From 82c13bdde0ef447280cc299b6c48bbd4c83a7306 Mon Sep 17 00:00:00 2001
From: Nyall Dawson <nyall.dawson@gmail.com>
Date: Fri, 14 Dec 2018 12:09:29 +1000
Subject: [PATCH] [processing] Fix error when running delete duplicate
 geometries and some input features have null geometries

Fixes #20788
---
 .../algs/qgis/DeleteDuplicateGeometries.py    | 16 +++++--
 .../expected/delete_duplicates_with_nulls.gml | 48 +++++++++++++++++++
 .../expected/delete_duplicates_with_nulls.xsd | 23 +++++++++
 .../tests/testdata/qgis_algorithm_tests.yaml  | 11 +++++
 4 files changed, 94 insertions(+), 4 deletions(-)
 create mode 100644 python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.gml
 create mode 100644 python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.xsd

diff --git a/python/plugins/processing/algs/qgis/DeleteDuplicateGeometries.py b/python/plugins/processing/algs/qgis/DeleteDuplicateGeometries.py
index 122f420df2a..cc533b5cdd5 100644
--- a/python/plugins/processing/algs/qgis/DeleteDuplicateGeometries.py
+++ b/python/plugins/processing/algs/qgis/DeleteDuplicateGeometries.py
@@ -76,11 +76,16 @@ class DeleteDuplicateGeometries(QgisAlgorithm):
 
         total = 100.0 / source.featureCount() if source.featureCount() else 0
         geoms = dict()
+        null_geom_features = set()
         index = QgsSpatialIndex()
         for current, f in enumerate(features):
             if feedback.isCanceled():
                 break
 
+            if not f.hasGeometry():
+                null_geom_features.add(f.id())
+                continue
+
             geoms[f.id()] = f.geometry()
             index.addFeature(f)
 
@@ -115,17 +120,20 @@ class DeleteDuplicateGeometries(QgisAlgorithm):
             current += 1
             feedback.setProgress(int(0.80 * current * total) + 10)  # takes about 80% of time
 
-        total = 100.0 / len(unique_features) if unique_features else 1
-
         # now, fetch all the feature attributes for the unique features only
         # be super-smart and don't re-fetch geometries
-        request = QgsFeatureRequest().setFilterFids(list(unique_features.keys())).setFlags(QgsFeatureRequest.NoGeometry)
+        distinct_geoms = set(unique_features.keys())
+        output_feature_ids = distinct_geoms.union(null_geom_features)
+        total = 100.0 / len(output_feature_ids) if output_feature_ids else 1
+
+        request = QgsFeatureRequest().setFilterFids(list(output_feature_ids)).setFlags(QgsFeatureRequest.NoGeometry)
         for current, f in enumerate(source.getFeatures(request)):
             if feedback.isCanceled():
                 break
 
             # use already fetched geometry
-            f.setGeometry(unique_features[f.id()])
+            if f.id() not in null_geom_features:
+                f.setGeometry(unique_features[f.id()])
             sink.addFeature(f, QgsFeatureSink.FastInsert)
 
             feedback.setProgress(int(0.10 * current * total) + 90) # takes about 10% of time
diff --git a/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.gml b/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.gml
new file mode 100644
index 00000000000..7610daf5ed6
--- /dev/null
+++ b/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.gml
@@ -0,0 +1,48 @@
+<?xml version="1.0" encoding="utf-8" ?>
+<ogr:FeatureCollection
+     xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+     xsi:schemaLocation="http://ogr.maptools.org/ delete_duplicates_with_nulls.xsd"
+     xmlns:ogr="http://ogr.maptools.org/"
+     xmlns:gml="http://www.opengis.net/gml">
+  <gml:boundedBy>
+    <gml:Box>
+      <gml:coord><gml:X>-1</gml:X><gml:Y>-3</gml:Y></gml:coord>
+      <gml:coord><gml:X>11</gml:X><gml:Y>5</gml:Y></gml:coord>
+    </gml:Box>
+  </gml:boundedBy>
+                                                                                                                                                             
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.0">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>6,2 9,2 9,3 11,5</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.1">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>-1,-1 1,-1</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.2">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>2,0 2,2 3,2 3,3</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.3">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>3,1 5,1</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.4">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>7,-3 10,-3</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.5">
+      <ogr:geometryProperty><gml:LineString srsName="EPSG:4326"><gml:coordinates>6,-3 10,1</gml:coordinates></gml:LineString></ogr:geometryProperty>
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+  <gml:featureMember>
+    <ogr:delete_duplicates_with_nulls fid="lines.6">
+    </ogr:delete_duplicates_with_nulls>
+  </gml:featureMember>
+</ogr:FeatureCollection>
diff --git a/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.xsd b/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.xsd
new file mode 100644
index 00000000000..1bedff81946
--- /dev/null
+++ b/python/plugins/processing/tests/testdata/expected/delete_duplicates_with_nulls.xsd
@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xs:schema targetNamespace="http://ogr.maptools.org/" xmlns:ogr="http://ogr.maptools.org/" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:gml="http://www.opengis.net/gml" elementFormDefault="qualified" version="1.0">
+<xs:import namespace="http://www.opengis.net/gml" schemaLocation="http://schemas.opengis.net/gml/2.1.2/feature.xsd"/>
+<xs:element name="FeatureCollection" type="ogr:FeatureCollectionType" substitutionGroup="gml:_FeatureCollection"/>
+<xs:complexType name="FeatureCollectionType">
+  <xs:complexContent>
+    <xs:extension base="gml:AbstractFeatureCollectionType">
+      <xs:attribute name="lockId" type="xs:string" use="optional"/>
+      <xs:attribute name="scope" type="xs:string" use="optional"/>
+    </xs:extension>
+  </xs:complexContent>
+</xs:complexType>
+<xs:element name="delete_duplicates_with_nulls" type="ogr:delete_duplicates_with_nulls_Type" substitutionGroup="gml:_Feature"/>
+<xs:complexType name="delete_duplicates_with_nulls_Type">
+  <xs:complexContent>
+    <xs:extension base="gml:AbstractFeatureType">
+      <xs:sequence>
+        <xs:element name="geometryProperty" type="gml:LineStringPropertyType" nillable="true" minOccurs="0" maxOccurs="1"/>
+      </xs:sequence>
+    </xs:extension>
+  </xs:complexContent>
+</xs:complexType>
+</xs:schema>
diff --git a/python/plugins/processing/tests/testdata/qgis_algorithm_tests.yaml b/python/plugins/processing/tests/testdata/qgis_algorithm_tests.yaml
index 1ef6a52172c..73ad4010cc9 100755
--- a/python/plugins/processing/tests/testdata/qgis_algorithm_tests.yaml
+++ b/python/plugins/processing/tests/testdata/qgis_algorithm_tests.yaml
@@ -6396,5 +6396,16 @@ tests:
         name: expected/split_linez_by_length.shp
         type: vector
 
+  - algorithm: qgis:deleteduplicategeometries
+    name: Delete Duplicates with null geometries
+    params:
+      INPUT:
+        name: lines.gml|layername=lines
+        type: vector
+    results:
+      OUTPUT:
+        name: expected/delete_duplicates_with_nulls.gml
+        type: vector
+
 
 # See ../README.md for a description of the file format