📰 What is pochi

🔑 Key idea

Almost traditional birthmark systems are usually simple input and output. For example, an user specifies the type of target birthmark, $p$ and $q$ as input, and obtains result as output. The process of such system is hard to use. Because, when the user would perform some process (ex. filtering results), it requires the update of birthmark system.

In general, the process of birthmarking is composed of extraction phase, and comparison phase. Also, the both phases are performed to huge amount of programs.

Therefore, the birthmark system accept the script file as input, and the user describes extraction and comparison of birthmarks in the script file. If the user would perform filtering process, it is easy to perform the process.

Go totop of the page

🍴 Usage

🏃‍♂️ CLI Interface

pochi [OPTIONS] [SCRIPT_FILE [ARGV...]]
OPTIONS
    -c, --classpath <CLASSPATH>      specifies the classpaths for Groovy (JVM) separated with colon (:).
    -C, --config <CONFIG_FILE>       specifies the configuration file.
    -e, --expression <EXPRESSION>    specifies one line script.
    -w, --working-dir <DIR>          specifies the working directory.
    -v, --verbose                    sets as verbose mode.

    -h, --help                       prints this message.
SCRIPT_FILE [ARGV...]
    Groovy script file name and its arguments.
    If no script files and no expression were given, pochi runs on interactive mode.

Script file

The script files are parsed by the Groovy. For more detail, see 🐜 Examples.

🐳 Docker

Container images of pochi for Docker are:

• ghcr.io/tamada/pochi
  • 2.6.0, latest
  • 2.5.2
  • 2.5.1
  • 2.5.0
  • 2.4.6
    • 2.4.0
    • 2.3.24
    • 2.3.23
    • 2.3.21
    • 2.3.19
    • 2.3.18
    • 2.3.17
    • 2.3.16
    • 2.3.10
    • 2.3.2
    • 2.3.1
    • 2.3.0
    • 2.2.0
    • 2.1.0
    • 2.0.0
      • accept only .groovy script files.
    • 1.0.0
      • accept only .js script files.

To run pochi on Docker container OS, type the following commands.

$ docker run --rm -it -v "$PWD":/home/pochi ghcr.io/tamada/pochi:latest [OPTIONS] [SCRIPT [ARGV...]]

• OPTIONS: the options for pochi.
• [SCRIPT [ARGV...]]: script file for pochi.
• --rm: remove the container after running.
• -it: interactive and tty mode.
• -v "$PWD":/home/pochi: share volume $PWD in host OS to /home/pochi in the container OS.
  • $PWD must be the absolute path.

ghcr.io/tamada/pochi does not include groovy interactive shell environment. Therefore, it does not work on interactive mode. If want to run pochi on the interactive mode, use ghcr.io/tamada/pochi-groovysh image instead.

Environments in the docker container

• USER: pochi
• WORKDIR: /home/pochi
• JAVA_HOME: /opt/java (symbolic link from /opt/openjdk-11-minimal)
  • This Java runtime environment do not include unnecessary modules.
• POCHI_HOME: /opt/pochi (symbolic link from /opt/pochi-x.x.x)
• GROOVY_HOME: /opt/groovy (symbolic link from /opt/groovy-x.x.x, only exist on ghcr.io/tamada/pochi-groovysh image)

Go totop of the page

🏊‍♂️ The birthmarking flow

The birthmarking process in pochi shows in Figure 1.

The followings are the description of the nodes and edges in the flowchart.

• Classes
  • is the Java class files, almost included in the jar files, and the directory. pochi treats them as the object of DataSource
• Birthmarks
  • shows the extracted characteristics from Java class files by certain method. In the pochi, Birthmarks and Birthmark show the extracted birthmarks. The string representation of Birthmark (the return value of toString method) is CSV format, therefore, we can store them into some csv file.
• Pair
  • shows the pair list of extracted birthmarks. pochi shows these objects as Pair<Birthmark>.
• Comparisons
  • represents the comparison results of birthmarks by the certain similarity calculation algorithm. pochi shows these objects as Comparisons and Comparison.
• Extractor
  • extracts birthmarks from given class files by the certain algorithm. In the script file, pochi.extractor("ALGORITHM_NAME") obtains the instance of Extractor.
• Parser
  • parses the given csv file and build Birthmarks object. To get the instance of BirthmarkParser, call pochi.parser() method in the script file.
• PairMatcher
  • matches the pair of birthmarks by some algorithm. pochi.matcher("ALGORITHM_NAME") returns the instance of PairMatcher.
• Comparator
  • calculates similarity between two given birthmaks by the certain algorithm. In the script file, pochi.comparator("ALGORITHM_NAME") gets the instance of Comparator.