This repository is part of the FrUITeR project. Please refer to FrUITeR's website for a complete guide on the latest version. The notes below are deprecated.

The source code for FrUITeR's implementation including different components (e.g., GUI Mapper, Test Processor); and Test Analyzer including ground truth mapping of GUI events to their canonical IDs.

• Static Test Analyzer is located at src/TestAnalyzer.java

• Perfect GUI Mapper is located at gui_mapper/gui_mapper_perfect.py

• AppFlow GUI Mapper is located at gui_mapper/gui_mapper_fse.py

• CraftDroid GUI Mapper is located at gui_mapper/gui_mapper_craftdroid.py

• Naive GUI Mapper is located at gui_mapper/gui_mapper_random.py

• Result Generator is located at gui_mapper/result_generator.ipynb

• Test Processor is located at other_components/test_processor.py

• Test Synthesizer is located at other_components/test_synthesizer.py

• Post-Synthesis Result Generator is located at gui_mapper/effectiveness_recalculator.ipynb

• Ground Truth Mapping is located at gui_mapper/ground_truth_mapping/

---

Note: Run Perfect Mapper first because it will output the missing GUI elements that are not labeled in the ground truth files :)

Instructions for getting final results of appflow, craftdroid, perfect, naive

1. For appflow, you need to first update its original mapping results by adding columns of original GUI event id, original GUI event xpath, prediction canonical id, correct canonical id.

Then run appflow/gui_mapper_appflow.py first to get the mapping results (e.g., transferred tests). Similarly, run baseline/gui_mapper_naive.py and gui_mapper_perfect.py to get corresponding mapping results.

For craftdroid, craftdroid/process_craftdroid_results.py helps with manually constructing craftdroid's raw mapping results of GUI elements (e.g., CraftDroid Transfer Results - Etsy-Geek.csv) based on the published results. With such results, run craftdroid/gui_mapper_craftdroid.py to get the mapping results in the same manner as the other gui mappers mentioned above.

2. After getting all the mapping results from step 1, run result_generator.ipynb to get the final results with all the calculated metrics. You need to run all the cells until combined_csv.to_csv("framework_results.csv") to output the final results to a csv file. (check the headers to delete the useless columns & achange 'event_array' to 'transferred'

3. go to process_final_results.py to add utility metrics

*process_final_results.py also has functions to process the final result files, such as adding columns for #src_test, #trans_test, #gt_test, calculating the avg of Naive, merging all results together to a single file.