Add basic blog-recommendation sample app

9 files changed, 408 insertions, 0 deletions

diff --git a/sample-apps/blog-recommendation/README.md b/sample-apps/blog-recommendation/README.md
new file mode 100644
index 00000000000..0b6a00c9743
--- /dev/null
+++ b/sample-apps/blog-recommendation/README.md
@@ -0,0 +1,55 @@
+Basic Search Application
+==================
+Start by [deploying a sample application](http://vespa.corp.yahoo.com/6/documentation/developing-with-vespa.html).
+
+### Find the endpoint
+
+When you have successfully deployed your own compiled version of the application above, you need to find the name of the "endpoint".
+The endpoint is used for feeding and searching for data.
+**Please allow a few minutes for the endpoint to appear after deployment**
+You can find this endpoint by doing:
+ ```sh
+
+ mvn vespa:endpoints | grep Endpoints
+ ```
+
+You can also find it by looking at the [Hosted Vespa Dashboard](http://dashboard.vespa.corp.yahoo.com).
+
+
+### Feed and search
+ 1. **Feed** the data that is to be searched
+ ```sh
+
+ # Feeding two documents
+ curl -X POST --data-binary  @music-data-1.json <endpoint url>/document/v1/music/music/docid/1 | python -m json.tool
+ curl -X POST --data-binary  @music-data-2.json <endpoint url>/document/v1/music/music/docid/2 | python -m json.tool
+
+  ```
+
+For feeding many documents fast and reliable, checkout [feeding example](https://git.corp.yahoo.com/vespa-samples/basic-feeding-client)
+
+ 2. **Visit documents
+
+ Since we do not have many documents we can list them all
+ ```sh
+
+ # All documents
+ curl <endpoint url>/document/v1/music/music/docid | python -m json.tool
+
+ # Document with id 1
+ curl <endpoint url>/document/v1/music/music/docid/1 | python -m json.tool
+
+  ```
+
+ 3. **Search**
+ We can also search for documents:
+    ```sh
+
+    curl '<endpoint url>/search/?query=bad' | python -m json.tool
+
+
+    ```
+
+### Next step: from development to production
+See [continuous deployments](http://vespa.corp.yahoo.com/6/documentation/continuous-deployment.html) for how to implement continuous deployments for production.
+See [RESTified Document Operation API](http://vespa.corp.yahoo.com/6/documentation/document_api_v1.html) for documentation about the REST API for document operations.
diff --git a/sample-apps/blog-recommendation/pom.xml b/sample-apps/blog-recommendation/pom.xml
new file mode 100644
index 00000000000..6e55e109f43
--- /dev/null
+++ b/sample-apps/blog-recommendation/pom.xml
@@ -0,0 +1,85 @@
+<?xml version="1.0"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0                              http://maven.apache.org/xsd/maven-4.0.0.xsd">
+  <modelVersion>4.0.0</modelVersion>
+  <groupId>com.yahoo.example</groupId>
+  <artifactId>recommendation</artifactId>
+  <packaging>container-plugin</packaging>
+  <version>0.0.1</version>
+  <name>application</name>
+
+  <properties>
+    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
+    <test.hide>true</test.hide>
+    <application>blog-recommendation</application>
+    <instance>default</instance>
+ </properties>
+
+  <parent>
+    <groupId>com.yahoo.vespa.tenant</groupId>
+    <artifactId>base</artifactId>
+    <version>RELEASE</version>
+  </parent>
+
+  <pluginRepositories>
+    <pluginRepository>
+      <id>ymaven</id>
+      <url>http://ymaven.corp.yahoo.com:9999/proximity/repository/public</url>
+     <snapshots>
+        <enabled>false</enabled>
+      </snapshots>
+    </pluginRepository>
+    <pluginRepository>
+      <id>maven2-repository.dev.java.net</id>
+      <url>http://download.java.net/maven/2</url>
+      <snapshots>
+        <enabled>false</enabled>
+      </snapshots>
+    </pluginRepository>
+    <pluginRepository>
+        <id>vespa-maven-release</id>
+        <name>vespa-maven-release</name>
