Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

CCFinderSW

github.com

  • Free plan
  • Windows
7.3 Club score Visit github.com

What CCFinderSW is

CCFinderSW is a free, MIT-licensed Java tool that finds similar code pairs through token-based clone detection. It runs from the command line: provide a source directory, language, and output file. The minimum clone-token threshold defaults to 50 and can be changed with the -t argument. It excludes clone pairs containing inserted or deleted sentences.

For each target language, execution requires a comment rule file. A reserved-word list is optional but recommended: without one, CCFinderSW detects type 1 clones; with one, it detects type 1 and type 2 clones. Users can modify optional files to configure lexical analysis, although the lexer does not use distinct mechanisms for each language and is not optimized for each one. The program recommends Java 8 or later and lists Windows as its platform. It can write CCFinder, CCFinderX, CCFinderSW text, or JSON output, which can be read by Gemini or CCFX. The repository offers a CCFinderSW-1.0.zip download.

Who it suits

It suits developers or researchers who need command-line token-based clone detection and can configure language rule files. Users should be comfortable with Java and Windows.

Good to know, and what to watch

  • Free, MIT-licensed software
  • Supports CCFinder, CCFinderX, text, and JSON output
  • Threshold can be changed with -t
  • Output can be read by Gemini or CCFX
  • Requires Java 8 or later
  • Requires a comment rule file per language
  • Lexer is not optimized for each language
  • PrettyPrinter and MetricsFiltering are not implemented

CCFinderSW

Free

Open source

Getting started

  1. Download CCFinderSW-1.0.zip from the repository
  2. Install or provide Java 8 or later
  3. Prepare a comment rule file for the target language
  4. Optionally add a reserved-word list for type 2 clone detection
  5. Run the command line with a source directory, language, and output file
  6. Choose a supported output format such as JSON or CCFinderSW text

Limits to check

Execution requires a comment rule file for the target language. Without a reserved-word list, detection is limited to type 1 clones; PrettyPrinter and MetricsFiltering arguments are marked not implemented.

The verdict

CCFinderSW provides configurable, command-line clone detection with several output formats. Its language setup requirements and stated lexer limits are key considerations for teams working across languages.

Questions people ask

Is CCFinderSW free?

Yes. It is free, open-source software under the MIT License.

What does it require to run?

The program is implemented in Java and recommends Java 8 or later. Windows is the listed platform.

Which clone types can it detect?

Without a reserved-word list it detects type 1 clones; with one it detects type 1 and type 2 clones.

What output formats are available?

It can output CCFinder, CCFinderX, CCFinderSW text, or JSON formats. Clone-pair output can be read by Gemini or CCFX.

Can the language analysis be changed?

Yes. Optional language files can be modified, while a comment rule file is required for the target language.

What the maker publishes

Affiliation
The README lists the author's affiliation as the Graduate School of Information Science and Technology, Osaka University, Osaka, Japan. source
Author
The repository identifies Yuichi Semura as the author. source
Bundled libraries
The README says the software contains libraries distributed under Apache License 2.0. source
Clone matching
The README says it excludes clone pairs containing inserted or deleted sentences. source
Clone types
Without a reserved-word list, the tool detects type 1 clones; with one, it detects type 1 and type 2 clones. source
Command line
The usage guide shows running the tool from the command line with a source directory, language, and output file arguments. source
Detection behavior
It excludes clone pairs that contain inserted or deleted sentences. source
Detection threshold
The minimum clone token threshold defaults to 50 and can be changed with the -t argument. source
Download
The repository provides a CCFinderSW-1.0.zip download. source
Encoding
The command line supports Shift-JIS, UTF-8, EUC-JP, and automatic encoding recognition. source
Execution
The tool runs from the command line and recommends Java 8 or later. source
Founded
2017 source
Language configuration
Language-specific lexical analysis is configured through optional files that users can modify. source
Language handling
It tailors lexical analysis to each language using optional files that users can modify. source
Language limitation
It does not use separate lexical analysis mechanisms for each language, and its lexer algorithm is not optimized for each language. source
Language setup requirement
Execution requires a comment rule file for the target language; a reserved-word list is optional but recommended. source
License
The repository license file includes the MIT License for CCFinderSW. source
License
The repository license file includes the MIT License for CCFinderSW and a BSD 3-clause license for ANTLR components. source
Output details
CCFinderSW output can represent clone information as pairs or sets in a text file. source
Output formats
It can output clone data in CCFinder, CCFinderX, CCFinderSW text, or JSON formats. source
Purpose
CCFinderSW is a token-based code clone detection tool that detects and outputs similar code pairs. source
Purpose
CCFinderSW is a token-based code clone detection tool that detects similar code fragments and outputs clone pairs. source
Runtime
The project recommends Java 8 or later and says the program is implemented only in Java. source
Support and limitations
The documentation provides troubleshooting instructions, while its PrettyPrinter and MetricsFiltering command-line arguments are marked not implemented. source

Checked against the maker's pages on 9 Oct 2026.