This script extracts the raw Pcode from Ghidra and puts it into dedicated Object which are then serialised using Gson. The result is a Json file containing the processed binary as Pcode Objects.
You need the third party library gson which is available at Maven Central
https://search.maven.org/artifact/com.google.code.gson/gson/2.8.6/jar
When running the script for the first time, remember to call it with the --gson flag which
automatically moves the gson library to the .../.ghidra/.ghidra_9.X.X_PUBLIC/plugins/ directory
and the --plugin flag to identify the location of .../.ghidra/.ghidra_9.X.X_PUBLIC/.
In case the plugins directory does not exist, it is created automatically.
The script start_analysis.py is used to call PcodeExtractor.java. It is necessary to pass the location of ghidra via the --ghidra flag and the location of the input binary via the --import flag.
Each time the script is called, it creates a new Ghidra project in a tmp folder in this project's root directory. The project is then deleted after the binary has been processed. The resulting Json file is then also put into the tmp folder.
python3 start_analysis.py --gson $Home/gson-2.8.6.jar --plugin $HOME/.ghidra/.ghidra_9.1.2_PUBLIC/
--ghidra $HOME/ghidra_9.1.2_PUBLIC/ --import $HOME/<binary>