Merge lp:~stevenk/python-debian/merge-and-stricter into lp:~launchpad-pqm/python-debian/devel

[Steve Kowalik]

Merge to the tip of lp:python-debian, and include a patch that makes BaseVersion stricter about versions.

[William Grant (wgrant)]

review: Approve (code*)

[Robert Collins (lifeless)]

Looks fine to me

review: Approve

[Robert Collins (lifeless)]

(the find() -> in change would be nice too, of course)

[lp:~stevenk/python-debian/merge-and-stricter]

188. By Steve Kowalik on 2011-03-29

Use in, rather than .find()

=== modified file 'debian/changelog'
--- debian/changelog	2010-08-02 22:12:54 +0000
+++ debian/changelog	2011-03-29 04:59:41 +0000
@@ -1,19 +1,23 @@
-python-debian (0.1.17~bzr184~launchpad2) hardy; urgency=low
+python-debian (0.1.18) unstable; urgency=low
+
+  * Support installation together with older versions of python-apt.
+    Original patch by Jelmer Vernooij. (Closes: #590805)
+
+ -- John Wright <[e-mail]>  Wed, 04 Aug 2010 01:16:52 -0700
+
+python-debian (0.1.17) unstable; urgency=low
 
   [ James Westby ]
   * Use os.path.normpath to hide python version differences
 
   [ John Wright ]
   * test_deb822.py: Make test_gpg_info more robust (Closes: #582878)
-  * Support installation together with older versions of python-apt.
-    Original patch by Jelmer Vernooij. (Closes: #590805)
-
-  [ Jelmer Vernooij ]
-  * Always use NativeVersion class to work around Debian bug #573592.
-  * Upload to Launchpad PPA.
-  * Downgrade source format to 1.0 to support building on Hardy.
-
- -- Jelmer Vernooij <[e-mail]>  Thu, 29 Jul 2010 14:23:02 +0200
+  * deb822: Use chardet to try to detect character encodings as necessary
+    - This is only used when the specified encoding doesn't work.  It's mainly
+      useful for files containing multiple deb822 paragraphs with mixed
+      encodings, like etch's Sources file. (Closes: #586021)
+
+ -- John Wright <[e-mail]>  Sun, 25 Jul 2010 02:07:11 -0700
 
 python-debian (0.1.16) unstable; urgency=low
 

=== modified file 'debian/control'
--- debian/control	2010-08-02 22:12:54 +0000
+++ debian/control	2011-03-29 04:59:41 +0000
@@ -8,7 +8,7 @@
  Reinhard Tartler <[e-mail]>,
  Stefano Zacchiroli <[e-mail]>,
  John Wright <[e-mail]>
-Build-Depends: debhelper (>= 5.0.37.2), python, python-setuptools
+Build-Depends: debhelper (>= 5.0.37.2), python, python-setuptools, python-chardet
 Build-Depends-Indep: python-support (>= 0.3)
 Standards-Version: 3.8.4
 Vcs-Browser: http://git.debian.org/?p=pkg-python-debian/python-debian.git
@@ -16,7 +16,7 @@
 
 Package: python-debian
 Architecture: all
-Depends: ${python:Depends}, ${misc:Depends}
+Depends: ${python:Depends}, ${misc:Depends}, python-chardet
 Recommends: python-apt
 Suggests: gpgv
 Provides: python-deb822

=== modified file 'lib/debian/changelog.py'
--- lib/debian/changelog.py	2010-03-14 09:16:30 +0000
+++ lib/debian/changelog.py	2011-03-29 04:59:41 +0000
@@ -23,7 +23,10 @@
 
 """This module implements facilities to deal with Debian changelogs."""
 
