pbernet / TravelCompanionScala

TravelCompanion is a Web Application developed in Scala with the Lift Framework. This branch adds the solr search engine.

Welcome to TravelCompanion in Scala / Lift.

This project was created during the bachelor thesis by Ralf Muri and Daniel Hobi in spring 2010.

The following technical aspects are implemented:

• JPA (based on lift-jpa-archetype)
• User Management (copy & paste from ProtoUser without mapper functionality)
• Validating (JSR 303)
• GUI Widgets (DistanceTraveler)
• Ajax & Comet (Blog as Single Page Application)
• REST

Please feel free to download & study our document in the doc directory (german!) ###You will get some additional stuff like:

1. Best practices in Lift & HowTos
2. Working with IntelliJ IDEA & github.com
3. General information about TravelCompanion
4. Experience made by the authors

A running demo is deployed on stax.net: http://travelcompanion.ralfmuri.staxapps.net.

###To run the app locally:

1. mvn install
2. cd web
3. mvn jetty:run

Then point your favorite browser to http://localhost:9090/.

###To create an offline version of the app (web/target/TravelCompanionScala-offline-1.0/):

1. mvn install
2. cd web
3. mvn package -Pjetty-offline

Notice: TravelCompanion is intended for demo purposes only.

Update Rel. 1.1 Integration Solr 1.4.1 search engine

Case

• The Solr search engine webapp shall run on the same Jetty Instance as the TravelCompanionScala webapp
• All the nessessary configurations (path to search index etc.) should be done inside the TravelCompanionScala webapp
• The TravelCompanionScala webapp does the search requests via the Solr XML-RPC API

Realisation

• Embedded Solr webapp in TravelCompanionScala module web
• New Dir /solr/home in TravelCompanionScala for for Solr Search-Index and configuration
• Enhanced pom.xml of module web: new contextHandler and systemProperty
• Basic Search interface created
• New and updated Stages are updated in the index and deleted stages are removed from the index

Issues

1. Because there are now two "memory-hungry" webapps, set your jetty run config in your IDE to: -XX:MaxPermSize=256m -Xms500m -Xmx500m

2. Use JRE 1.6 or higher to run the webapps

3. The class TestTravelGenerator.scala generates testdata for the DB and the Solr Search-Index It's important that the DB and the Search-Index have the same Data - if the data is inconsistent:

   • Delete DB: [UserHome]
   • Delete Solr Index: [ProjectHome]\solr\home\data
   • Jetty Run
   • Run TestTravelGenerator.scala: the tours and stages are added
   • Jetty restart: needs to be done, because the Test-Class uses another "JPA Session"

4. Note that this integration is not using the more common (and probably faster) solrj lib but uses the scala databinder lib instead.