Alejandro Martínez
0bfdeae147
Move "username" column to the last position on analysis_catalog - fixes #276
...
Due to the way it was first implemented, the "username" column would be
on a different position depending on if it was an extension upgrade or
a fresh install.
This caused problems with pg_dumping databases and restoring them.
pg_dump does not include the extension source (so this table's schema is
not included on the dump) but does include this contents, using a COPY
without column names by default and failing due to the order difference.
After this has run, all tables (whether updated or not) will have the
"username" column on the last position.
2016-10-20 14:30:02 +02:00
Rafa de la Torre
bc5e23b143
Replace qmax by nominal_quota
...
Since the analysis quota factor can be greater than 1, `qmax` can be a
misleading name. Thus the change in var name.
2016-10-19 12:41:19 +02:00
Rafa de la Torre
64fae71a37
Default factor for analysis size from 0.2 to 2
...
Change the default value for the factor of analysis tables size from 0.2
to 2. I also checked it is applied on the "nominal" user quota.
2016-10-19 12:36:57 +02:00
Rafa de la Torre
ce762f41ac
Merge pull request #280 from CartoDB/279-check-analysis
...
Implement CDB_CheckAnalysisQuota
2016-10-17 17:55:45 +02:00
Rafa de la Torre
529b12af20
Cosmetic fix: s/INT8/bigint #279
2016-10-17 17:16:04 +02:00
Rafa de la Torre
f98b6fb0a1
Merge pull request #270 from CartoDB/269-overviews-non-null
...
Exclude null geometries to create overviews
2016-10-17 17:07:17 +02:00
Javier Goizueta
066c574709
Fix: cdb_conf table is not accessible to regular users
2016-10-11 17:52:11 +02:00
Javier Goizueta
fa6f9a8a66
Implement CDB_CheckAnalysisQuota
...
see #279
2016-10-11 15:14:44 +02:00
Antonio Zamorano
b462e969a1
Adding cache_tables to the create table too
2016-08-12 11:52:24 +02:00
Antonio Zamorano
5d323456ee
Chaging the type of the cached tables to a regclass array, fixes #273
2016-08-12 09:42:50 +02:00
Antonio Zamorano
6e130c336e
Adding a new column to have the name of the cached analysis in case there is one, fixes #273
2016-08-11 16:27:18 +02:00
Javier Goizueta
457b614d96
Exclude null geometries to create overviews
...
Fixes #269
2016-07-05 16:45:12 +02:00
Raul Ochoa
5eddf5ce8e
Merge pull request #268 from CartoDB/cdb_analysis_catalog-export-config
...
Add export config for cdb_analysis_catalog table
2016-07-01 14:58:34 +02:00
Raul Ochoa
2b24390a8a
Do not raise anything on existing columns
2016-07-01 12:56:44 +02:00
Raul Ochoa
69f04bb8b0
Add export config for cdb_analysis_catalog table
...
Closes #251
2016-07-01 12:36:07 +02:00
Raul Ochoa
c96bf7c7d5
Do not use plpgsql language
2016-07-01 12:34:40 +02:00
Raul Ochoa
15a8876d06
Rename to last_* pattern
2016-07-01 12:25:05 +02:00
Raul Ochoa
89e991aae9
Document new fields
2016-07-01 12:24:44 +02:00
Raul Ochoa
064b26ccd3
Add some extra fields to cdb_analysis_catalog table
...
Track user, error_message for failures, and last entity modifying the node.
2016-07-01 12:13:31 +02:00
Javier Goizueta
2b69823949
💄 indent/parenthesize for clarity
2016-06-13 10:39:05 +02:00
Javier Goizueta
1f01ecae30
Exclude overviews from user data size
...
Fixes #261
Some internal functions from the Overviews module
have been moved to a separate file because they're now
used from Quaota function.
2016-06-13 10:04:13 +02:00
Javier Goizueta
a2a1ff6ae8
Merge pull request #258 from CartoDB/257-max-overviews-level
...
Fix problems computing feature density for overviews creation
2016-05-27 16:15:59 +02:00
Javier Goizueta
2a30eb2fd3
Fix zoom from scale condition for NULL result
...
And rewrite in cleaner form.
2016-05-27 15:10:20 +02:00
Javier Goizueta
0b3ad5e569
Limit the maximum zoom level
...
Avoid returning zoom levels greater than the maximum
'safe' level. For a zero denominator (which would imply
and infinite zoom level) return the maximum level too.
2016-05-27 12:29:52 +02:00
Paul Norman
aa302c237d
Use standard formula for CDB_ZoomFromScale
...
postgis-vt-util comes with a standard formula for zoom from
scaleDenominator, and this is well tested. This also fixes the
function throwing an error on NULL input.