+      <url>http://edge.artifactory.yahoo.com:8000/artifactory/vespa-maven-libs-release-local</url>
+      <releases>
+        <enabled>true</enabled>
+      </releases>
+      <snapshots>
+        <enabled>false</enabled>
+      </snapshots>
+    </pluginRepository>
+  </pluginRepositories>
+  <repositories>
+    <repository>
+      <id>ymaven</id>
+      <url>http://ymaven.corp.yahoo.com:9999/proximity/repository/public</url>
+      <snapshots>
+        <enabled>false</enabled>
+      </snapshots>
+    </repository>
+    <repository>
+      <id>vespa-maven-release</id>
+      <name>vespa-maven-release</name>
+      <url>http://edge.artifactory.yahoo.com:8000/artifactory/vespa-maven-libs-release-local</url>
+      <releases>
+        <enabled>true</enabled>
+      </releases>
+      <snapshots>
+        <enabled>false</enabled>
+      </snapshots>
+    </repository>
+  </repositories>
+
+  <build>
+    <plugins>
+      <plugin>
+        <groupId>org.apache.maven.plugins</groupId>
+        <artifactId>maven-compiler-plugin</artifactId>
+        <configuration>
+          <compilerArgs>
+            <arg>-Xlint:all</arg>
+            <arg>-Werror</arg>
+          </compilerArgs>
+        </configuration>
+      </plugin>
+    </plugins>
+  </build>
+</project>
diff --git a/sample-apps/blog-recommendation/src/main/application/deployment.xml b/sample-apps/blog-recommendation/src/main/application/deployment.xml
new file mode 100644
index 00000000000..0a2397bbca9
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/application/deployment.xml
@@ -0,0 +1,5 @@
+<deployment version='1.0'>
+  <test />
+  <staging />
+  <prod/>
+</deployment>
diff --git a/sample-apps/blog-recommendation/src/main/application/search/query-profiles/types/root.xml b/sample-apps/blog-recommendation/src/main/application/search/query-profiles/types/root.xml
new file mode 100644
index 00000000000..6362bbe48ce
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/application/search/query-profiles/types/root.xml
@@ -0,0 +1,3 @@
+<query-profile-type id="root" inherits="native">
+    <field name="ranking.features.query(user_item_cf)" type="tensor(user_item_cf[10])" />
+</query-profile-type>
diff --git a/sample-apps/blog-recommendation/src/main/application/searchdefinitions/blog_post.sd b/sample-apps/blog-recommendation/src/main/application/searchdefinitions/blog_post.sd
new file mode 100644
index 00000000000..1b92822e425
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/application/searchdefinitions/blog_post.sd
@@ -0,0 +1,66 @@
+search blog_post {
+    document blog_post {
+        field date_gmt type string {
+            indexing: summary
+        }
+        field language type string {
+            indexing: summary
+        }
+
+        field author type string {
+            indexing: summary
+        }
+
+        field url type string {
+            indexing: summary
+        }
+
+        field title type string {
+            indexing: summary | index
+        }
+
+	    field blog type string {
+	        indexing: summary
+	    }
+
+	    field post_id type string {
+	        indexing: summary
+	    }
+
+        field tags type array<string> {
+            indexing: summary
+        }
+
+        field blogname type string {
+            indexing: summary
+        }
+
+        field content type string {
+            indexing: summary | index
+        }
+
+        field categories type array<string> {
+            indexing: summary | index
+        }
+
+        field user_item_cf type tensor {
+            indexing: summary | attribute
+            attribute: tensor(user_item_cf[10])
+        }
+
+        field has_user_item_cf type byte {
+            indexing: summary | attribute
+            attribute: fast-search
+        }
+
+    }
+
+    rank-profile tensor {
+        first-phase {
+            expression {
