)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"3c8ecd920d92d390148a4c9a18247236a21d866d","unresolved":true,"context_lines":[{"line_number":7,"context_line":"db: Drop \"remove\" of index"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"We were simply referencing this function - not calling it. This has been"},{"line_number":10,"context_line":"this way since commit 817baeb9374 was first merged way back in 2013 so"},{"line_number":11,"context_line":"it clearly has not been missed."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I2623bd4fe91906e928e891fb3250c02873d7fb09"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":4,"id":"420d1002_8d9f2e68","line":10,"range":{"start_line":10,"start_character":22,"end_line":10,"end_character":33},"updated":"2021-01-19 10:42:47.000000000","message":"nit: SHA1s commit hashes are good for looking directly to git, but I prefer to provide Gerrit Change-IDs. Anyway, looked at it and agreed with you.","commit_id":"4701d28da39f2fe843f90ce02084132de9170c52"}],"nova/db/sqlalchemy/migrate_repo/versions/216_havana.py":[{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"59afd508db483553cca0ce30d35357c0b3a9277d","unresolved":false,"context_lines":[{"line_number":1380,"context_line":"        else:"},{"line_number":1381,"context_line":"            index.create(migrate_engine)"},{"line_number":1382,"context_line":""},{"line_number":1383,"context_line":"    Index(\u0027project_id\u0027, dns_domains.c.project_id).drop"},{"line_number":1384,"context_line":""},{"line_number":1385,"context_line":"    # Common foreign keys"},{"line_number":1386,"context_line":"    fkeys \u003d ["}],"source_content_type":"text/x-python","patch_set":2,"id":"3f65232a_5eb1698f","side":"PARENT","line":1383,"updated":"2020-10-22 23:06:36.000000000","message":"Hol\u0027up, this isn\u0027t actually *calling* drop, it\u0027s just referencing the function. I\u0027m not sure what you\u0027re doing is legit...","commit_id":"dc053578da0f5181ae71ae4ac79162e27b105b27"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"f48cc81a0867625c399674360f979134adbce917","unresolved":false,"context_lines":[{"line_number":1380,"context_line":"        else:"},{"line_number":1381,"context_line":"            index.create(migrate_engine)"},{"line_number":1382,"context_line":""},{"line_number":1383,"context_line":"    Index(\u0027project_id\u0027, dns_domains.c.project_id).drop"},{"line_number":1384,"context_line":""},{"line_number":1385,"context_line":"    # Common foreign keys"},{"line_number":1386,"context_line":"    fkeys \u003d ["}],"source_content_type":"text/x-python","patch_set":2,"id":"3f65232a_841ba719","side":"PARENT","line":1383,"in_reply_to":"3f65232a_35a2c39b","updated":"2020-10-23 16:33:08.000000000","message":"Right, this results in one less index\n\n  --- nova_before.sql     2020-10-23 17:23:15.930290991 +0100\n  +++ nova_after.sql      2020-10-23 17:23:40.247759060 +0100\n  @@ -276,7 +276,6 @@\n     `availability_zone` varchar(255) DEFAULT NULL,\n     `project_id` varchar(255) DEFAULT NULL,\n     PRIMARY KEY (`domain`),\n  -  KEY `project_id` (`project_id`),\n     KEY `dns_domains_domain_deleted_idx` (`domain`,`deleted`),\n     CONSTRAINT `CONSTRAINT_1` CHECK (`deleted` in (0,1))\n   ) ENGINE\u003dInnoDB DEFAULT CHARSET\u003dutf8;\n\nIt\u0027s later removed in 271 [1], so I didn\u0027t see this in the diff of the end result.\n\nI\u0027ll rework this just to drop this line, since it\u0027s clearly not doing what we think, and then wait until we squash 271 to drop the index above.\n\n[1] https://github.com/openstack/nova/blob/6a5e158756edc6c01bb23c26dcfc4c3a3df703dc/nova/db/sqlalchemy/migrate_repo/versions/271_sqlite_postgresql_indexes.py#L61-L63","commit_id":"dc053578da0f5181ae71ae4ac79162e27b105b27"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"edfa8fd65aea0b848e39cce34d2f5a32a076641b","unresolved":false,"context_lines":[{"line_number":1380,"context_line":"        else:"},{"line_number":1381,"context_line":"            index.create(migrate_engine)"},{"line_number":1382,"context_line":""},{"line_number":1383,"context_line":"    Index(\u0027project_id\u0027, dns_domains.c.project_id).drop"},{"line_number":1384,"context_line":""},{"line_number":1385,"context_line":"    # Common foreign keys"},{"line_number":1386,"context_line":"    fkeys \u003d ["}],"source_content_type":"text/x-python","patch_set":2,"id":"3f65232a_f5596bfe","side":"PARENT","line":1383,"in_reply_to":"3f65232a_5eb1698f","updated":"2020-10-23 11:48:26.000000000","message":"Wow, this is here for the past seven years [1]. I think it\u0027s a no-op since the diff looked the same before and after though. Perhaps I need to rework the commit message to reflect this. Will investigate.\n\n[1] 817baeb9374c66bd5c1fc7a60810335bf9e9338d","commit_id":"dc053578da0f5181ae71ae4ac79162e27b105b27"},{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"e5311c27c1a8ee732dd177fec63dbce9bf6f4ca9","unresolved":false,"context_lines":[{"line_number":1380,"context_line":"        else:"},{"line_number":1381,"context_line":"            index.create(migrate_engine)"},{"line_number":1382,"context_line":""},{"line_number":1383,"context_line":"    Index(\u0027project_id\u0027, dns_domains.c.project_id).drop"},{"line_number":1384,"context_line":""},{"line_number":1385,"context_line":"    # Common foreign keys"},{"line_number":1386,"context_line":"    fkeys \u003d ["}],"source_content_type":"text/x-python","patch_set":2,"id":"3f65232a_35a2c39b","side":"PARENT","line":1383,"in_reply_to":"3f65232a_f5596bfe","updated":"2020-10-23 12:11:14.000000000","message":"Well, you\u0027re definitely *no longer creating* the index on L1138, so unless there\u0027s a duplicate create somewhere else (is that even possible? Wouldn\u0027t sqlalchemy/mysql itself complain?), we\u0027re now short an index after this patch. So if the diff is the same, maybe there\u0027s something wrong with the diff process?","commit_id":"dc053578da0f5181ae71ae4ac79162e27b105b27"}]}