Age | Commit message (Collapse) | Author | Files | Lines |
|
5.1.1
=====
- Use AF_UNSPEC for name resolution
5.1.0
=====
- Improve performance of _get_free_channel_id, fix channel max bug
- Document memoryview usage, minor frame_writer.write_frame refactor
- Start dropping python 3.6
- Added experimental __slots__ to some classes
- Relaxed vine version for upcoming release.
- Upgraded topytest 7
|
|
They now have a tool dependency on py-setuptools instead of a DEPENDS
|
|
5.0.9:
- Append to _used_channel_ids in _used_channel_ids
|
|
5.0.8
=====
- Reduce memory usage of Connection
- Add additional error handling around code where an OSError
may be raised on failed connections. Fixes
5.0.7
=====
- Remove dependency to case
- Bugfix: not closing socket after server disconnect
|
|
All checksums have been double-checked against existing RMD160 and
SHA512 hashes
Not committed (merge conflicts...):
net/radsecproxy/distinfo
The following distfiles could not be fetched (fetched conditionally?):
./net/citrix_ica/distinfo citrix_ica-10.6.115659/en.linuxx86.tar.gz
./net/djbdns/distinfo dnscache-1.05-multiple-ip.patch
./net/djbdns/distinfo djbdns-1.05-test28.diff.xz
./net/djbdns/distinfo djbdns-1.05-ignoreip2.patch
./net/djbdns/distinfo djbdns-1.05-multiip.diff
./net/djbdns/distinfo djbdns-cachestats.patch
|
|
|
|
5.0.6
=====
- Change the order in which context.check_hostname and context.verify_mode get set
in SSLTransport._wrap_socket_sni. Fixes bug introduced in 5.0.3 where setting
context.verify_mode = ssl.CERT_NONE would raise
"ValueError: Cannot set verify_mode to CERT_NONE when check_hostname is enabled."
Setting context.check_hostname prior to setting context.verify_mode resolves the
issue.
- Remove TCP_USER_TIMEOUT option for Solaris
- Pass long_description to setup()
- Fix for tox-docker 2.0
- Moved to GitHub actions CI
5.0.5
=====
- Removed mistakenly introduced code which was causing import errors
5.0.4
=====
- Add missing load_default_certs() call to fix a regression in v5.0.3 release.
5.0.3
=====
- Change the default value of ssl_version to None. When not set, the
proper value between ssl.PROTOCOL_TLS_CLIENT and ssl.PROTOCOL_TLS_SERVER
will be selected based on the param server_side in order to create
a TLS Context object with better defaults that fit the desired
connection side.
- Change the default value of cert_reqs to None. The default value
of ctx.verify_mode is ssl.CERT_NONE, but when ssl.PROTOCOL_TLS_CLIENT
is used, ctx.verify_mode defaults to ssl.CERT_REQUIRED.
- Fix context.check_hostname logic. Checking the hostname depends on
having support of the SNI TLS extension and being provided with a
server_hostname value. Another important thing to mention is that
enabling hostname checking automatically sets verify_mode from
ssl.CERT_NONE to ssl.CERT_REQUIRED in the stdlib ssl and it cannot
be set back to ssl.CERT_NONE as long as hostname checking is enabled.
- Refactor the SNI tests to test one thing at a time and removing some
tests that were being repeated over and over.
5.0.2
=====
- Whhels are no longer universal.
- Added debug representation to Connection and *Transport classes
- Reintroduce ca_certs and ciphers parameters of SSLTransport._wrap_socket_sni()
- Fix infinite wait when using confirm_publish
5.0.1
=====
- Require vine 5.0.0.
5.0.0
=====
- Stop to use deprecated method ssl.wrap_socket.
5.0.0b1
=======
- Dropped Python 3.5 support.
- Removed additional compatibility code.
5.0.0a1
=======
- Dropped Python 2.x support.
- Dropped Python 3.4 support.
- Depend on :pypi:`vine` 5.0.0a1.
