Merge pull request #131 from CartoDB/develop

Release 0.4.1

doc/08_interpolation.md

@@ -2,7 +2,7 @@
 
 Function to interpolate a numeric attribute of a point in a scatter dataset of points, using one of three methos:
 
-* [Nearest neighbor](https://en.wikipedia.org/wiki/Nearest-neighbor_interpolation)
+* [Nearest neighbor(s)](https://en.wikipedia.org/wiki/Nearest-neighbor_interpolation)
 * [Barycentric](https://en.wikipedia.org/wiki/Barycentric_coordinate_system)
 * [IDW](https://en.wikipedia.org/wiki/Inverse_distance_weighting)
 
@@ -15,7 +15,7 @@ Function to interpolate a numeric attribute of a point in a scatter dataset of p
 | query   | text | query that returns at least `the_geom` and a numeric value as `attrib` |
 | point   | geometry | The target point to calc the value |
 | method   | integer     | 0:nearest neighbor, 1: barycentric, 2: IDW|
-| p1   | integer     | IDW: limit the number of neighbors, 0->no limit|
+| p1   | integer     | limit the number of neighbors, IDW: 0->no limit, NN: 0-> closest one|
 | p2   | integer     | IDW: order of distance decay, 0-> order 1|
 
 ### CDB_SpatialInterpolation (geom geometry[], values numeric[], point geometry, method integer DEFAULT 1, p1 integer DEFAULT 0, ps integer DEFAULT 0)
@@ -28,7 +28,7 @@ Function to interpolate a numeric attribute of a point in a scatter dataset of p
 | values | numeric[]   | Array of points' values for the param under study|
 | point   | geometry | The target point to calc the value |
 | method   | integer     | 0:nearest neighbor, 1: barycentric, 2: IDW|
-| p1   | integer     | IDW: limit the number of neighbors, 0->no limit|
+| p1   | integer     | limit the number of neighbors, IDW: 0->no limit, NN: 0-> closest one|
 | p2   | integer     | IDW: order of distance decay, 0-> order 1|
 
 ### Returns
@@ -37,6 +37,9 @@ Function to interpolate a numeric attribute of a point in a scatter dataset of p
 |-------------|------|-------------|
 | value  | numeric | Interpolated value at the given point, `-888.888` if the given point is out of the boundaries of the source points set |
 
+Default values:
+* -888.888: when using Barycentric, the target point is out of the realm of the input points
+* -777.777: asking for a method not available
 
 #### Example Usage
 

doc/19_contour.md

@@ -18,7 +18,7 @@ Function to generate a contour map from an scatter dataset of points, using one
 | method   | integer     | 0:nearest neighbor, 1: barycentric, 2: IDW|
 | classmethod   | integer     | 0:equals, 1: heads&tails, 2:jenks, 3:quantiles |
 | steps   | integer     | Number of steps in the classification|
-| max_time   | integer     | Max time in milliseconds for processing time
+| max_time   | integer     | if <= 0: max processing time in seconds (smart resolution) , if >0: resolution in meters
 
 ### Returns
 Returns a table object

src/pg/sql/08_interpolation.sql

@@ -1,6 +1,8 @@
--- 0: nearest neighbor
+-- 0: nearest neighbor(s)
 -- 1: barymetric
 -- 2: IDW
+-- 3: krigin ---> TO DO
+
 
 CREATE OR REPLACE FUNCTION CDB_SpatialInterpolation(
     IN query text,
@@ -50,12 +52,19 @@ DECLARE
     vc numeric;
     output numeric;
 BEGIN
-    output :=  -999.999;
+    -- output :=  -999.999;
-    -- nearest
+
+    -- nearest neighbors
+    -- p1: limit the number of neighbors, 0-> closest one
     IF method = 0 THEN
 
-        WITH    a as (SELECT unnest(geomin) as g, unnest(colin) as v)
+        IF p1 = 0 THEN
-        SELECT a.v INTO output FROM a ORDER BY point<->a.g LIMIT 1;
+            p1 := 1;
+        END IF;
+
+        WITH    a as (SELECT unnest(geomin) as g, unnest(colin) as v),
+                b as (SELECT a.v as v FROM a ORDER BY point<->a.g LIMIT p1::integer)
+        SELECT avg(b.v) INTO output FROM b;
         RETURN output;
 
     -- barymetric
@@ -121,6 +130,11 @@ BEGIN
         SELECT sum(b.f)/sum(b.k) INTO output FROM b;
         RETURN output;
 
+    -- krigin
+    ELSIF method = 3 THEN
+
+    --  TO DO
+
     END IF;
 
     RETURN -777.777;

src/pg/sql/19_contour.sql

@@ -17,16 +17,15 @@ RETURNS TABLE(
 DECLARE
     cell_count integer;
     tin geometry[];
+    resolution integer;
 BEGIN
-    -- calc the cell size in web mercator units
+
-    -- WITH center as (
+    -- nasty trick to override issue #121
-    --     SELECT ST_centroid(ST_Collect(geomin)) as c
+    IF max_time = 0 THEN
-    -- )
+        max_time = -90;
-    -- SELECT
+    END IF;
-    --     round(resolution / cos(ST_y(c) * pi()/180))
+    resolution := max_time;
-    --     INTO cell
+    max_time := -1 * resolution;
-    -- FROM center;
-    -- raise notice 'Resol: %', cell;
 
