84a9b8a3fa
the naming of the function is wrong, need future refactor, this logic is valid
2015-03-24 20:14:05 -06:00
c21ca156e9
this logic should never happen
2015-03-24 20:14:05 -06:00
7be574534c
Fixing intervals for group by. Fixing bad PR
2015-03-24 20:13:05 -06:00
dc0d8a35dc
wip
2015-03-20 16:51:41 -06:00
3001448f16
Merge branch 'master' into limit-group-by
...
Conflicts:
CHANGELOG.md
2015-03-19 15:17:50 -07:00
3496105033
Remove resultTimes as it's not used
2015-03-19 16:48:47 -04:00
c13bdd5e86
refactor, bug fix, move tests to integration
2015-03-19 14:33:47 -06:00
d5f04d82c6
remove debug code
2015-03-19 13:43:04 -06:00
a6171b3382
refactor of limit/offset
2015-03-19 13:31:46 -06:00
56281fbfd5
WIP: add IsRawQuery to select and fix LIMIT and OFFSET on GROUP BY time
2015-03-19 11:41:18 -04:00
46f5521070
oops - forgot to remove temp import
2015-03-19 10:31:27 -05:00
105d0aa503
fixing failed integration tests - change how number of points calculated and the expected result of the server
2015-03-19 10:28:47 -05:00
5f29565ce2
still allow crazy limits if total points is less than max group limit
2015-03-18 18:44:31 -06:00
e78e5f3e19
Limit group by to MaxGroupByPoints (currently 100,000)
2015-03-18 18:44:30 -06:00
5ca9bb5f85
Update comment for clarity
2015-03-12 11:39:57 -07:00
63111a374c
Fix issue with time buckets and improve performance
...
When the start time fell in the middle of a group by bucket, the times for all buckets would be off by the amount of that stat time. This fixes it so group by windows are always on the bucket. However, if the user specifies a start time in the middle of a bucket, the very first bucket will only have whatever data in it that is past the start time (so it'll be a partial)
Also fixed a performance bug where if someone issued a query like: `SELECT mean(value) FROM foo GROUP BY time(10s) LIMIT 3`. Previously, the engine would create emtpy values for every 10s window since 1970, then limit it afterwards. Now, it'll just create value holders for the 3 buckets we need.
2015-03-11 19:33:55 -07:00
06e8c1b3e8
Add fill to select statements.
...
Fixes #1913
Add fill(none), fill(<num>), and fill(previous) options
2015-03-11 18:05:31 -07:00
1a545a1cc0
Update Slimit and Soffset to SLimit and SOffset
2015-03-09 20:09:47 -07:00
a12f028de4
Wire up limit and offset per series.
...
Fixes #1709 and fixes #1103
2015-03-09 19:17:03 -07:00
8d2abc5c86
Add Slimit and Soffset. Update tests and behavior to use those.
2015-03-09 18:46:05 -07:00
a032d77fa9
influxql: make walkFunctionCalls a func not method
2015-03-09 20:03:49 -04:00
3628fbe7cd
Updates based on PR feedback
2015-03-09 16:32:55 -07:00
08ecaabc16
Fix decodes on multiple fields. Remove debug info. Update tests to match new query engine output.
2015-03-09 11:17:36 -07:00
eca4af5514
Wire ip limits, fix server_test, remove warns
2015-03-07 16:30:42 -08:00
0b98d6e1d2
WIP: wired up math processing on queries
2015-03-07 15:29:57 -08:00
93f3a229c6
Fix merge conflicts
2015-03-06 16:30:26 -08:00
df3c0099e1
WIP: wire up raw queries
2015-03-06 11:23:58 -08:00
74e6f1f4db
WIP: query language refactoring compiles. mores soon
2015-03-01 22:37:09 -08:00
889f0a323f
Fix problem with merging series that have unequal number of points in group by interval
2015-02-27 01:02:34 -05:00
a8a0b9f24e
Fix raw data query where some values are null
...
I also removed the engine test for the raw planner. It's getting tested elsewhere and it would have been too difficult to make it work. Besides, that's getting replaced soon anyway.
2015-02-24 00:36:07 -05:00
b69597f376
Fix queries that return raw data without aggregates.
2015-02-24 00:36:07 -05:00
89b80dcd66
WIP: insane refactor of engine. Must come back later.
2015-02-23 18:07:01 -05:00
6ed70ebbe3
WIP: fix for ordering of queries of raw values
2015-02-17 21:40:53 -05:00
fec6764b09
WIP: wire up continuous queries
...
* Fix run to work with CQ broker
* Fix CQ handler
* Fix SetTimeRange to use RFC3339Nano
* Fix the time range tests
* Fix the parser to parse for RFC3339Nano literals in addition to the other format
* Add logic for running CQs
* Remove duplicate WriteBufferSize default setting from config
2015-02-17 21:02:18 -05:00
dcdb4b77a1
Merge branch 'master' into support_all_value_types
...
Conflicts:
database.go
server.go
2015-02-16 09:21:10 -08:00
4cd3494ddc
Update interface on iterator.Next() to pass raw data along with key and value
2015-02-14 17:12:38 -05:00
b394fc203f
WIP: updating the signature for iterators to pass around raw bytes for field values.
2015-02-13 14:04:45 -05:00
aa7073bfb0
make the fields public to all for serializing them later
2015-02-12 15:12:14 -07:00
61d4cd6a5a
shortcut calculating spread if not necessary
2015-02-12 10:40:24 -07:00
7c12afb9e1
fixing comment
2015-02-12 09:56:39 -07:00
3dd53ebe80
wire up spread function
2015-02-12 09:39:41 -07:00
59a953db97
Merge pull request #1575 from influxdb/batch-std-dev
...
emit in batches of 1000
2015-02-11 18:19:39 -07:00
79fe16fb6c
add comment about why we batch
2015-02-11 16:03:01 -07:00
9e80b2173a
Wire up first/last aggregate functions
2015-02-11 15:55:51 -07:00
4b18ddedec
magic numbers are evil
2015-02-11 15:11:45 -07:00
5911b38e94
emit in batches of 1000
2015-02-11 15:00:39 -07:00
6b9ba8baef
check to see if we found values to calculate
2015-02-11 12:37:14 -07:00
d108af5a94
fix standard deviation calculation
2015-02-11 12:33:10 -07:00
bf0c4599ba
first pass at stddev. tests incoming
2015-02-11 11:21:55 -07:00
8421d59eb0
no more pointers, refactored min/max
2015-02-11 10:45:44 -07:00
Cory LaNou
Philip O'Toole
Paul Dix
Jonathan Nutzmann
David Norton
Philip O'Toole