Code Cleanups & Improvements:
2.6.1
=====
- Fix buffer overflow in frame_writer after frame_max is increased. `frame_writer`
allocates a `bytearray` on intialization with a length based on the `connection.frame_max`
value. If `connection.frame_max` is changed to a larger value, this causes an
error like `pack_into requires a buffer of at least 408736 bytes`.
2.6.0
=====
- Implement speedups in cython
- Updated some tests & code improvements
- Separate logger for Connection.heartbeat_tick method
- Cython generic content
- Improve documentation a_global parameter of basic_qos() method.
- Fix saving partial read buffer on windows during socket timeout.
- Fix deserialization of long string field values that are not utf-8.
- Added simple cythonization of abstract_channel.py
- Speedups of serialization.py are more restrictive
|
|
|
|
2.5.2
- Ignore all methods except Close and Close-OK when channel/connection is closing
- Fix faulty ssl sni intiation parameters
- Undeprecate auto_delete flag for exchanges.
- Improved tests and testing environments
|
|
2.5.1
- Ignore all methods except Close and Close-OK when channel/connection is closing
- Fix faulty ssl sni intiation parameters
- Undeprecate auto_delete flag for exchanges.
- Improved tests and testing environments
|
|
|
|
2.5.0
- Drop Python 3.4
- Add new platform
- Numerious bug fixes
|
|
2.4.2:
- Added support for the Cygwin platform
- Correct offset incrementation when parsing bitmaps.
- Consequent bitmaps are now parsed correctly.
Previously the bit counter was reset with every bit.
We now reset it once per 8 bits, when we consume the next byte.
|
|
2.4.1:
- To avoid breaking the API basic_consume() now returns the consumer tag
instead of a tuple when nowait is True.
- Fix crash in basic_publish when broker does not support connection.blocked
capability.
- read_frame() is now Python 3 compatible for large payloads.
- Support float read_timeout/write_timeout.
- Always treat SSLError timeouts as socket timeouts.
- Treat EWOULDBLOCK as timeout.
This fixes a regression on Windows from 2.4.0.
|
|
2.4.0
- Fix inconsistent frame_handler return value.
The function returned by frame_handler is meant to return True
once the complete message is received and the callback is called,
False otherwise.
This fixes the return value for messages with a body split across
multiple frames, and heartbeat frames.
- Don't default content_encoding to utf-8 for bytes.
This is not an acceptable default as the content may not be
valid utf-8, and even if it is, the producer likely does not
expect the message to be decoded by the consumer.
- Fix encoding of messages with multibyte characters.
Body length was previously calculated using string length,
which may be less than the length of the encoded body when
it contains multibyte sequences. This caused the body of
the frame to be truncated.
- Respect content_encoding when encoding messages.
Previously the content_encoding was ignored and messages
were always encoded as utf-8. This caused messages to be
incorrectly decoded if content_encoding is properly respected
when decoding.
- Fix AMQP protocol header for AMQP 0-9-1.
- Add support for Python 3.7.
Change direct SSLSocket instantiation with wrap_socket.
Added Python 3.7 to CI.
- Add support for field type "x" (byte array).
- If there is an exception raised on Connection.connect or Connection.close,
ensure that the underlying transport socket is closed.
- TCP_USER_TIMEOUT has to be excluded from KNOWN_TCP_OPTS in BSD platforms.
- Handle negative acknowledgments.
- Added integration tests.
- Fix basic_consume() with no consumer_tag provided.
- Improved empty AMQPError string representation.
- Drain events before publish.
- Don't revive channel when connection is closing.
|
|
2.3.1
- Fix a regression that occurs when running amqp under Python 2.7.
2.3.2
- Fix a regression that occurs when running amqp under Python 2.7 on OSX.
TCP_USER_TIMEOUT is not available when running on OSX.
We now remove it from the set of known TCP options.
|
|
|
|
Changes 2.2.2:
- Sending empty messages no longer hangs. Instead an empty message is sent correctly.
- Fixed compatibility issues in UTF-8 encoding behavior between Py2/Py3
|
|
- Fix implicit conversion from bytes to string on the connection object.
This issue has caused Celery to crash on connection to RabbitMQ.
|
|
- Fix random delays in task execution.
- Calling ``conn.collect()`` multiple times will no longer raise an ``AttributeError`` when no channels exist.
