ASPA - Abstrakti SyntaksiPuu Analysaattori

Abstrakti SyntaksiPuu Analysaattori (ASPA) == Abstract Syntax Tree Analyser. It is a static analyser to support self-study and complying with recommended coding conventions at Fundamentals of Programming (CS1) course at LUT University.

Citation

If you wish to cite related research (preferred), cite a conference article at ICSE DOI link, same citation as text: R. Luukkainen, J. Kasurinen, U. Nikula, V. Lenarduzzi, ASPA: A static analyser to support learning and continuous feedback on programming courses. an empirical validation, in: Proceedings of the ACM/IEEE 44th International Conference on Software Engineering: Software Engineering Education and Training, ICSE-SEET ’22, Association for Computing Machinery, New York, NY, USA, 2022, p. 29–39. doi:10.1145/3510456.3514149. and same citation in bibtex format: @inproceedings{luukkainen_aspa_2022, address = {New York, NY, USA}, series = {{ICSE}-{SEET} '22}, title = {{ASPA}: {A} static analyser to support learning and continuous feedback on programming courses. {An} empirical validation}, isbn = {978-1-4503-9225-9}, shorttitle = {{ASPA}}, url = {https://dl.acm.org/doi/10.1145/3510456.3514149}, doi = {10.1145/3510456.3514149}, abstract = {For decades there have been arguments how to teach programming in the basic courses. Supportive intervention methods to improve students' learning and methods to improve assessment process have been widely studied. There are various successful methods to each topic separately, but only a few of them fit for both. In this work, we aimed at validating ASPA a static analyser tool that supports learning and continuous feedback on programming courses. For this purpose, we designed and conduct an empirical study among 236 students enrolled in the basic programming course, that voluntary adopted the tools during the project development activities. We first profiled the students, then, evaluated the attitude toward using ASPA, the perceived ease of use, and the perceived usefulness. Results showed that ASPA is a good helper for the entire course and especially the student's programming assignments, and it also helps to improve the students' grades.}, urldate = {2023-03-30}, booktitle = {Proceedings of the {ACM}/{IEEE} 44th {International} {Conference} on {Software} {Engineering}: {Software} {Engineering} {Education} and {Training}}, publisher = {Association for Computing Machinery}, author = {Luukkainen, Roope and Kasurinen, Jussi and Nikula, Uolevi and Lenarduzzi, Valentina}, month = oct, year = {2022}, keywords = {CS1, programming education, empirical software engineering, software education, static analysis tools}, pages = {29--39}, }

However, if you wish to cite directly this repository use "Cite this repository" to get Zenodo link.

Repository structure

bash . ├── ASPA_main.py ├── misc │ ├── ast_examples │ └── preanalyse_structures ├── results ├── src │ ├── BKT │ ├── analysers │ └── config └── tests ├── AR_functions ├── MR_file_structure ├── OK_files ├── PK_exception_handling ├── PT_basic_command ├── TK_file_handling └── TR_data_structure

ROOT

The root directory contains the main file, i.e. ASPA_main.py, which is used to run the program. In addition settings.json will be generated here after the initial execution. Finally there are repository related general files such as 1. .gitignore 2. LICENCE 3. README.md (this file)

misc

In this directory, there are all the miscellaneous material, which did not belong to any other directory. This include helper documentation files e.g. 1. examples of Abstract Syntax Trees (AST) which are used to check what type of tree some piece of codes created. 2. List of code structures which are detected with StructureDetector class.

results

This directory is for the result files, i.e. output of analysis will be written here.

src

This directory and its subdirectories contain all of the source code files, excluding ASPA_main.py.

tests

This directory contains expected output files and subdirectories containing files which are analysed to get these results. Each subdirectory is related to a single AST analyser, excluding OK_files, which contain files with only valid structures, i.e. they should not create any errors, warnings or notes. Excpeted output files can be used as a primitive regression testing to check that previous checks did not fail due to the new check or change. No unit tests, yet. 1. ARfunction == Function structure 2. MRfilestructure == Module/file structure 3. PKexceptionhandling == Exception handling 4. PTbasiccommand == Basic commands, e.g. loops, naming and unreachable code 5. TKfilehandling == File handling 6. TRdata_structure == Data structures e.g. classes (and objects)