Code Visualizer

Code Visualizer is a development support tool to generate sequence diagram and class diagram from java source code. Sequence diagrams are generated for each public method which belong to specified class. Class diagrams are generated for each sequence diagram and include classes which appeare in it as parameters and return types.

This (https://sitoolkit.github.io/sit-cv/designdocs/) is generated diagrams from this project.

Diagrams are provided as static web contents so you can see them with web browser. You can generate diagrams with two ways.

1. Server mode
2. Report mode

Server mode

Code Visualizer can be run as an application server. In this mode, you can see diagrams via http://localhost:8080/ and the diagram will be automatically re-generated after source code modification.

Report mode

Code Visualizer can be run as an batch application. After execution of report mode, diagrams are generated in project_root/docs/designdocs directory. These report files can be accessed with browser as local files and also published with web server or GitHub Pages.

Required Software

• JDK v11

• Graphviz(https://www.graphviz.org/)

  • For Windows, Graphviz is installed automatically
  • For macOS or Linux, you need to install manually with following commands.

• macOS

brew install graphviz

• Ubuntu

sudo apt update
sudo apt install graphviz

Demo

If you want to know how Code Visualizer works, try to run it for DDD Sample app (https://github.com/citerus/dddsample-core). You can run Code Visualizer on Windows and macOS using following commands of each OS. Before you try it, install JDK8 for DDD Sample app.

• Windows

git clone https://github.com/citerus/dddsample-core.git
cd dddsample-core
git checkout f01db3d2d8be14233403f363f128645b633d2952
1>.env echo JAVA_HOME=C:\\path\u0020to\\jdk1.8
curl -o sit-cv-app-1.1.1-exec.jar -G http://repo1.maven.org/maven2/io/sitoolkit/cv/sit-cv-app/1.1.1/sit-cv-app-1.1.1-exec.jar
java -jar sit-cv-app-1.1.1-exec.jar --cv.analyze-sql

• macOS, Linux

git clone https://github.com/citerus/dddsample-core.git
cd dddsample-core
git checkout f01db3d2d8be14233403f363f128645b633d2952
echo JAVA_HOME=/path/to/jdk1.8 > .env
curl -o sit-cv-app-1.1.1-exec.jar -G http://repo1.maven.org/maven2/io/sitoolkit/cv/sit-cv-app/1.1.1/sit-cv-app-1.1.1-exec.jar
java -jar sit-cv-app-1.1.1-exec.jar --cv.analyze-sql

Exclude methods owned by the parent repository class (HibernateRepository) from the CRUD matrix, because all methods based this class converge to the same CRUD.

Download the configuration file to project root.

curl -o sit-cv-config.json -G https://raw.githubusercontent.com/Xenuzever/sit-cv-config-dddsample-core/master/sit-cv-config-dddsample-core.json

After running last java command, you can see following log on your console.

Started SitCvApplication in 00.000 seconds (JVM running for 00.000)

Then you can access http://localhost:8080 with browser and see UML diagrams of DDD Sample app.

• Report Mode

Running java command with --cv.report option, static report files(html, css, js) are generated to docs/designdoc directory.

java -jar sit-cv-app-1.1.1-exec.jar --cv.report

You can see diagrams by opening docs/designdoc/index.html with browser.

• Generate CRUD matrix

Run with the --cv.analyze-sql option to generate a CRUD matrix. Use this option to automatically run tests and analyze logs to get SQL.

# Server Mode
java -jar sit-cv-app-1.1.1-exec.jar --cv.analyze-sql

# or Report Mode
java -jar sit-cv-app-1.1.1-exec.jar --cv.analyze-sql --cv.report

How to Use in Your Project

Maven Project

If your project uses Maven, add plugin to pom.xml of your project.

<project>
  ...
  <build>
    <plugins>
        <plugin>
            <groupId>io.sitoolkit.cv</groupId>
            <artifactId>sit-cv-maven-plugin</artifactId>
            <version>1.1.1</version>
        </plugin>
    </plugins>
  </build>
  ...
</project>

Then you can use following commands in your project directory.

• Server mode

mvn sit-cv:run

• Report mode

mvn sit-cv:report

• Generate CRUD matrix

# Server Mode
mvn sit-cv:run -Danalyze-sql=true

# or Report Mode
mvn sit-cv:report -Danalyze-sql=true

Gradle Project

If your project uses Gradle, add plugin to build.gradle of your project.

• Using the plugins DSL:

plugins {
  id "io.sitoolkit.cv.sit-cv-gradle-plugin" version "1.1.1"
}

test {
  testLogging {
    showStandardStreams = true
  }
}

• Using legacy plugin application:

buildscript {
  repositories {
    maven {
      url "https://plugins.gradle.org/m2/"
    }
  }
  dependencies {
    classpath group: 'io.sitoolkit.cv', name: 'sit-cv-gradle-plugin', version:'1.1.1'
  }
}

apply plugin: 'sit-cv-gradle-plugin'

Then you can use following commands in your project directory.

• Server mode

gradlew cvRun

• Report mode

gradlew cvReport

• Generate CRUD matrix

# Server Mode
gradlew cvRun --analyze-sql

# or Report Mode
gradlew cvReport --analyze-sql

Configuration for Your Project

If you want to customize filter condition to draw classes on diagrams, put sit-cv-config.json in your project root directory. It's JSON structure is as follows.

• sit-cv-config.json

{
  "override": false,
  "lifelines": [
    {
      "name": ".*Controller.*",
      "annotation": ".*Controller",
      "entryPoint": true
    },
    {
      "name": ".*Service.*"
    },
    {
      "name": ".*Repository.*",
      "annotation": ".*(Repository|Named)",
      "dbAccess": true
    },
    {
      "name": ".*Factory.*",
      "lifelineOnly": true
    },
    {
      "name": ".*Specification.*",
      "lifelineOnly": true
    }
  ],
  "asyncAnnotations": ["Async", "Asynchronous"],
  "sqlLogPattern": {
    "start": ".*Pattern before SQL starts.*",
    "end": ".*Pattern after SQL ends.*"
  },
  "showAccessor": false
}

Key	Description	Default value
override	Ignore default configuration.	false
lifelines	Specify classes to draw as a lifeline in the sequence diagram.
name	Pattern to match class qualified name.
annotation	Pattern to match qualified annotation name of class.
entryPoint	Set true to recognize as a entry point, i.e. left end class of the sequence diagram.	false
lifelineOnly	Set true to hide internal processing, i.e. messages to itself.	false
dbAccess	Set true to recognize as a repository class. This is used to generate CRUD matrix.	false
exclude	Excludes methods owned by classes matching the pattern from the CRUD matrix.	false
asyncAnnotations	Annotaion names to recognize as asynchronous method.
sqlLogPattern	Filter rule to find SQL from test log. This is used to generate CRUD matrix.
start	Pattern to match the line just before SQL starts.
end	Pattern to match the line just after SQL ends.
match	Pattern to match the string enclosed in parentheses.
showAccessor	Set true to show setter and getter method in the class diagram.	false