- Fix compatibility code for Python 2.7.6.
- When running in Windows, py-amqp will no longer use the unsupported TCP option TCP_MAXSEG.
- Added support for setting the SNI hostname header.
- Authentication mechanisms were refactored to be more modular. GSSAPI authentication is now supported.
- Do not reconnect on collect.
|
|
|
|
Removes byte string comparison warnings when running under python -b.
Fix contributed by Jon Dufresne.
Linux version parsing broke when the version included a ‘+’ character (Issue 119).
Now sets default TCP settings for platforms that support them (e.g. Linux).
|
|
|
|
Problems found with existing digests:
Package haproxy distfile haproxy-1.5.14.tar.gz
159f5beb8fdc6b8059ae51b53dc935d91c0fb51f [recorded]
da39a3ee5e6b4b0d3255bfef95601890afd80709 [calculated]
Problems found locating distfiles:
Package bsddip: missing distfile bsddip-1.02.tar.Z
Package citrix_ica: missing distfile citrix_ica-10.6.115659/en.linuxx86.tar.gz
Package djbdns: missing distfile djbdns-1.05-test25.diff.bz2
Package djbdns: missing distfile djbdns-cachestats.patch
Package djbdns: missing distfile 0002-dnscache-cache-soa-records.patch
Package gated: missing distfile gated-3-5-11.tar.gz
Package owncloudclient: missing distfile owncloudclient-2.0.2.tar.xz
Package poink: missing distfile poink-1.6.tar.gz
Package ra-rtsp-proxy: missing distfile rtspd-src-1.0.0.0.tar.gz
Package ucspi-ssl: missing distfile ucspi-ssl-0.70-ucspitls-0.1.patch
Package waste: missing distfile waste-source.tar.gz
Otherwise, existing SHA1 digests verified and found to be the same on
the machine holding the existing distfiles (morden). All existing
SHA1 digests retained for now as an audit trail.
|
|
=====
- Now keeps buffer when socket times out.
- Adds ``Connection.Transport`` attribute that can be used to specify
a different transport implementation.
|
|
- Can now deserialize more AMQP types.
Now handles types ``short string``, ``short short int``,
``short short unsigned int``, ``short int``, ``short unsigned int``,
``long unsigned int``, ``long long int``, ``long long unsigned int``
and ``float`` which for some reason was missing, even in the original
amqplib module.
- SSL: Workaround for Python SSL bug.
A bug in the python socket library causes ``ssl.read/write()``
on a closed socket to raise :exc:`AttributeError` instead of
:exc:`IOError`.
Fix contributed by Craig Jellick.
- ``Transport.__del_`` now handles errors occurring at late interpreter
shutdown (Issue #36).
|
|
MASTER_SITES. Simplify Makefile based on those changes. From Changelog:
- SSL transport accidentally disconnected after read timeout.
|
|
=====
- Fixed bug where more data was requested from the socket
than was actually needed.
|
|
have a load of py-amqp directories rather than just throwing all the
distfiles in one. Hindsight 20/20.
|
|
|
|
distfile. From Changelog:
1.4.2
=====
- Heartbeat negotiation would use heartbeat value from server even
if heartbeat disabled (Issue #31).
1.4.1
=====
- Fixed error occurring when heartbeats disabled.
1.4.0
=====
- Heartbeat implementation improved (Issue #6).
The new heartbeat behavior is the same approach as taken by the
RabbitMQ java library.
This also means that clients should preferably call the ``heartbeat_tick``
method more frequently (like every second) instead of using the old
``rate`` argument (which is now ignored).
- Heartbeat interval is negotiated with the server.
- Some delay is allowed if the heartbeat is late.
- Monotonic time is used to keep track of the heartbeat
instead of relying on the caller to call the checking function
at the right time.
- NoneType is now supported in tables and arrays.
- SSLTransport: Now handles ``ENOENT``.
|
|
|
|
This is a fork of amqplib which was originally written by Barry Pederson. It is
maintained by the Celery project, and used by kombu as a pure python alternative
when librabbitmq is not available.
This library should be API compatible with librabbitmq.
|