2016-05-27 12:29:43 +02:00
Javier Goizueta
9526f0448f
Fix bug in feature density recursive query
...
If the table had x and/or y columns they were picked by an inner
select instead of the recursive arguments.
Fixes #256
2016-05-26 18:25:11 +02:00
Javier Goizueta
3399f2b9a5
Fix max overviews level usage
...
Note that _CDB_Feature_Density_Ref_Z_Strategy returns the first
level for which overviews should not be used, and that in some
cases _CDB_Feature_Density should look beyond the max level to
compute a feature density estimate.
2016-05-25 14:00:55 +02:00
Javier Goizueta
803b3671d0
Fixes for feature density computation
...
* The initial iteration of the recursive feature density query shouldn't
start beyond the maximum level
* Correct off-by-one limit the rest of iterations
2016-05-25 11:21:12 +02:00
Javier Goizueta
c3fada29a8
Comment how the max overview level is defined
2016-05-25 11:07:06 +02:00
Javier Goizueta
86e5f6d317
Limit the maximum overview level to 23
...
Fixes #257
23 = 32 bits per integer - 1 sign bit - 8 bits px/tile
2016-05-24 18:18:58 +02:00
Javier Goizueta
8d7860dc7a
Fixes #252
2016-05-09 11:54:56 +02:00
Javier Goizueta
b5427c65c8
Drop aggregate to be defined
...
Otherwise future versions will fail to recreate the aggregate
2016-04-29 08:46:01 +02:00
Javier Goizueta
a067cc7da1
Generate stats used to identify category columns in overviews if needed
...
This only generates the stats if no stats are available for a table.
This doesn't warrant that the stats are up to date or accurate.
2016-04-27 15:06:09 +02:00
Javier Goizueta
2c43943df6
Fix syntax
2016-04-26 18:27:52 +02:00
Javier Goizueta
417cbe7902
Fix category columns aggregation in overviews
...
Overviews are created in cascade, each one from the inmediate
lower level, but the stats to decide if a column is a category
should be taken always from the base table.
2016-04-26 18:02:25 +02:00
Javier Goizueta
9a73703954
Use mode to aggregate categorical columns in overviews
...
Fixes #245
2016-04-26 15:15:24 +02:00
Javier Goizueta
efe381ad94
Merge pull request #243 from CartoDB/241-webmercator
...
Compute webmercator resolution with full accuracy
2016-04-25 17:30:40 +02:00
Javier Goizueta
f7cce21eb7
Merge pull request #242 from CartoDB/240-overviews-pixels
...
Adjust overview points to pixel centers
2016-04-25 17:30:25 +02:00
Javier Goizueta
18267477da
Merge pull request #238 from CartoDB/235-column-names
...
Optimize column information functions
2016-04-25 17:30:07 +02:00
Javier Goizueta
11ad45306f
Remove unneeded pg_catalog schema name
2016-04-25 16:30:58 +02:00
Javier Goizueta
75c7ae98e4
Compute webmercator resolution with full accuracy
...
Fixes #241
2016-04-25 14:02:26 +02:00
Javier Goizueta
3c12cf629f
Optimize overview pixel adjustment for integer-pixel cells
2016-04-25 13:53:59 +02:00
Javier Goizueta
7b2100b51e
Adjust overview coordinates to pixel centers
...
This makes the adjustment for all grid sizes, not only
for integral number of pixels.
2016-04-25 13:33:43 +02:00
Javier Goizueta
580ec38ab8
Adjust overview clustered point to pixel centers
...
Fixes #240
2016-04-23 15:07:06 +02:00
Javier Goizueta
65415bb335
Optimize funcion CDB_COlumnType
2016-04-18 19:07:33 +02:00
Javier Goizueta
06ebb27160
Optimize internal funcion _cdb_unlimited_text_column
2016-04-18 18:50:37 +02:00
Javier Goizueta
bd5ae84e90
Optimize CDB_ColumnNames
...
This implementation is about 1000 times faster
2016-04-18 18:49:58 +02:00
Raul Ochoa
de5a702510
Adds table for storing camshaft analysis nodes
2016-04-18 17:41:39 +02:00
Javier Goizueta
ef43623f77
Remove unneeded variable
2016-04-15 17:58:03 +02:00
Javier Goizueta
1b0f77aa96
Always retain single-valued aggregated texts
...
This makes columns which have the same value in a group to be aggregated
maintain that value (rather than replacing it by the multiple-value
indicator *) whatever the group value is. (Previously this happend
only for small groups)
2016-04-15 17:49:00 +02:00