+import os
+import pwd
 import re
+import socket
 import warnings
 
 import debian_support
@@ -470,3 +473,77 @@
 
     def write_to_open_file(self, file):
         file.write(self.__str__())
+
+
+def get_maintainer():
+    """Get the maintainer information in the same manner as dch.
+
+    This function gets the information about the current user for
+    the maintainer field using environment variables of gecos
+    informations as approriate.
+
+    It uses the same methods as dch to get the information, namely
+    DEBEMAIL, DEBFULLNAME, EMAIL, NAME, /etc/mailname and gecos.
+
+    :returns: a tuple of the full name, email pair as strings.
+        Either of the pair may be None if that value couldn't
+        be determined.
+    """
+    env = os.environ
+    regex = re.compile(r"^(.*)\s+<(.*)>$")
+
+    # Split email and name
+    if 'DEBEMAIL' in env:
+        match_obj = regex.match(env['DEBEMAIL'])
+        if match_obj:
+            if not 'DEBFULLNAME' in env:
+                env['DEBFULLNAME'] = match_obj.group(1)
+            env['DEBEMAIL'] = match_obj.group(2)
+    if 'DEBEMAIL' not in env or 'DEBFULLNAME' not in env:
+        if 'EMAIL' in env:
+            match_obj = regex.match(env['EMAIL'])
+            if match_obj:
+                if not 'DEBFULLNAME' in env:
+                    env['DEBFULLNAME'] = match_obj.group(1)
+                env['EMAIL'] = match_obj.group(2)
+
+    # Get maintainer's name
+    if 'DEBFULLNAME' in env:
+        maintainer = env['DEBFULLNAME']
+    elif 'NAME' in env:
+        maintainer = env['NAME']
+    else:
+        # Use password database if no data in environment variables
+        try:
+            maintainer = re.sub(r',.*', '', pwd.getpwuid(os.getuid()).pw_gecos)
+        except (KeyError, AttributeError):
+            maintainer = None
+
+    # Get maintainer's mail address
+    if 'DEBEMAIL' in env:
+        email = env['DEBEMAIL']
+    elif 'EMAIL' in env:
+        email = env['EMAIL']
+    else:
+        addr = None
+        if os.path.exists('/etc/mailname'):
+            f = open('/etc/mailname')
+            try:
+                addr = f.readline().strip()
+            finally:
+                f.close()
+        if not addr:
+            addr = socket.getfqdn()
+        if addr:
+            user = pwd.getpwuid(os.getuid()).pw_name
+            if not user:
+                addr = None
+            else:
+                addr = "%s@%s" % (user, addr)
+
+        if addr:
+            email = addr
+        else:
+            email = None
+
+    return (maintainer, email)

=== modified file 'lib/debian/deb822.py'
--- lib/debian/deb822.py	2010-08-02 22:12:54 +0000
+++ lib/debian/deb822.py	2011-03-29 04:59:41 +0000
@@ -4,7 +4,7 @@
 # (.changes, .dsc, Packages, Sources, etc)
 #
 # Copyright (C) 2005-2006  dann frazier <[e-mail]>
-# Copyright (C) 2006-2008  John Wright <[e-mail]>
+# Copyright (C) 2006-2010  John Wright <[e-mail]>
 # Copyright (C) 2006       Adeodato Simó <[e-mail]>
 # Copyright (C) 2008       Stefano Zacchiroli <[e-mail]>
 #
@@ -32,10 +32,13 @@
 except (ImportError, AttributeError):
     _have_apt_pkg = False
 
+import chardet
 import new
 import re
 import string
 import sys
+import warnings
+
 import StringIO
 import UserDict
 
@@ -178,7 +181,25 @@
 
         if isinstance(value, str):
             # Always return unicode objects instead of strings
-            value = value.decode(self.encoding)
+            try:
+                value = value.decode(self.encoding)
+            except UnicodeDecodeError, e:
+                # Evidently, the value wasn't encoded with the encoding the
+                # user specified.  Try detecting it.
+                warnings.warn('decoding from %s failed; attempting to detect '
+                              'the true encoding' % self.encoding,
+                              UnicodeWarning)
+                result = chardet.detect(value)
+                try:
+                    value = value.decode(result['encoding'])
+                except UnicodeDecodeError:
+                    raise e
+                else:
+                    # Assume the rest of the paragraph is in this encoding as
+                    # well (there's no sense in repeating this exercise for
+                    # every field).
+                    self.encoding = result['encoding']
+
         return value
 
     def __delitem__(self, key):
@@ -309,33 +330,29 @@
         curkey = None
         content = ""
         for line in self.gpg_stripped_paragraph(sequence):
-            if isinstance(line, str):
-                line = line.decode(self.encoding)
             m = single.match(line)
             if m:
                 if curkey:
-                    self[curkey] += content
+                    self[curkey] = content
 
                 if not wanted_field(m.group('key')):
                     curkey = None
                     continue
 
                 curkey = m.group('key')
-                self[curkey] = m.group('data')
-                content = ""
+                content = m.group('data')
                 continue
 
             m = multi.match(line)
             if m:
                 if curkey:
-                    self[curkey] += content
+                    self[curkey] = content
 
                 if not wanted_field(m.group('key')):
                     curkey = None
                     continue
 
                 curkey = m.group('key')
-                self[curkey] = ""
                 content = ""
                 continue
 
@@ -345,7 +362,7 @@
                 continue
 
         if curkey:
-            self[curkey] += content
+            self[curkey] = content
 
     def __str__(self):
         return self.dump()

=== modified file 'lib/debian/debian_support.py'
--- lib/debian/debian_support.py	2010-03-14 11:37:42 +0000
+++ lib/debian/debian_support.py	2011-03-29 04:59:41 +0000
@@ -94,6 +94,10 @@
         m = self.re_valid_version.match(version)
         if not m:
             raise ValueError("Invalid version string %r" % version)
+        # If there no epoch ("1:..."), then the upstream version can not
+        # contain a :.
+        if m.group("epoch") is None and ':' in m.group("upstream_version"):
+            raise ValueError("Invalid version string %r" % version)
 
         self.__full_version = version
         self.__epoch = m.group("epoch")

