)]}'
{"midonet/os_vif/mm_ctl.py":[{"author":{"_account_id":156,"name":"Ryu Ishimoto","email":"ryu@midokura.com","username":"ryu-midokura"},"change_message_id":"aeaa555b3946c02133d66eba66255bbe1cfe0167","unresolved":false,"context_lines":[{"line_number":1,"context_line":"#! /usr/bin/env python"},{"line_number":2,"context_line":""},{"line_number":3,"context_line":"# Copyright 2016 Midokura SARL"},{"line_number":4,"context_line":"#"}],"source_content_type":"text/x-python","patch_set":1,"id":"fa31d9ce_f63f2585","line":1,"updated":"2017-02-16 13:51:22.000000000","message":"I want to discuss how this script should be distributed.  If we put this here, do you plan to remove the current mm-ctl script from midonet?\n\nWhy not keep mm-ctl as part of midolman?  That seems like the most appropriate separation of what belongs in midonet and what belongs in openstack.  This code contains the mechanism in which midonet ports are bound, and IMO, such code should remain as part of MidoNet instead of the consuming projects like OpenStack, containers, etc.\n\nThoughts?","commit_id":"260266a5da9f61b270a8e5912b4fd5202bdc64c6"},{"author":{"_account_id":6854,"name":"YAMAMOTO Takashi","email":"yamamoto@midokura.com","username":"yamamoto"},"change_message_id":"8bf9b7bba9633881452ec5ff2ef092713041d08b","unresolved":false,"context_lines":[{"line_number":1,"context_line":"#! /usr/bin/env python"},{"line_number":2,"context_line":""},{"line_number":3,"context_line":"# Copyright 2016 Midokura SARL"},{"line_number":4,"context_line":"#"}],"source_content_type":"text/x-python","patch_set":1,"id":"fa31d9ce_e0f4a288","line":1,"in_reply_to":"fa31d9ce_40872e12","updated":"2017-02-16 23:51:01.000000000","message":"this is a python client library for the binding api.\nthe one found in midolman is a command for human beings.\nthey happen to be similar, but it\u0027s just an implementation detail.\nit might or might not be better if midolman provides a library for this, but it isn\u0027t the case right now.","commit_id":"260266a5da9f61b270a8e5912b4fd5202bdc64c6"},{"author":{"_account_id":156,"name":"Ryu Ishimoto","email":"ryu@midokura.com","username":"ryu-midokura"},"change_message_id":"e2c32e0f2e0c07e1a814e5f2fb91911f51dcf550","unresolved":false,"context_lines":[{"line_number":1,"context_line":"#! /usr/bin/env python"},{"line_number":2,"context_line":""},{"line_number":3,"context_line":"# Copyright 2016 Midokura SARL"},{"line_number":4,"context_line":"#"}],"source_content_type":"text/x-python","patch_set":1,"id":"fa31d9ce_40872e12","line":1,"in_reply_to":"fa31d9ce_55be12d5","updated":"2017-02-16 23:45:12.000000000","message":"Not at all, the question is whether this script should be bundled with midolman instead since we have things like mdts and libvirt (though probably should be removed) that rely on it, and maybe others that I have forgotten about.   Why do you think it\u0027s better to be here instead of midolman?","commit_id":"260266a5da9f61b270a8e5912b4fd5202bdc64c6"},{"author":{"_account_id":156,"name":"Ryu Ishimoto","email":"ryu@midokura.com","username":"ryu-midokura"},"change_message_id":"317bf2b107850ea58a8ca536a7c40f5c4e9bb7ca","unresolved":false,"context_lines":[{"line_number":1,"context_line":"#! /usr/bin/env python"},{"line_number":2,"context_line":""},{"line_number":3,"context_line":"# Copyright 2016 Midokura SARL"},{"line_number":4,"context_line":"#"}],"source_content_type":"text/x-python","patch_set":1,"id":"fa31d9ce_c05dde49","line":1,"in_reply_to":"fa31d9ce_e0f4a288","updated":"2017-02-17 00:03:59.000000000","message":"ok so that means that mm-ctl of midolman will remain where it is right now, just not used by os-vif?  If so, that removes the main concern.\n\nAs for whether it\u0027s better if midonet provides the lib, it feels like it to me since, while they happen to be similar as you said, I don\u0027t think there are too many ways to code up http-over-socket either (and would be annoying to have to do it each time). But I suppose that\u0027s more about refactoring and should not prevent this patch from going in.","commit_id":"260266a5da9f61b270a8e5912b4fd5202bdc64c6"},{"author":{"_account_id":6854,"name":"YAMAMOTO Takashi","email":"yamamoto@midokura.com","username":"yamamoto"},"change_message_id":"95e957aa8bbb607991974ceac334113e2a021176","unresolved":false,"context_lines":[{"line_number":1,"context_line":"#! /usr/bin/env python"},{"line_number":2,"context_line":""},{"line_number":3,"context_line":"# Copyright 2016 Midokura SARL"},{"line_number":4,"context_line":"#"}],"source_content_type":"text/x-python","patch_set":1,"id":"fa31d9ce_55be12d5","line":1,"in_reply_to":"fa31d9ce_f63f2585","updated":"2017-02-16 23:04:46.000000000","message":"do you mean that the binding api is private and should not be used directly here?","commit_id":"260266a5da9f61b270a8e5912b4fd5202bdc64c6"}]}
