Skip to content

danielduduta/ninka

 
 

Repository files navigation

-*-org-*-

Contact information.

Any feedback will be appreciated. You can email us at Daniel M. German <dmg@uvic.ca> and Yuki Manabe <y-manabe@ist.osaka-u.ac.jp>

Introduction

Ninka is license identification tool that identifies the license(s) under which a source file is made available.

This tool uses a source file as input and outputs the licenses identified within that file.

If you need to know the detail of Ninka, please see the following paper:

Daniel M. German, Yuki Manabe and Katsuro Inoue. A sentence-matching method for automatic license identification of source code files. In 25nd IEEE/ACM International Conference on Automated Software Engineering (ASE 2010). You can email me (dmg@uvic.ca) for a copy or download it from

http://turingmachine.org/~dmg/papers/dmg2010ninka.pdf

If you use Ninka for research purposes, we would appreciate you cite the above paper.

Contributors

  • Anthony Kohan for writing the excel and sqlite backends.
  • Armijn Hemel from Tjaldur Software Governance Solutions for multiple bug reports and suggestions

License

Except for the directories comments and splitter, Ninka is licensed under the GPLv2+

Copyright (C) 2009-2010 Yuki Manabe and Daniel M. German

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

  • splitter.pl is a derivative work of the Rule-based sentence splitter script by Paul Paul Clough. Please see splitter/README for details.
  • comments is based on a program to remove comments by Jon Newman, it is released under the GNU General Public License Version 2 or (at your option) any later version.

Requirements

How to install

  1. Unpack the distribution in a directory.
  2. Build and install comments (make sure it is somwehere in the path) (see directory comments)
  3. Build splitter.pl (see splitter/README for instructions)

Usage:

Ninka uses a pipe model (see below). Each step of the “pipe” creates a file, but

ninka.pl [options] [filename]

Available options -v verbose -d delete intermediate files -C force creation of comments file -c stop after creation of comments -S force creation of sentences file -s stop after creation of sentences -G force creation of goodsent file -g stop after creation of goodsent -T force creation of senttok file -t stop after creation of senttok -L force creation of license file -f force all processing

Example:

ninka.pl foo.c

It will create five files:

  1. foo.c.comments: extracted the first two comments blocks, where the license is usually
  2. foo.c.sentences: creates the list of sentences in the license statement
  3. foo.c.goodsent: contains sentences that are likely to be part of a license statement
  4. foo.c.badsent: contains the sentences that are not part of foo.c.goodsent
  5. foo.c.senttok: Each sentence in *.goodsent is converted into a tokenized sentence (or unmatched, when none matches)
  6. foo.c.license: List of licenses found in the file. Its contains a single line with 3 fields (semicolon delimited):
    • Licenses
    • Unmatched sentences in *.senttok that were not matched

Ninka model

Ninka uses a pipe-model. Each stage of the pipe does something very specific:

  1. Comment extractor.
    • directory: extComments
    • command: extComments.pl, might use comments (included in distribution)
    • Purpose: Extracts top comments of source code. If no comment extractor is known for the language, then extracts top lines from source (currently 700)
    • Creates <filename>.comments file
  1. Split sentences in comments
    • directory: splitter
    • command: splitter.pl
    • Purpose: Ninka works by matching sentences of licenses, hence it needs to properly break text into sentences.
    • Outputs <filename>.sentences
  2. Filter “good” sentences.
    • directory filter
    • command: filter.pl
    • Purpose: some sentences are related to a license, some are not. It is valuable to know if a file contains lines that look like a license or not (e.g. to know that a file has no license)
    • Outputs: <filename>.goodsent, and <filename>.badsent (not used)
  3. Tokenizes sentences
    • Directory senttok
    • command: senttok.pl
    • Purpose: It creates a file that corresponds to the recognized sentence tokens. For each sentence, it outputs its sentence token, or unknown otherwise.
    • Outputs: <filename>.senttok
  4. Matches sentences to licenses
    • Directory matcher
    • Command: matcher.pl
    • Purpose: looks at the sequence of sentence tokens and outputs the licenses found
    • Output: <filename>.license

The script ninka.pl takes care of all these steps, and optionally removes intermediary files, and writes to the stdout the licenses found.


How to read the output:

Assume, for example, this output:

eq.c;MITX11noNotice;1;2;2;6;0;Copyright,-1,-1,DualLicenseIntention,GPLorOpenBSDTypeVer2,BSDpre,BSDcondSource,BSDcondBinary

So Ninka detects all the sentences, including the MIT variant, it finds the GPL bsd intention. But the license is not really BSD.

The disclaimers are not what you expect. Now, in all fairness, maybe this is another license.

Let me translate the output for you:

file: eq.c; License(s) found: MITX11noNotice

;1;2;2;6;0; Found 1 license Composed of 2 lines (tokens) 2 tokens were ignored 6 tokens were not mached: Copyright,-1,-1,DualLicenseIntention,GPLorOpenBSDTypeVer2,BSDpre,BSDcondSource,BSDcondBinary (-1 indicates where a match happened) 0 tokens were unknown

Another example:

nsAccessibilityUtils.cpp;MPLv1_1;1;1;3;7;2;UNKNOWN,MPL1_1_GPL2_LGPL2_1intentionVer0,1,-1,-1,MPLsee,Copyright,-1,Altern,UNKNOWN,MPLoptionNOTGPLVer0,MPLoptionIfNotDelete3licsVer0,licenseBlockEnd

License matched:MPLv1_1; One license: 1; Composed of one token: 1; 3 token were ignored 3; 7 tokens were matched but not recognized as a license: UNKNOWN,MPL1_1_GPL2_LGPL2_1intentionVer0,1,-1,-1,MPLsee,Copyright,-1,Altern,UNKNOWN,MPLoptionNOTGPLVer0,MPLoptionIfNotDelete3licsVer0,licenseBlockEnd 2 of those tokens were unknown

About

a license identification tool for Source Code

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Perl 61.6%
  • C++ 32.3%
  • Roff 4.0%
  • Makefile 1.2%
  • C 0.9%