+                sum(query(user_item_cf) * attribute(user_item_cf))
+            }
+        }
+    }
+
+}
diff --git a/sample-apps/blog-recommendation/src/main/application/searchdefinitions/user.sd b/sample-apps/blog-recommendation/src/main/application/searchdefinitions/user.sd
new file mode 100644
index 00000000000..0c2a6e50aa6
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/application/searchdefinitions/user.sd
@@ -0,0 +1,26 @@
+search user {
+    document user {
+
+        field user_id type string {
+            indexing: summary | attribute
+            attribute: fast-search
+        }
+
+        field has_read_items type array<string> {
+            indexing: summary | attribute
+        }
+
+        # see http://vespa.corp.yahoo.com/6/documentation/reference/search-definitions.html#tensor-type-spec
+        field user_item_cf type tensor {
+            indexing: summary | attribute
+            attribute: tensor(user_item_cf[10])
+        }
+
+        field has_user_item_cf type byte {
+            indexing: summary | attribute
+            attribute: fast-search
+        }
+
+    }
+
+}
diff --git a/sample-apps/blog-recommendation/src/main/application/services.xml b/sample-apps/blog-recommendation/src/main/application/services.xml
new file mode 100644
index 00000000000..d1650eaca8c
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/application/services.xml
@@ -0,0 +1,36 @@
+<?xml version='1.0' encoding='UTF-8'?>
+<services version='1.0' xmlns:deploy="vespa" xmlns:preprocess="properties">
+
+  <jdisc id='default' version='1.0'>
+
+    <search>
+      <chain id='blog' inherits='vespa'>
+        <searcher bundle='recommendation' id='com.yahoo.example.BlogTensorSearcher' />
+      </chain>
+      <chain id='user' inherits='vespa'>
+        <searcher bundle='recommendation' id='com.yahoo.example.UserProfileSearcher' />
+      </chain>
+      <chain id='default' inherits='vespa'>
+        <searcher bundle='recommendation' id='com.yahoo.example.UserProfileSearcher' />
+        <searcher bundle='recommendation' id='com.yahoo.example.BlogTensorSearcher' />
+      </chain>
+    </search>
+    <document-api/>
+    <nodes count='1'/>
+  </jdisc>
+
+  <content id='content' version='1.0'>
+    <redundancy>1</redundancy>
+    <documents>
+      <document mode='index' type='blog_post'/>
+      <document mode='index' type='user'/>
+    </documents>
+    <nodes count='1'/>
+    <engine>
+      <proton>
+        <searchable-copies>1</searchable-copies>
+      </proton>
+    </engine>
+  </content>
+
+</services>
diff --git a/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/BlogTensorSearcher.java b/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/BlogTensorSearcher.java
new file mode 100644
index 00000000000..f9ff27c4813
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/BlogTensorSearcher.java
@@ -0,0 +1,43 @@
+package com.yahoo.example;
+
+import com.yahoo.prelude.query.IntItem;
+import com.yahoo.processing.request.CompoundName;
+import com.yahoo.search.Query;
+import com.yahoo.search.Result;
+import com.yahoo.search.Searcher;
+import com.yahoo.search.querytransform.QueryTreeUtil;
+import com.yahoo.search.searchchain.Execution;
+import com.yahoo.tensor.Tensor;
+
+public class BlogTensorSearcher extends Searcher {
+
+    @Override
+    public Result search(Query query, Execution execution) {
+
+        Object userItemCfProperty = query.properties().get("user_item_cf");
+        if (userItemCfProperty != null) {
+
+            // Modify the query by restricting to blog_posts...
+            query.getModel().setRestrict("blog_post");
+
+            // ... that has a tensor field fed
+            QueryTreeUtil.andQueryItemWithRoot(query, new IntItem(1, "has_user_item_cf"));
+
+            // Modify the ranking by using the 'tensor' rank-profile (as defined in blog_post.sd)...
