)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":3012,"name":"Dina Belova","email":"dbelova@mirantis.com","username":"dbelova"},"change_message_id":"f804d6d0d8febf021a1ddbcf80de2931be7001ed","unresolved":false,"context_lines":[{"line_number":9,"context_line":"table_prefix_separator is hardcoded to be an underscore (happybase"},{"line_number":10,"context_line":"default). This fix adds the possibility to parametrize this value using"},{"line_number":11,"context_line":"a parameter in ceilometer database configuration. An example is:"},{"line_number":12,"context_line":"hbase://localhost:9090/?table_prefix\u003dceilo\u0026table_prefix_separator\u003d."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Change-Id: I761977dadce102641f91930ebf75db8177d03a6b"},{"line_number":15,"context_line":"Closes-Bug: #1331374"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"1ae5cdf2_bb6e5e94","line":12,"updated":"2014-06-19 07:01:52.000000000","message":"please add to this change the modification of the following line too:\n\nhttps://github.com/openstack/ceilometer/blob/a2139d85181bcdc4a86933f394fad279371e6378/ceilometer/tests/db.py#L71","commit_id":"c2f06908cd768648aea224fd3d46aca73d1a7a05"}],"ceilometer/tests/db.py":[{"author":{"_account_id":6676,"name":"zqfan","email":"aji.zqfan@gmail.com","username":"aji-zqfan"},"change_message_id":"9538e92c1c23f8059036238e95db7b3f12c203d4","unresolved":false,"context_lines":[{"line_number":72,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":73,"context_line":"            self._url,"},{"line_number":74,"context_line":"            uuid.uuid4().hex,"},{"line_number":75,"context_line":"            random.choice(\"_.-\")  # NOTE (szilli) Only valid separators"},{"line_number":76,"context_line":"        )"},{"line_number":77,"context_line":""},{"line_number":78,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"1ae5cdf2_840bac23","line":75,"updated":"2014-06-23 06:49:28.000000000","message":"in source code, it seems accepts anything, but test code doesn\u0027t cover the invalid case, is this OK?\n\nand if an user configure an unsuitable character (or may be multiple characters but source code only accepts the first one), and after run for a while, he meets some errors, then he want to use a better separator, then what should he do? 1) just restart the service 2) he needs some extra steps to migrate/update the database 3) he needs to drop all the old data and start from ground?","commit_id":"e4790f9c7736d2de2ed4177a18055bad7dfd06be"},{"author":{"_account_id":8748,"name":"Stefano Zilli","email":"stefano.zilli@cern.ch","username":"szilli"},"change_message_id":"062b57baaabe16778b171b5be0e1837948dd9c04","unresolved":false,"context_lines":[{"line_number":72,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":73,"context_line":"            self._url,"},{"line_number":74,"context_line":"            uuid.uuid4().hex,"},{"line_number":75,"context_line":"            random.choice(\"_.-\")  # NOTE (szilli) Only valid separators"},{"line_number":76,"context_line":"        )"},{"line_number":77,"context_line":""},{"line_number":78,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"baada198_97630c41","line":75,"in_reply_to":"1ae5cdf2_840bac23","updated":"2014-07-07 12:07:42.000000000","message":"Probably I should revrite the test part and put a random sting there. Happybase itself does not do any check on the content of the table_prefix and table_prefix_separator except that it should be a string. The table name will be just \"table_prefix + table_prefix_separator + name\". In case you put an invalid character in any of these strings you will receive an error from hbase itself. As the real name of the table in hbase is the sum of these 3 fields, if this concatenation is different, they should indeed migrate the table on hbase (snapshot-\u003eclone_from_snapshot ?). Of course this is valid only when table_prefix is set otherwise only the name will be used.","commit_id":"e4790f9c7736d2de2ed4177a18055bad7dfd06be"},{"author":{"_account_id":7478,"name":"Nadya Shakhat","email":"nadmi4@gmail.com","username":"nprivalova"},"change_message_id":"961cbdc2386bc8ea0f3952d6ce97a09f0f675979","unresolved":false,"context_lines":[{"line_number":74,"context_line":"            self.url, \u0027ceilometer.alarm.storage\u0027)"},{"line_number":75,"context_line":""},{"line_number":76,"context_line":"    @property"},{"line_number":77,"context_line":"    def url(self):"},{"line_number":78,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":79,"context_line":"            self._url,"},{"line_number":80,"context_line":"            uuid.uuid4().hex,"}],"source_content_type":"text/x-python","patch_set":5,"id":"baada198_fb0a6725","line":77,"updated":"2014-08-04 17:17:01.000000000","message":"So every time in tests we will get different separators? I think that random in tests is not a good idea... I understand that separator in the name of table unlikely affect tests` result but anyway. If test fails because of this it would be extremely hard to determine the reason. Maybe it would be better to create a separate test for separators? What do you think?","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"},{"author":{"_account_id":7478,"name":"Nadya Shakhat","email":"nadmi4@gmail.com","username":"nprivalova"},"change_message_id":"5290848a5df5de66b379858c958ea79212bbdd24","unresolved":false,"context_lines":[{"line_number":74,"context_line":"            