PLC CODE ANALISYS FOR SIMILARITY. Olivér Hornyák and Gábor Sáfrány

Transcription

1 PLC CODE ANALISYS FOR SIMILARITY Olivér Hornyák and Gábor Sáfrány

2 Draft Introduction PLC programming languages Computer algorithms for detection of similarity Local alignment detection using Smith- Waterman algorithm Finding similarity in Structured Text

3 Introduction Code reuse Software libraries approach Coding standards Concepts taken from plagiarism detection

4 PLC programming languages CP contact plan: a PLC programming language with the support in net-and relay switch plans. FBD Function Block Diagram: - FBD is another graphical programming language. The main concept is the data flow that start from inputs and passes in block(s) and generate the output. LD Ladder Diagram: Ladder diagrams are specialized schematics commonly used to document industrial control logic systems. They are called "ladder" diagrams because they resemble a ladder, with two vertical rails (supply power) and as many "rungs" (horizontal lines) as there are control circuits to represent IL instruction list: Instruction list (IL) programming is defined as part of the IEC standard. It uses very simple instructions similar to the original mnemonic programming languages developed for PLCs. ST Structured Text: Structural Text is a high level PLC programming language such as Pascal. SFC - Sequential Function Charts: Sequential Function Charts have long been established as a means of designing and implementing sequential control systems utilising programmable controllers. The Programming Standard IEC includes a graphical implementation of SFC s in its suite of programming languages.

5 Computer algorithms Metrics: n 1 = number of unique or distinct operators. n 2 = number of unique or distinct operands. N 1 = total usage of all the operators. N 2 = total usage of all the operands. Using these metrics we can calculate: V=(N 1 +N 2 ) log 2 (n 1 +n 2 ) (1) E=(n 1 N 2 (N 1 +N 2 ) log 2 (n 1 +n 2 ))/2n 2 (2) where V refers to the volume of the program and E refers to the efforts to create the program

6 Levels of software similarity

7 Smith-Waterman algorithm To compare two molecular sequences: A=a 1 a 2... a n and B=b 1 b 2... b m. The algorithm works as follows: 1. Set up a H n+1 x m+1 matrix whose first row and column have the 0 index and are zeroed. H k0 =H 0l =0 for 0 k n and 0 l m

8 Smith-Waterman algorithm Then H ij is the maximum similarity of two segments ending in a i and b j respectively. It is calculated as H ij Hi 1, max max max j 1 k 1 l 1 s( ai, bj ), Hi jk Wk, Hi, j 1 Wl where s(a i,b j ) is a score function for similarity, W k is a weight (cost) function for a k-long deletion and W l is a cost function for inserting l length of new sequence

9 Smith-Waterman algorithm The highest score in the matrix indicates the maximum local alignment of the two sequences. Once the matrix elements are calculated the maximum element has to be found. That refers to the end of the maximum alignment. The traceback algorithm will find the way back. Find the next highest score. A diagonal jump implies there is an alignment (either a match or a mismatch). A top-down jump implies there is a deletion. A left-right jump implies there is an insertion. See Figure for the traceback steps.

10 Example: Structured Text INTERFACE USEPACKAGE CAM; PROGRAM StartUp; **** PROGRAM Movement; END_INTERFACE IMPLEMENTATION PROGRAM startup ** HMI_rotaryknife_show_position := 200; *** END_PROGRAM *** PROGRAM movement VAR backcurrpos :INT := 0; END_VAR LABEL1: ; HMI_rotaryknife_show_position INTERFACE USEPACKAGE CAM; PROGRAM Movement; END_INTERFACE IMPLEMENTATION PROGRAM movement VAR backcurrpos :INT := 0; END_VAR LABEL1: ; HMI_rotaryknife_show_position := HMI_rotaryknife_show_position - 5; IF HMI_rotaryknife_show_position = 50 THEN current_offset[0] := 95; backcurrpos := current_offset[0]; END_IF;

11 Example: Structured Text HMI_rotaryknife_show_position - 5; IF HMI_rotaryknife_show_position = 50 THEN current_offset[0] := 95; backcurrpos := current_offset[0]; END_IF; IF HMI_rotaryknife_show_position = 0 THEN HMI_rotaryknife_show_position := 200; END_IF; GOTO label1; END_PROGRAM END_IMPLEMENTATION IF HMI_rotaryknife_show_position = 0 THEN HMI_rotaryknife_show_position := 200; END_IF; GOTO label1; END_PROGRAM END_IMPLEMENTATION

12 Keywords coding INTERFACE - A USEPACKAGE - B CAM - C PROGRAM -D StartUp - E PROGRAM - D Movement - F END_INTERFACE - G IMPLEMENTATION - H PROGRAM - D startup - E HMI_rotaryknife_show_position - I END_PROGRAM - J PROGRAM - D movement - F VAR - K backcurrpos - L INT - M END_VAR - N LABEL1 - O HMI_rotaryknife_show_position - I HMI_rotaryknife_show_position - I IF - P HMI_rotaryknife_show_position - I THEN - Q current_offset - R backcurrpos - L current_offset - R END_IF - S IF - P HMI_rotaryknife_show_position - I THEN - Q END_IF - S GOTO - T LABEL1 - O END_PROGRAM - U END_IMPLEMENTATION V

13 Result ABCDEFGHDEIJDFKLMNOIIPIQRLRSPIQSTOUV ABC FGHD FKLMNOIIPIQRLRSPIQSTOUV

14 Similarity Analyzer Application

15 Similarity Analyzer Application green color stands for the match, blue color stands for delete and red color stands for insert.

16 Conclusions Smith and Waterman algorithm is capable of detecting local alignment in ST code. Some further development is required to automate keyword detection and to improve similarity detection.

17 Acknowledgements Thank you for your kind attention! The described work was carried out as part of the TÁMOP-4.2.2/B-10/ project in the framework of the New Hungarian Development Plan. The realization of this project is supported by the European Union, cofinanced by the European Social Fund.