Skip to content

Latest commit

 

History

History
87 lines (55 loc) · 3.94 KB

README.md

File metadata and controls

87 lines (55 loc) · 3.94 KB
Raw

🛠 Stereocode 1.0

💡 What does it do?

Stereocode classifies methods and classes based on their stereotypes.

Class and method stereotypes are defined in three papers presented at the IEEE International Conference on Software Maintenance (ICSM) in 2006, 2009, and 2010 by Dragan, Collard, and Maletic.

Stereocode takes a srcML archive as input, performs static analysis, and annotates each function and class tag in the XML input with an attribute indicating the detected stereotype. For example:

<class st:stereotype="entity"> ... </class>
<function st:stereotype="get"> ... </function>

This output can be used for further processing or analysis. For instance, the stereotype information can be embedded as documentation in the source code (using comments, Doxygen, Javadoc, etc.).

🔧 Installation and Build

  1. Prerequisites

  1. Clone or download this repository.

  2. Build using cmake:

cmake CMakeLists.txt -B build_path
cd build_path
make

Note: Stereocode is compatible with srcML v1.0, but might not work as intended in certain rare cases.

🚀 Usage

Stereocode is a command-line tool. It can process individual source files or entire systems written in C++, Java, and/or C#.

Demo:

# This line converts the PowerShell system to the srcML format
srcml PowerShell.zip -o PowerShell.xml

# Saves the output to PowerShell-output.xml
./stereocode PowerShell.xml -o PowerShell-output.xml

# For more options and help:
./stereocode --help

📜 Stereocode Options

-o, --output-file: File name of output - srcML archive with stereotypes.

-p, --primitive-file: File name of user supplied primitive types (one per line).

-g, --ignore-call-file: File name of user supplied calls to ignore (one per line).

-t, --type-token-file: File name of user supplied data type tokens to remove (one per line).

-i, --enable-interface: Identify stereotypes for interfaces (C# and Java).

-s, --enable-struct: Identify stereotypes for structs (C# and Java).

-e, --input-overwrite: Overwrite input with stereotype information.

-x, --txt-report: Output optional TXT report file containing stereotype information.

-z, --csv-report: Output optional CSV report file containing stereotype information.

-c, --comment: Annotates stereotypes as a comment before method and class definitetions (/** @stereotype stereotype */).

-l, --large-class [int]: Method threshold for the large-class stereotype (default = 21).

📓 Developer Notes:

The initial version of this code base was developed by Doleh and documented in his MS Thesis December 2021 at Kent State University. Later, Al-Ramadan updated Stereocode by adding support for additional programming languages (i.e., C# and Java), enabling the stereotyping of complete systems, and incorporating deep static analysis. This work was also documented in MS Thesis May 2024 at Kent State University. Previously, a prototype of Stereocode was developed by Collard and Dragan for Dragan's dissertation in December 2010 at KSU. This prototype was further extended for DySDoc 2018 by Decker and Collard.

Developers of stereocode 1.0:

  • Ali Al-Ramadan - Kent State University
  • Michael L. Collard - University of Akron
  • Michael Decker - Bowling Green State University
  • Zane Doleh - Kent State University
  • Jonathan I. Maletic - Kent State University
  • Nick Weber - Kent State University