)]}'
{"oslo/messaging/_drivers/impl_rabbit.py":[{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"b68b76d179104b3be1eca4526461abf74dd0649f","unresolved":false,"context_lines":[{"line_number":689,"context_line":"    def reset(self):"},{"line_number":690,"context_line":"        \"\"\"Reset a connection so it can be used again.\"\"\""},{"line_number":691,"context_line":"        if len(self.consumers):"},{"line_number":692,"context_line":"            self.channel.close()"},{"line_number":693,"context_line":"            self.channel \u003d self.connection.channel()"},{"line_number":694,"context_line":"            # work around \u0027memory\u0027 transport bug in 1.1.3"},{"line_number":695,"context_line":"            if self.memory_transport:"}],"source_content_type":"text/x-python","patch_set":1,"id":"baada198_9c581034","line":692,"updated":"2014-07-02 13:40:41.000000000","message":"Isn\u0027t the same approach applicable to Qpid driver?","commit_id":"c6d367f4514e16881fec1b20b1201ca471001286"},{"author":{"_account_id":8784,"name":"Gordon Sim","email":"gsim@redhat.com","username":"gordonsim"},"change_message_id":"969ad4de4319e1ecb1c8733fe2e53d1725a2dcb5","unresolved":false,"context_lines":[{"line_number":689,"context_line":"    def reset(self):"},{"line_number":690,"context_line":"        \"\"\"Reset a connection so it can be used again.\"\"\""},{"line_number":691,"context_line":"        if len(self.consumers):"},{"line_number":692,"context_line":"            self.channel.close()"},{"line_number":693,"context_line":"            self.channel \u003d self.connection.channel()"},{"line_number":694,"context_line":"            # work around \u0027memory\u0027 transport bug in 1.1.3"},{"line_number":695,"context_line":"            if self.memory_transport:"}],"source_content_type":"text/x-python","patch_set":1,"id":"baada198_cf772cd3","line":692,"in_reply_to":"baada198_2fa9887b","updated":"2014-07-02 14:22:58.000000000","message":"I think the drivers are already hugely divergent. They use two completely different APIs, with different models for messaging. This makes understanding the qpid driver, and all the impact a given change may have, quite a challenge, at least for me.\n\n(As a small example, the qpid.messaging Session is not stateless for sending as it has explicit sender entities attached to the session).","commit_id":"c6d367f4514e16881fec1b20b1201ca471001286"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"3c6291b3c3c565490ca440919265c709ae81ff84","unresolved":false,"context_lines":[{"line_number":689,"context_line":"    def reset(self):"},{"line_number":690,"context_line":"        \"\"\"Reset a connection so it can be used again.\"\"\""},{"line_number":691,"context_line":"        if len(self.consumers):"},{"line_number":692,"context_line":"            self.channel.close()"},{"line_number":693,"context_line":"            self.channel \u003d self.connection.channel()"},{"line_number":694,"context_line":"            # work around \u0027memory\u0027 transport bug in 1.1.3"},{"line_number":695,"context_line":"            if self.memory_transport:"}],"source_content_type":"text/x-python","patch_set":1,"id":"baada198_2fa9887b","line":692,"in_reply_to":"baada198_6f3b809c","updated":"2014-07-02 14:13:13.000000000","message":"My concern is that logic of the drivers will diverge. In theory, this wouldn\u0027t even need synchronised changes in two separate files, but since it\u0027s the case atm due to vague layer isolation, I would keep drivers as identical as possible until common code is introduced to handle the layer broker-agnostically.","commit_id":"c6d367f4514e16881fec1b20b1201ca471001286"},{"author":{"_account_id":8784,"name":"Gordon Sim","email":"gsim@redhat.com","username":"gordonsim"},"change_message_id":"cf630edd9263141f37f23cbf0956ea4864fe6f8a","unresolved":false,"context_lines":[{"line_number":689,"context_line":"    def reset(self):"},{"line_number":690,"context_line":"        \"\"\"Reset a connection so it can be used again.\"\"\""},{"line_number":691,"context_line":"        if len(self.consumers):"},{"line_number":692,"context_line":"            self.channel.close()"},{"line_number":693,"context_line":"            self.channel \u003d self.connection.channel()"},{"line_number":694,"context_line":"            # work around \u0027memory\u0027 transport bug in 1.1.3"},{"line_number":695,"context_line":"            if self.memory_transport:"}],"source_content_type":"text/x-python","patch_set":1,"id":"baada198_6f3b809c","line":692,"in_reply_to":"baada198_9c581034","updated":"2014-07-02 14:07:32.000000000","message":"It is, but it has little effect there as there on its own. The send is synchronous, and a \u0027sender\u0027 is created for each produced message, which involves (at present) querying the broker regarding exchange and queue information (also synchronous). These could in theory be fixed as well of course, but the change gets larger and more likely to have unforeseen problems (in my view).","commit_id":"c6d367f4514e16881fec1b20b1201ca471001286"}]}