=== added file 'tests/test_Sources.mixed_encoding'
--- tests/test_Sources.mixed_encoding	1970-01-01 00:00:00 +0000
+++ tests/test_Sources.mixed_encoding	2011-03-29 04:59:41 +0000
@@ -0,0 +1,34 @@
+Package: amarok
+Binary: amarok, amarok-engines, amarok-xine
+Version: 1.4.4-4etch1
+Priority: optional
+Section: kde
+Maintainer: Adeodato Simó <[e-mail]>
+Build-Depends: cdbs, debhelper (>= 5), quilt, bzip2, automake1.9, libtool, kdelibs4-dev, kdemultimedia-dev, kdebase-dev, libxine-dev, libtag1-dev (>> 1.4), libsqlite3-dev, libtunepimp3-dev, libmysqlclient15-dev, libpq-dev, xmms-dev, libvisual-0.4-dev, libsdl1.2-dev, libifp-dev, libusb-dev, libgpod-dev, libnjb-dev, ruby, ruby1.8-dev, dpkg-dev (>= 1.13.19)
+Architecture: any
+Standards-Version: 3.7.2
+Format: 1.0
+Directory: pool/main/a/amarok
+Files:
+ f8e80af55fbd8386e6b13b0b12d798f4 986 amarok_1.4.4-4etch1.dsc
+ 0adbbd8373da2198b80e509618a2dab9 17628566 amarok_1.4.4.orig.tar.gz
+ c29b0538c033ededacc6d31339d17700 42402 amarok_1.4.4-4etch1.diff.gz
+Uploaders: Ana Beatriz Guerrero Lopez <[e-mail]>
+
+Package: texinfo
+Binary: texinfo, info
+Version: 4.8.dfsg.1-4
+Priority: important
+Section: doc
+Maintainer: Norbert Preining <[e-mail]>
+Build-Depends: debhelper (>= 5), dpatch, libncurses5-dev | libncurses-dev, gettext
+Architecture: any
+Standards-Version: 3.7.2
+Format: 1.0
+Directory: pool/main/t/texinfo
+Files:
+ 2c233d2bf6627eac32deb9bb87726ea1 680 texinfo_4.8.dfsg.1-4.dsc
+ 614273ac8568a25926aae374cd9a6683 1926534 texinfo_4.8.dfsg.1.orig.tar.gz
+ e01520524bc114d90a2a1e5eefe71b50 101211 texinfo_4.8.dfsg.1-4.diff.gz
+Uploaders: Frank Küster <[e-mail]>
+

=== modified file 'tests/test_changelog.py'
--- tests/test_changelog.py	2010-03-14 09:16:30 +0000
+++ tests/test_changelog.py	2011-03-29 04:59:41 +0000
@@ -202,6 +202,8 @@
         self._test_version('2:1.0.4+svn26-1ubuntu1', '2', '1.0.4+svn26',
                            '1ubuntu1')
         self._test_version('2:1.0.4~rc2-1', '2', '1.0.4~rc2', '1')
+        self.assertRaises(
+            ValueError, changelog.Version, 'a1:1.8.8-070403-1~priv1')
 
     def test_version_updating(self):
         v = changelog.Version('1:1.4.1-1')

=== modified file 'tests/test_deb822.py'
--- tests/test_deb822.py	2010-07-25 07:12:28 +0000
+++ tests/test_deb822.py	2011-03-29 04:59:41 +0000
@@ -21,6 +21,7 @@
 import re
 import sys
 import unittest
+import warnings
 from StringIO import StringIO
 
 sys.path.insert(0, '../lib/debian/')
@@ -702,6 +703,29 @@
         self.assertEqual(utf8_contents, latin1_to_utf8.getvalue())
         self.assertEqual(latin1_contents, utf8_to_latin1.getvalue())
 
+    def test_mixed_encodings(self):
+        """Test that we can handle a simple case of mixed encodings
+
+        In general, this isn't guaranteed to work.  It uses the chardet
+        package, which tries to determine heuristically the encoding of the
+        text given to it.  But as far as I've seen, it's reliable for mixed
+        latin1 and utf-8 in maintainer names in old Sources files...
+        """
+
+        # Avoid spitting out the encoding warning during testing.
+        warnings.filterwarnings(action='ignore', category=UnicodeWarning)
+
+        filename = 'test_Sources.mixed_encoding'
+        for paragraphs in [deb822.Sources.iter_paragraphs(file(filename)),
+                           deb822.Sources.iter_paragraphs(file(filename),
+                                                          use_apt_pkg=False)]:
+            p1 = paragraphs.next()
+            self.assertEqual(p1['maintainer'],
+                             u'Adeodato Simó <[e-mail]>')
+            p2 = paragraphs.next()
+            self.assertEqual(p2['uploaders'],
+                             u'Frank Küster <[e-mail]>')
+
 class TestPkgRelations(unittest.TestCase):
 
     def test_packages(self):