self.url, \u0027ceilometer.alarm.storage\u0027)"},{"line_number":75,"context_line":""},{"line_number":76,"context_line":"    @property"},{"line_number":77,"context_line":"    def url(self):"},{"line_number":78,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":79,"context_line":"            self._url,"},{"line_number":80,"context_line":"            uuid.uuid4().hex,"}],"source_content_type":"text/x-python","patch_set":5,"id":"baada198_893cc293","line":77,"in_reply_to":"baada198_46445ab7","updated":"2014-08-04 19:11:38.000000000","message":"hm, let me think about it once more, I will comment tomorrow","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"},{"author":{"_account_id":3012,"name":"Dina Belova","email":"dbelova@mirantis.com","username":"dbelova"},"change_message_id":"99ef9ac0d9795be218ea2bef43041ff7f1b0ff9e","unresolved":false,"context_lines":[{"line_number":74,"context_line":"            self.url, \u0027ceilometer.alarm.storage\u0027)"},{"line_number":75,"context_line":""},{"line_number":76,"context_line":"    @property"},{"line_number":77,"context_line":"    def url(self):"},{"line_number":78,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":79,"context_line":"            self._url,"},{"line_number":80,"context_line":"            uuid.uuid4().hex,"}],"source_content_type":"text/x-python","patch_set":5,"id":"fa98f980_658b7ac3","line":77,"in_reply_to":"baada198_893cc293","updated":"2014-09-03 12:46:18.000000000","message":"Nadya, any new comments here? For me random separator looks weird in tests as well, but I hess we need your opinion here for sure :)","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"},{"author":{"_account_id":8748,"name":"Stefano Zilli","email":"stefano.zilli@cern.ch","username":"szilli"},"change_message_id":"64836e7a6e58d289bdcf7b715d77733903af742f","unresolved":false,"context_lines":[{"line_number":74,"context_line":"            self.url, \u0027ceilometer.alarm.storage\u0027)"},{"line_number":75,"context_line":""},{"line_number":76,"context_line":"    @property"},{"line_number":77,"context_line":"    def url(self):"},{"line_number":78,"context_line":"        return \u0027%s?table_prefix\u003d%s\u0026table_prefix_separator\u003d%s\u0027 % ("},{"line_number":79,"context_line":"            self._url,"},{"line_number":80,"context_line":"            uuid.uuid4().hex,"}],"source_content_type":"text/x-python","patch_set":5,"id":"baada198_46445ab7","line":77,"in_reply_to":"baada198_fb0a6725","updated":"2014-08-04 17:33:55.000000000","message":"I haven\u0027t really thought about this as the prefix is also a random string. I can write a separate test but I don\u0027t thing it would change much.","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"}],"doc/source/install/manual.rst":[{"author":{"_account_id":7478,"name":"Nadya Shakhat","email":"nadmi4@gmail.com","username":"nprivalova"},"change_message_id":"961cbdc2386bc8ea0f3952d6ce97a09f0f675979","unresolved":false,"context_lines":[{"line_number":100,"context_line":"    [database]"},{"line_number":101,"context_line":"    connection \u003d hbase://hbase-thrift-host:9090"},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"   It is possible to customize happybase\u0027s `table_prefix` and `table_prefix_separator`"},{"line_number":104,"context_line":"   via query string. Default values are `None` for `table_prefix` and \u0027_\u0027 for"},{"line_number":105,"context_line":"   `table_prefix_separator`."},{"line_number":106,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"baada198_8681124f","line":103,"updated":"2014-08-04 17:17:01.000000000","message":"It would be clearer if you provide an example in \u0027None\u0027 case:\nDefault values are `None` for `table_prefix` and \u0027_\u0027 for `table_prefix_separator`. So the result table name will be e.g. \u0027resource\u0027 by default. \n\nBy the way, am I right :)? \u0027resource\u0027 or \u0027_resource\u0027?","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"},{"author":{"_account_id":8748,"name":"Stefano Zilli","email":"stefano.zilli@cern.ch","username":"szilli"},"change_message_id":"64836e7a6e58d289bdcf7b715d77733903af742f","unresolved":false,"context_lines":[{"line_number":100,"context_line":"    [database]"},{"line_number":101,"context_line":"    connection \u003d hbase://hbase-thrift-host:9090"},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"   It is possible to customize happybase\u0027s `table_prefix` and `table_prefix_separator`"},{"line_number":104,"context_line":"   via query string. Default values are `None` for `table_prefix` and \u0027_\u0027 for"},{"line_number":105,"context_line":"   `table_prefix_separator`."},{"line_number":106,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"baada198_666cfe6b","line":103,"in_reply_to":"baada198_8681124f","updated":"2014-08-04 17:33:55.000000000","message":"Yes, in case of \u0027None\u0027 it\u0027s just \u0027resource\u0027.\n\nThe example with \u0027None\u0027 would be ..9090:?table_prefix_separator\u003d. and it would actually do nothing as the separator wouldn\u0027t be used. I think maybe it\u0027s better to change the example in the text with a positive one and maintain the example with both the parameters.\n\nAlso, reading again I don\u0027t like the \u0027None\u0027. It seems like you need to put a \u0027None\u0027 as a string. I will change the entire line.","commit_id":"4ea69ea6b75a2d111fc1b72bc79dce77fcb49830"}]}