ('mrf RFC_SPEC', 'username+lastname@example.com')),
                   ('username <user@example.com>',
                     ('username', 'user@example.com')),
                   ('username <user@example.com',
                    ('username', 'user@example.com')),
                   ('broken missing@example.com',
                    ('broken', 'missing@example.com')),
                   ('<justemail@example.com>',
                    ('', 'justemail@example.com')),
                   ('justname',
                    ('justname', '')),
                   ('Mr Double Name withemail@example.com ',
                    ('Mr Double Name', 'withemail@example.com')),
                   (u'John Doe <джондо à éẋàṁṗłê.ç°ḿ>',
-                   (u'John Doe <\u0434\u0436\u043e\u043d\u0434\u043e \xe0 \xe9\u1e8b\xe0\u1e41\u1e57\u0142\xea.\xe7\xb0\u1e3f>', '\xd0\xb4\xd0\xb6\xd0\xbe\xd0\xbd\xd0\xb4\xd0\xbe \xc3\xa0 \xc3\xa9\xe1\xba\x8b\xc3\xa0\xe1\xb9\x81\xe1\xb9\x97\xc5\x82\xc3\xaa.\xc3\xa7\xc2\xb0\xe1\xb8\xbf')),
                    (u'John Doe <\u0434\u0436\u043e\u043d\u0434\u043e \xe0 \xe9\u1e8b\xe0\u1e41\u1e57\u0142\xea.\xe7\xb0\u1e3f>', '')),
+                  ]
     def test_author_email(self):
         for test_str, result in self.TEST_AUTHORS:
             self.assertEqual(result[1], author_email(test_str))
     def test_author_name(self):
         for test_str, result in self.TEST_AUTHORS:
             self.assertEqual(result[0], author_name(test_str))
 class TestGetDictForAttrs(unittest.TestCase):
     def test_returned_dict_has_expected_attrs(self):
         obj = mock.Mock()
         obj.NOT_INCLUDED = 'this key/value should not be included'
         obj.CONST = True
         obj.foo = 'aaa'
         obj.attrs = {'foo': 'bar'}
         obj.date = datetime.datetime(2010, 12, 31)
         obj.count = 1001

0 comments (0 inline, 0 general)