     -- calc the optimal number of cells for the current dataset
     SELECT
@@ -70,9 +69,13 @@ BEGIN
     ),
     resolution as(
         SELECT
-             round(|/ (
+            CASE WHEN resolution <= 0  THEN
-                 ST_area(geom) / cell_count
+                round(|/ (
-             )) as cell
+                 ST_area(geom) / abs(cell_count)
+             ))
+            ELSE
+                resolution
+            END AS cell
         FROM envelope3857
     ),
     grid as(

src/pg/test/expected/08_interpolation_test.out

@@ -1,7 +1,7 @@
 SET client_min_messages TO WARNING;
 \set ECHO none
- nn  |           nni            |       idw       
+          nn          |           nni            |       idw       
------+--------------------------+-----------------
+----------------------+--------------------------+-----------------
- 200 | 238.41059602632179224595 | 341.46260750526
+ 200.0000000000000000 | 238.41059602632179224595 | 341.46260750526
 (1 row)
 

src/pg/test/sql/19_contour_test.sql

@@ -12,6 +12,6 @@ SELECT
     foo.*
 FROM
     a,
-    cdb_crankshaft.CDB_contour(a.g, a.vals,  0.0, 1, 3, 5, 60) foo
+    cdb_crankshaft.CDB_contour(a.g, a.vals,  0.0, 1, 3, 5, -60) foo
 )
 SELECT bin, avg_value from b order by bin;

src/py/crankshaft/crankshaft/clustering/moran.py

@@ -30,18 +30,13 @@ def moran(subquery, attr_name,
 
     query = pu.construct_neighbor_query(w_type, qvals)
 
-    plpy.notice('** Query: %s' % query)
-
     try:
         result = plpy.execute(query)
         # if there are no neighbors, exit
         if len(result) == 0:
             return pu.empty_zipped_array(2)
-        plpy.notice('** Query returned with %d rows' % len(result))
+    except plpy.SPIError, e:
-    except plpy.SPIError:
+        plpy.error('Analysis failed: %s' % e)
-        plpy.error('Error: areas of interest query failed, check input parameters')
-        plpy.notice('** Query failed: "%s"' % query)
-        plpy.notice('** Error: %s' % plpy.SPIError)
         return pu.empty_zipped_array(2)
 
     ## collect attributes
@@ -79,9 +74,8 @@ def moran_local(subquery, attr,
         # if there are no neighbors, exit
         if len(result) == 0:
             return pu.empty_zipped_array(5)
-    except plpy.SPIError:
+    except plpy.SPIError, e:
-        plpy.error('Error: areas of interest query failed, check input parameters')
+        plpy.error('Analysis failed: %s' % e)
-        plpy.notice('** Query failed: "%s"' % query)
         return pu.empty_zipped_array(5)
 
     attr_vals = pu.get_attributes(result)
@@ -111,18 +105,13 @@ def moran_rate(subquery, numerator, denominator,
 
     query = pu.construct_neighbor_query(w_type, qvals)
 
-    plpy.notice('** Query: %s' % query)
-
     try:
         result = plpy.execute(query)
         # if there are no neighbors, exit
         if len(result) == 0:
             return pu.empty_zipped_array(2)
-        plpy.notice('** Query returned with %d rows' % len(result))
+    except plpy.SPIError, e:
-    except plpy.SPIError:
+        plpy.error('Analysis failed: %s' % e)
-        plpy.error('Error: areas of interest query failed, check input parameters')
-        plpy.notice('** Query failed: "%s"' % query)
-        plpy.notice('** Error: %s' % plpy.SPIError)
         return pu.empty_zipped_array(2)
 
     ## collect attributes
@@ -160,10 +149,8 @@ def moran_local_rate(subquery, numerator, denominator,
         # if there are no neighbors, exit
         if len(result) == 0:
             return pu.empty_zipped_array(5)
-    except plpy.SPIError:
+    except plpy.SPIError, e:
-        plpy.error('Error: areas of interest query failed, check input parameters')
+        plpy.error('Analysis failed: %s' % e)
-        plpy.notice('** Query failed: "%s"' % query)
-        plpy.notice('** Error: %s' % plpy.SPIError)
         return pu.empty_zipped_array(5)
 
     ## collect attributes
@@ -186,7 +173,6 @@ def moran_local_bv(subquery, attr1, attr2,
     """
         Moran's I (local) Bivariate (untested)
     """
-    plpy.notice('** Constructing query')
 
     qvals = OrderedDict([("id_col", id_col),
                          ("attr1", attr1),
@@ -205,7 +191,6 @@ def moran_local_bv(subquery, attr1, attr2,
     except plpy.SPIError:
         plpy.error("Error: areas of interest query failed, " \
                    "check input parameters")
-        plpy.notice('** Query failed: "%s"' % query)
         return pu.empty_zipped_array(4)
 
     ## collect attributes
@@ -219,13 +204,9 @@ def moran_local_bv(subquery, attr1, attr2,
     lisa = ps.esda.moran.Moran_Local_BV(attr1_vals, attr2_vals, weight,
                                         permutations=permutations)
 
-    plpy.notice("len of Is: %d" % len(lisa.Is))
-
     # find clustering of significance
     lisa_sig = quad_position(lisa.q)
 
-    plpy.notice('** Finished calculations')
-
     return zip(lisa.Is, lisa_sig, lisa.p_sim, weight.id_order)
 
 # Low level functions ----------------------------------------

src/py/crankshaft/crankshaft/space_time_dynamics/markov.py

@@ -56,9 +56,9 @@ def spatial_markov_trend(subquery, time_cols, num_classes=7,
         )
         if len(query_result) == 0:
             return zip([None], [None], [None], [None], [None])
-    except plpy.SPIError, err:
+    except plpy.SPIError, e:
         plpy.debug('Query failed with exception %s: %s' % (err, pu.construct_neighbor_query(w_type, qvals)))
-        plpy.error('Query failed, check the input parameters')
+        plpy.error('Analysis failed: %s' % e)
         return zip([None], [None], [None], [None], [None])
 
     ## build weight