python-dnspython-1.15.0-bp152.2.18

- license update: ISC
  See LICENSE (and compare http://www.spdx.org/licenses/MIT with
  http://www.spdx.org/licenses/ISC)

- Update to new upstream release 1.10.0
  * Added dns.resolver.LRUCache.
  * dns.resolver.query() will try TCP if a UDP response is truncated.
  * The python socket module's DNS methods can be now be overriden
  with implementations that use dnspython's resolver.
- Remove redundant tags/sections from specfile
- Correct license field

- Update to 1.9.4:
  - There is no new functionality in this release;
    just a few bug fixes in RRSIG and SIG code.
  - Legacy code will be eliminated for earlier versions of
    DNSSEC in a future release of dnspython.

- Fix SLE_10 build.

- Update to 1.9.3:
  Bugs fixed since 1.9.2:
  - Dnspython was erroneously doing case-insensitive comparisons
    of the names in NSEC and RRSIG RRs.
  - We now use "is" and not "==" when testing what section an RR
    is in.
  - The resolver now disallows metaqueries.
  New since 1.9.2:
  - A boolean parameter, 'raise_on_no_answer', has been added to
    the query() methods.  In no-error, no-data situations, this
    parameter determines whether NoAnswer should be raised or not.
    If True, NoAnswer is raised.  If False, then an Answer()
    object with a None rrset will be returned.
  - Resolver Answer() objects now have a canonical_name field.
  - Rdata now have a __hash__ method.
- Regenerated spec file with py2pack;
- Removed LICENSE PKG-INFO and TODO files from documentation.

- Update to 1.9.2:
  - The fix for the import problems was actually bad, but didn't
    show up in testing because the test suite's conditional
    importing code hid the problem.

- Update to 1.9.1:
  Bugs fixed since 1.9.0:
  - The dns.dnssec module didn't work with DSA due to namespace
    contamination from a "from"-style import.

- Update to 1.9.0:
  New since 1.8.0:
  - dnspython now uses poll() instead of select() when available.
  - Basic DNSSEC validation can be done using dns.dnsec.validate()
    and dns.dnssec.validate_rrsig() if you have PyCrypto 2.3 or
    later installed. Complete secure resolution is not yet
    available.
  - Added key_id() to the DNSSEC module, which computes the DNSSEC
    key id of a DNSKEY rdata.
  - Added make_ds() to the DNSSEC module, which returns the DS RR
    for a given DNSKEY rdata.
  - dnspython now raises an exception if HMAC-SHA284 or
    HMAC-SHA512 are used with a Python older than 2.5.2. (Older
    Pythons do not compute the correct value.)
  - Symbolic constants are now available for TSIG algorithm names.
  Bugs fixed since 1.8.0:
  - dns.resolver.zone_for_name() didn't handle a query response
    with a CNAME or DNAME correctly in some cases.
  - When specifying rdata types and classes as text, Unicode
    strings may now be used.
  - Hashlib compatibility issues have been fixed.
  - dns.message now imports dns.edns.
  - The TSIG algorithm value was passed incorrectly to use_tsig()
    in some cases.

- Update to 1.8.0;
- Building as noarch for openSUSE >= 11.2.

- update to version 1.5.0

- Update for the multipython build.
- Rename python-dnspython.asc to python-dnspython.keyring.
- Spec cleanup.

- use and recommend python-ecdsa and python-pycrypto for dnssec support

- New upstream release 1.15.0
  * IDNA 2008 support is now available if the "idna" module has been
    installed and IDNA 2008 is requested.  The default IDNA behavior
    is still IDNA 2003.  The new IDNA codec mechanism is currently
    only useful for direct calls to dns.name.from_text() or
    dns.name.from_unicode(), but in future releases it will be
    deployed throughout dnspython, e.g. so that you can read a
    masterfile with an IDNA 2008 codec in force.
  * By default, dns.name.to_unicode() is not strict about which
    version of IDNA the input complies with.  Strictness can be
    requested by using one of the strict IDNA codecs.
  * Add AVC RR support.
  * Some problems with newlines in various output modes have been
    addressed.
  * dns.name.to_text() now returns text and not bytes on Python 3.x
  * More miscellaneous fixes for the Python 2/3 codeline merge.
- Include readme with readme.patch as not included in upstream tarball
- Fix a bug in the tests code with 210.patch with upstream pull request #210

- New upstream release 1.14.0

- New upstream release 1.13.0

- fix build on SLE11

- update to 1.12.0:
  * The test system can now run the tests without requiring dnspython
    to be installed.
  * When reading from a masterfile, if the first content line
    started with leading whitespace, we raised an ugly exception
    instead of doing the right thing, namely using the zone origin as
    the name. [#73]  Thanks to Tassatux for reporting the issue.
  * Added dns.zone.to_text() convenience method.  Thanks to Brandon
    Whaley <redkrieg@gmail.com> for the patch.
  * The /etc/resolv.conf setting "options rotate" is now understood
    by the resolver.  If present, the resolver will shuffle the
    nameserver list each time dns.resolver.query() is called.  Thanks
    to underrun for the patch.  Note that you don't want to add
    "options rotate" to your /etc/resolv.conf if your system's
    resolver library does not understand it.  In this case, just set
    resolver.rotate = True by hand.
  * Escaping of Unicode has been corrected.  Previously we escaped
    and then converted to Unicode, but the right thing to do is
    convert to Unicode, then escape.  Also, characters > 0x7f should
    NOT be escaped in Unicode mode.  Thanks to Martin Basti for the
    patch.
  * dns.rdtypes.ANY.DNSKEY now has helpers functions to convert
    between the numeric form of the flags and a set of human-friendly
    strings.  Thanks to Petr Spacek for the patch.
  * RRSIGs did not respect relativization settings in to_text().
    Thanks to Brian Smith for reporting the bug and submitting a
    (slightly different) patch.
  * dns/rdtypes/IN/APL.py: The APL from_wire() method did not accept an
    rdata length of 0 as valid.  Thanks to salzmdan for reporting the
    problem.
  * dns/ipv6.py: Add is_mapped()
  * dns/reversename.py: Lookup IPv6 mapped IPv4 addresses in the v4
    reverse namespace.  Thanks to Devin Bayer.  Yes, I finally fixed
    this one :)
  * dns/zone.py: Do not put back an unescaped token.  This was
    causing escape processing for domain names to break.  Thanks to
    connormclaud for reporting the problem.
  * dns/message.py: Making a response didn't work correctly if the
    query was signed with TSIG and we knew the key.  Thanks to Jeffrey
    Stiles for reporting the problem.
  * dns/query.py: Fix problems with the IXFR state machine which caused
    long diffs to fail.  Thanks to James Raftery for the fix and the
    repeated prodding to get it applied :)
- enable some tests
- add python-dnspython-rpmlintrc file
- add python-ecdsa and python-pycrypto to BuildRequires for the
  DNSSec tests
- use /usr/bin/python instead of /usr/bin/env python for the example
  scripts to avoid additional dependencies

- update to 1.11.1:
  * fix syntax error exception on dns server failure
  * various fixlets for DNSSEC support

- Package LICENSE, use proper upstream URL
- Run testsuite
- Cleanup macro usage

- update to 1.11.0:
  * TLSA RR support
  * Added set_flags() method to dns.resolver.Resolver
  * Names with offsets >= 2^14 are no longer added to the compression
    table.
  * The "::" syntax is not used to shorten a single 16-bit section of
  the text form an IPv6 address.
  * Empty rdatasets are not printed.
  * DNSKEY key tags are no longer assumed to be unique.