paprika-Project

You can access to the site on the: https://spirals-somca.inria.lille.fr/paprika/index/

Description:

This open-source project is the Web version of Paprika. This version is cut to three parts. The native Paprika is here: https://github.com/GeoffreyHecht/paprika

The project use Docker and Maven, you need to have installed these functionality for run the project.

• paprika-analyze: Part who run the analyze with file.apk and send data to neo4J database.

• paprika-web: Part who launch the web page and launch paprika-analyze when it needed.

• paprika-tandoori: Part who run the analyze for GitHub.

Run:

For run Paprika-web, you need to compile paprika-analyze, paprika-web ad paprika-Tandoori,CodesmellsAnno per the command:

mvn clean package

Then move new jar of target/file.jar on dockerweb,dockeranalyze or dockerTandoori CodesmellsAnno.jar go only on dockerTandoori.

If you want use with GitHub: You need also to read also readme of dockerweb,analyze,tandoori For have "token", you need to create a GitHub account, then create a token who have access to repo.

When done, you need just to launch:

$./run.sh

Then you need to go on your docker-machine, for me, this is

docker-machine ssh spirals-somca.

Then install docker-compose if not installed, see https://docs.docker.com/compose/install/ For finish, just launch:

sudo docker-compose up web

The first run launch always a error you need to: go on

:7474 for create a small account. ( address is "localhost" per default) Where the username need to be neo4j(the default) and you need to put the same password than the "neo4j_pwd" put on your json.

CSAnnotations, Paprika, Tandoori:

On the folder of each. After the mvn clean package.

Install the jar:

mvn install:install-file -Dfile=./target/CodeSmellsAnnotations.jar -DpomFile=./pom.xml

mvn install:install-file -Dfile=./target/Paprika.jar -DpomFile=./pom.xml

mvn install:install-file -Dfile=./target/Tandoori.jar -DpomFile=./pom.xml

Then on your pom.xml, you can put the dependency: Maven:

<dependency>
	<groupId>codesmells.annotations</groupId>
	<artifactId>CSAnnotations</artifactId>
	<version>1</version>
</dependency>
<dependency>
	<groupId>paprika</groupId>
	<artifactId>Paprika</artifactId>
	<version>latest</version>
</dependency>
<dependency>
	<groupId>tandoori</groupId>
	<artifactId>Tandoori</artifactId>
	<version>latest</version>
</dependency>

You need made also a info.json: $ touch info.json And fill these values:

{ 
   "token":"——-",
   "token_botname":"—",
   "email_pwd":"——-",
   "email_username":"=====",
   "email_email":"[email protected]",
   "email_smtp":"smtp.gmail.com",
   "token_private_key_captcha":"--------------------------------",
   "keystore_key":"——-",
   "neo4j_pwd":"———————"
 }

{ 
   "token" <— token key of a GitHub account than you have created.

   "token_botname" <- the username of the GitHub account of the token.

   "email_pwd" <— password of the email of the bot account.

   "email_username" <- username of the email of the bot account.

   "email_email" <- E-mail of the bot account.

   "email_smtp" <- Service of the e-mail for stmp.

   "token_private_key_captcha" <- Not used, this is the private key of the Recaptcha than you have when you create a service between you and Google.
 https://www.google.com/recaptcha/

   "keystore_key" <— The key than you have put for your keystore created per:
keytool -genkeypair -keystore keystore2.jks -alias toto -keyalg RSA -keysize 2048 -dname "CN=toto"

   "neo4j_pwd" <- The password of your neo4j database.
 }

Paprika-Web use:

• Spark-core framework web.
• Neo4J graph database.(With Cypher).
• Spark Bolt for linked Java to Neo4J.
• Velocity template for page html.
• Docker
• Spoon of Inria
• Paprika of Inria

Bug:

-On the menu->Project, new versions can be inverted with the old, only if 0 versions have be analyzed on the project.

Requirement of the site:

• Paprika-analyze: ** accept only File.apk. ** Can be very long.

• Paprika-Tandoori ** require these Github Link "https://github.com/Snrasha/paprika-web.git" ** If you have two files who have the same name and the same package, theses files will no be analyzed correctly.