gigfork / dupfind Goto Github PK

Version 1.0.1

dupfind: Scan JavaScript code and find duplicated sections.

Synopsis
--------
dupfind [file_set ...] [file_pattern ...]

Description
-----------
dupfind will read multiple JS files and look for duplicated segments of code.
It matches tokens, so it will ignore comments and it's insensitive to
whitespace.  It can do a fuzzy match, which will also ignore changes to
variable names.  You can define filesets in a configuration file, for often
used files.

When run with no arguments, default filesets from the configuration file will
be scanned.  When run with arguments, the arguments are either the names of
filesets in the configuration file, or file patterns to match.

Configuration
-------------
You can create a configuration file for dupfind to use.  dupfind will search
for the configuration file in these locations, in this order:

   ./dupfind.cfg
   ~/.dupfind.cfg
   /home/y/conf/dupfind/dupfind.cfg

If it doesn't find a configuration file, it will search all .js files in or
below the current directory.

The configuration file should contain JSON that looks like this:

{
    min:        30,
    max:        500,
    increment:  10,
    fuzzy:      true,

    sources:
    [
        {
            name:   "yui",
            def:    true,
            root:   "/home/y/share/htdocs/yui3",
            directories:
            [
                "build"
            ],
            include:
            [
                "*.js"
            ],
            exclude:
            [
                "*/.svn",
                "*simpleyui.js",
                "*-[^/]*.js",
                "*yui.js",
                "*datatype*"
            ]
        }
    ]
}

min:         The minimum number of consecutive duplicated tokens required to
             report the duplication.
max:         The maximim number of consecutive tokens to check
increment:   dupfind looks for duplication multiple times.  The first time, it
             looks for 'max' consecutive tokens.  It reduces the number by
             'increment' and checks again (I guess 'increment' should really be
             'decrement'...)  It continues until the number being checked is
             less than 'min'.
fuzzy:       'true' to do a fuzzy match.  A fuzzy match ignores changes to
             variable names.
sources:     An array of objects describing the files to scan.

Each object in the sources array contains:

name:        The name of the file set.  This name can be provided as an
             argument on the command line to limit scanning to this fileset.
def:         'true' if this is a default fileset.  All default filesets will be
             scanned when dupfind is executed with no arguments.
root:        The root directory for the fileset.
directories: An array of strings.  These are subdirectories under the root to
             scan.  If this member isn't present, all subdirectories are
             scanned.
include:     Files to include.  This is a DOS style regular expression - '.'
             means '.', '*' means '.*' and '?' means '.'.  The expression is
             matched against the full path to each regular file (not
             directories), and it has to match for the file to be scanned.
exclude:     Files and directories to exclude.  This is also a DOS style
             regular expression.  Matching files are excluded.  Matching
             directories aren't scanned at all (meaning they're not recursed
             into, either.)

-Steve Francis
[email protected]