Duplicate code analysis

In large program systems often occure duplicate codes, which is a computer programming term for a sequence of source code that occurs more than once. There are two ways in which two code sequences can be duplicates of each other: syntactically and functionally. This new feature can detect the syntactically similar duplicates, that only differ in atoms, constants (integer, float and string) and the name of variables. A lot of identical or very similar code fragments can result the application of copy and paste, for example.

Parameters

Parameters can be defined by a proplist. The possible properties are summed up in the table below. All the properties are optional.

Property	Description	Type	Default value	Example
files	files in which the search is carried out	[Module::atom() [[BR]] |Filepath::string() |RegExp::string() |File::string()]	all files from the database	{files,["/usr/local/lib/erlang/lib/mnesia-4.5/src/mnesia_log.erl", module]}
minlen	minimal length of a clone(length is in tokens)	integer()	10	{minlen,50}
minnum	minimal number of clones in one clone group	integer()	2	{minnum,5}
overlap	maximum length that duplicates can overlap each other (length is in tokens)	integer()	0	{overlap,1}
output	name of the file in which to save the result of the analysis	string()	-	{output,"result.txt"}
name	name of the result in the table	atom()	the string "temp" concatenated with the timestamp	{name,referl}

Interfaces

Web interface

See Duplicate code analysis on the web interface.

Console interface

We currently have two interface functions in the ri module: search_duplicates/0 and search_duplicates/1. The first uses the default values of the parameters. The second takes a proplist as parameter described above. Both interface function provide information about the progress of the process. The result is the list of clone groups. Every clones are defined by the path of the file and information about the start and end positions (line and column number). An example can be found at the bottom of the page.

Examples

ri:search_duplicates().

ri:search_duplicates([
        {files, ["/home/user/dups/dup1.erl",
                 "/home/[0-9a-zA-Z/_.\-]+/src"]},
        {minlen, 50},
        {minnum, 3},
        {overlap, 1},
        {output, "result.txt"},
        {name, filtered}]).

Small example of the return value:

ri:search_duplicates([{files,[dup1, dup2]}]).

Initial clone detection started.
Initial clone detection finished.
Trimming clones started.
Trimming clones finished.
Filter clones finished.
Calculating positions started.
Calculating positions finished.
[[[{filepath,"/home/user/dups/dup1.erl"},
   {startpos,{15,1}},
   {endpos,{17,24}}],
  [{filepath,"/home/user/dups/dup2.erl"},
   {startpos,{5,1}},
   {endpos,{7,24}}]],
 [[{filepath,"/home/user/dups/dup1.erl"},
   {startpos,{8,1}},
   {endpos,{14,11}}],
  [{filepath,"/home/user/dups/dup2.erl"},
   {startpos,{12,1}},
   {endpos,{18,11}}]]]