version of Maxime Laug and Charlie Pauvré

oracle-of-bacon-backend/.project

@@ -0,0 +1,28 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+	<name>oracle-of-bacon-backend</name>
+	<comment>Project oracle-of-bacon-backend created by Buildship.</comment>
+	<projects>
+	</projects>
+	<buildSpec>
+		<buildCommand>
+			<name>org.eclipse.buildship.core.gradleprojectbuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
+	</buildSpec>
+	<natures>
+		<nature>org.eclipse.buildship.core.gradleprojectnature</nature>
+	</natures>
+	<filteredResources>
+		<filter>
+			<id>1616599270626</id>
+			<name></name>
+			<type>30</type>
+			<matcher>
+				<id>org.eclipse.core.resources.regexFilterMatcher</id>
+				<arguments>node_modules|.git|__CREATED_BY_JAVA_LANGUAGE_SERVER__</arguments>
+			</matcher>
+		</filter>
+	</filteredResources>
+</projectDescription>

oracle-of-bacon-backend/.settings/org.eclipse.buildship.core.prefs

@@ -0,0 +1,13 @@
+arguments=
+auto.sync=false
+build.scans.enabled=false
+connection.gradle.distribution=GRADLE_DISTRIBUTION(WRAPPER)
+connection.project.dir=
+eclipse.preferences.version=1
+gradle.user.home=
+java.home=C\:/Program Files/Java/jdk-11.0.10
+jvm.arguments=
+offline.mode=false
+override.workspace.settings=true
+show.console.view=true
+show.executions.view=true

script/insert.js

@@ -2,22 +2,37 @@ const csv = require("csv-parser");
 const fs = require("fs");
 const { Client } = require("@elastic/elasticsearch");
 
-const client = new Client({ node: "http://localhost:9200" });
 
 async function insert() {
-  // TODO créer l'index (et plus pour la ré-exécution ?)
+
+
+const client = new Client({ node: "http://localhost:9200" });
+
+  await client.indices.delete({
+    index: 'bacon',
+    ignore_unavailable: true
+  });
+
+  await client.indices.create({ index: 'bacon' });
 
   let actors = [];
   let first = true;
   fs.createReadStream("./imdb-data/actors.csv")
     .pipe(csv())
     // Pour chaque ligne on créé un document JSON pour l'acteur correspondant
-    .on("data", async ({ name }) => {
-      // TODO ajouter les acteurs au tableau
+    .on("data", async (data) => {
+      //console.log(data['name:ID'])
+      actors.push({
+        name: data['name:ID']
+      });
     })
     // A la fin on créé l'ensemble des acteurs dans ElasticSearch
     .on("end", () => {
-      // TODO insérer dans elastic (les fonctions ci-dessous peuvent vous aider)
+      client.bulk(createBulkInsertQueries(actors.map(actor => actor.name), 10000), (err, resp) => {
+        if (err) console.trace(err.message);
+        else console.log(`Inserted ${resp.body.items.length} actors`);
+        client.close();
+      });
     });
 }
 
@@ -28,29 +43,35 @@ function recBulk(client, bulks) {
   }
 
   const first = bulks.pop();
-  return client.bulk(first).then((r) => {
+  return client.bulk(first).then(() => {
     console.log("Done inserting " + first.body.length / 2);
     return recBulk(client, bulks);
   });
 }
 
 function createBulkInsertQueries(names, length) {
-  // TODO
+  const arrays = [];
+  while (names.length > 0) {
+    arrays.push(names.splice(0, length));
+  }
+
+  return arrays.map((arr) => createBulkInsertQuery(arr));
 }
 
 // Fonction utilitaire permettant de formatter les données pour l'insertion "bulk" dans elastic
 function createBulkInsertQuery(names) {
+  //console.log(names);
   const body = names.reduce((acc, name) => {
-    const suggest = [name];
-    const parts = name.split(",");
-    parts.forEach((part) => suggest.push(part.trim()));
-    parts.reverse().forEach((part) => suggest.push(part.trim()));
+    // const suggest = [name];
+    // const parts = name.split(",");
+    // parts.forEach((part) => suggest.push(part.trim()));
+    // parts.reverse().forEach((part) => suggest.push(part.trim()));
 
-    acc.push({ index: { _index: "actor", _type: "_doc" } });
-    acc.push({ name, suggest });
+    acc.push({ index: { _index: "bacon" } });
+    acc.push({ name });
     return acc;
   }, []);
-
+  //console.log(body);
   return { body };
 }