+            query.properties().set(new CompoundName("ranking"), "tensor");
+
+            // ... and setting 'query(user_item_cf)' used in that rank-profile
+            query.getRanking().getFeatures().put("query(user_item_cf)", toTensor(userItemCfProperty));
+        }
+
+        return execution.search(query);
+    }
+
+    private Tensor toTensor(Object tensor) {
+        if (tensor instanceof Tensor) {
+            return (Tensor) tensor;
+        }
+        return Tensor.from(tensor.toString());
+    }
+
+}
diff --git a/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/UserProfileSearcher.java b/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/UserProfileSearcher.java
new file mode 100644
index 00000000000..63bf4110e67
--- /dev/null
+++ b/sample-apps/blog-recommendation/src/main/java/com/yahoo/example/UserProfileSearcher.java
@@ -0,0 +1,89 @@
+package com.yahoo.example;
+
+import com.yahoo.data.access.Inspectable;
+import com.yahoo.data.access.Inspector;
+import com.yahoo.prelude.query.WordItem;
+import com.yahoo.processing.request.CompoundName;
+import com.yahoo.search.Query;
+import com.yahoo.search.Result;
+import com.yahoo.search.Searcher;
+import com.yahoo.search.result.Hit;
+import com.yahoo.search.searchchain.Execution;
+import com.yahoo.search.searchchain.SearchChain;
+import com.yahoo.tensor.MapTensorBuilder;
+import com.yahoo.tensor.Tensor;
+
+import java.util.Iterator;
+import java.util.Map;
+
+public class UserProfileSearcher extends Searcher {
+
+    public Result search(Query query, Execution execution) {
+
+        Object userIdProperty = query.properties().get("user_id");
+        if (userIdProperty != null) {
+
+            // Retrieve user profile...
+            Tensor userProfile = retrieveUserProfile(userIdProperty.toString(), execution);
+
+            // ... and add user profile to query properties so BlogTensorSearcher can pick it up
+            query.properties().set(new CompoundName("user_item_cf"), userProfile);
+
+            if (query.isTraceable(9)) {
+                String tensorRepresentation = userProfile != null ? userProfile.toString() : "";
+                query.trace("Setting user profile to :" + tensorRepresentation, 9);
+            }
+        }
+
+        return execution.search(query);
+    }
+
+
+    private Tensor retrieveUserProfile(String userId, Execution execution) {
+        Query query = new Query();
+        query.getModel().setRestrict("user");
+        query.getModel().getQueryTree().setRoot(new WordItem(userId, "user_id"));
+        query.setHits(1);
+
+        SearchChain vespaChain = execution.searchChainRegistry().getComponent("vespa");
+        Result result = new Execution(vespaChain, execution.context()).search(query);
+
+        // This is needed to get the actual summary data
+        execution.fill(result);
+
+        Hit hit = getFirstHit(result);
+        if (hit != null) {
+            Object userItemCf = hit.getField("user_item_cf");
+            if (userItemCf instanceof Inspectable) {
+                return convertTensor((Inspectable) userItemCf);
+            }
+        }
+        return null;
+    }
+
+    private Hit getFirstHit(Result result) {
+        Iterator<Hit> hiterator = result.hits().deepIterator();
+        return hiterator.hasNext() ? hiterator.next() : null;
+    }
+
+    private Tensor convertTensor(Inspectable field) {
+        MapTensorBuilder tensorBuilder = new MapTensorBuilder();
+
+        Inspector cells = field.inspect().field("cells");
+        for (Inspector cell : cells.entries()) {
+            MapTensorBuilder.CellBuilder cellBuilder = tensorBuilder.cell();
+
+            Inspector address = cell.field("address");
+            for (Map.Entry<String, Inspector> entry : address.fields()) {
+                String dim = entry.getKey();
+                String label = entry.getValue().asString();
+                cellBuilder.label(dim, label);
+            }
+
+            Inspector value = cell.field("value");
+            cellBuilder.value(value.asDouble());
+        }
+        return tensorBuilder.build();
+    }
+
+}