From 6e50e43e1c2ad3fa2c4bf7923386dcba614d78a5 Mon Sep 17 00:00:00 2001
From: Andy Eschbacher <andy.eschbacher@gmail.com>
Date: Mon, 28 Nov 2016 11:07:23 -0500
Subject: [PATCH] add query for gwr

---
 .../crankshaft/pysal_utils/pysal_utils.py     | 59 +++++++++++++++----
 .../crankshaft/regression/gwr/gwr.py          |  8 +--
 2 files changed, 51 insertions(+), 16 deletions(-)

diff --git a/src/py/crankshaft/crankshaft/pysal_utils/pysal_utils.py b/src/py/crankshaft/crankshaft/pysal_utils/pysal_utils.py
index 6dbcbd8..db75703 100644
--- a/src/py/crankshaft/crankshaft/pysal_utils/pysal_utils.py
+++ b/src/py/crankshaft/crankshaft/pysal_utils/pysal_utils.py
@@ -42,7 +42,7 @@ def get_weight(query_res, w_type='knn', num_ngbrs=5):
     return built_weight
 
 
-def query_attr_select(params):
+def query_attr_select(params, table_ref="i"):
     """
         Create portion of SELECT statement for attributes inolved in query.
         @param params: dict of information used in query (column names,
@@ -50,11 +50,15 @@ def query_attr_select(params):
     """
 
     attr_string = ""
-    template = "i.\"%(col)s\"::numeric As attr%(alias_num)s, "
+    template = "\"%(col)s\"::numeric As attr%(alias_num)s, "
 
-    if 'time_cols' in params:
-        # if markov analysis
-        attrs = params['time_cols']
+    if table_ref is not None:
+        template = "{table_ref}.".format(table_ref=table_ref) + template
+
+    if ('time_cols' in params) or ('ind_vars' in params):
+        # if markov or gwr analysis
+        attrs = (params['time_cols'] if 'time_cols' in params
+                 else params['ind_vars'])
 
         for idx, val in enumerate(attrs):
             attr_string += template % {"col": val, "alias_num": idx + 1}
@@ -71,7 +75,7 @@ def query_attr_select(params):
     return attr_string
 
 
-def query_attr_where(params):
+def query_attr_where(params, table_ref=None):
     """
       Construct where conditions when building neighbors query
         Create portion of WHERE clauses for weeding out NULL-valued geometries
@@ -90,11 +94,14 @@ def query_attr_where(params):
           NULL AND idx_replace."time3" IS NOT NULL'
     """
     attr_string = []
-    template = "idx_replace.\"%s\" IS NOT NULL"
+    template = "\"%s\" IS NOT NULL"
+    if table_ref is not None:
+        template = "idx_replace." + template
 
-    if 'time_cols' in params:
-        # markov where clauses
-        attrs = params['time_cols']
+    if ('time_cols' in params) or ('ind_vars' in params):
+        # markov or gwr where clauses
+        attrs = (params['time_cols'] if 'time_cols' in params
+                 else params['ind_vars'])
         # add values to template
         for attr in attrs:
             attr_string.append(template % attr)
@@ -104,13 +111,16 @@ def query_attr_where(params):
         # get keys
         attrs = sorted([k for k in params
                         if k not in ('id_col', 'geom_col', 'subquery',
-                                     'num_ngbrs', 'subquery')])
+                                     'num_ngbrs', 'ind_vars')])
         # add values to template
         for attr in attrs:
             attr_string.append(template % params[attr])
 
         if len(attrs) == 2:
-            attr_string.append("idx_replace.\"%s\" <> 0" % params[attrs[1]])
+            check_zero = "\"%s\" <> 0" % params[attrs[1]]
+            if table_ref is not None:
+                check_zero = "idx_replace." + check_zero
+            attr_string.append(check_zero)
 
     out = " AND ".join(attr_string)
 
@@ -177,6 +187,31 @@ def queen(params):
 
     return query.format(**params)
 
+def gwr_query(params):
+    """
+    GWR query
+    """
+
+    replacements = {"ind_vars_select": query_attr_select(params,
+                                                         table_ref=None),
+                    "ind_vars_where": query_attr_where(params,
+                                                       table_ref=None)}
+
+    query = '''
+      SELECT
+        array_agg(ST_X({geom_col})) As x,
+        array_agg(ST_Y({geom_col})) As y,
+        array_agg({dep_var}) As dep_var,
+        %(ind_vars_select)s,
+        array_agg({id_col}) As rowid
+      FROM ({subquery}) As q
+      WHERE
+        {dep_var} IS NOT NULL AND
+        %(ind_vars_where)s
+        ''' % replacements
+
+    return query.format(**params).strip()
+
 # to add more weight methods open a ticket or pull request
 
 
diff --git a/src/py/crankshaft/crankshaft/regression/gwr/gwr.py b/src/py/crankshaft/crankshaft/regression/gwr/gwr.py
index 155b95b..9cf6fa8 100644
--- a/src/py/crankshaft/crankshaft/regression/gwr/gwr.py
+++ b/src/py/crankshaft/crankshaft/regression/gwr/gwr.py
@@ -16,14 +16,14 @@ def gwr(subquery, dep_var, ind_vars, fixed=False, kernel='bisquare'):
 
     x = np.array(query_result[0]['x'])
     y = np.array(query_result[0]['y'])
-    coords = zip(x,y)
-        
-    Y = query_result[0]['dep'].reshape((-1,1))
+    coords = zip(x, y)
+
+    Y = query_result[0]['dep'].reshape((-1, 1))
     n = Y.shape[0]
     k = len(ind_vars)
     X = np.zeros((n, k))
 
-    for attr in range(0,k):
+    for attr in range(0, k):
         attr_name = 'attr' + str(attr+1)
         X[:, attr] = np.array(query_result[0][attr_name]).flatten()