# Program slicing

In computer programming, program slicing is the computation of the set of program statements, the program slice, that may affect the values at some point of interest, referred to as a slicing criterion. Program slicing can be used in debugging to locate source of errors more easily. Other applications of slicing include software maintenance, optimization, program analysis, and information flow control.

Slicing techniques have been seeing a rapid development since the original definition by Mark Weiser. At first, slicing was only static, i.e., applied on the source code with no other information than the source code. Bogdan Korel and Janusz Laski introduced dynamic slicing, which works on a specific execution of the program (for a given execution trace). Other forms of slicing exist, for instance path slicing.

## Static slicing

Based on the original definition of Weiser, informally, a static program slice S consists of all statements in program P that may affect the value of variable v in a statement x. The slice is defined for a slicing criterion C=(x,v) where x is a statement in program P and v is variable in x. A static slice includes all the statements that can affect the value of variable v at statement x for any possible input. Static slices are computed by backtracking dependencies between statements. More specifically, to compute the static slice for (x,v), we first find all statements that can directly affect the value of v before statement x is encountered. Recursively, for each statement y which can affect the value of v in statement x, we compute the slices for all variables z in y that affect the value of v. The union of all those slices is the static slice for (x,v).

### Example

For example, consider the C program below. Let's compute the slice for ( write(sum), sum ). The value of sum is directly affected by the statements "sum = sum + i + w" if N>1 and "int sum = 0" if N <= 1. So, slice( write(sum), sum) is the union of three slices and the "int sum = 0" statement which has no dependencies:

1. slice( sum = sum + i + w, sum)
2. ,
3. slice( sum = sum + i + w, i)
4. ,
5. slice( sum = sum + i + w, w)
6. , and
7. { int sum=0 }.

It is fairly easy to see that slice( sum = sum + i + w, sum) consists of "sum = sum + i + w" and "int sum = 0" because those are the only two prior statements that can affect the value of sum at "sum = sum + i + w". Similarly, slice( sum = sum + i + w, i) only contains "for(i = 1; i < N; ++i) {" and slice( sum = sum + i + w, w) only contains the statement "int w = 7".

When we union all of those statements, we do not have executable code, so to make the slice an executable slice we merely add the end brace for the for loop and the declaration of i. The resulting static executable slice is shown below the original code below.

```int i;
int sum = 0;
int product = 1;
int w = 7;
for(i = 1; i < N; ++i) {
sum = sum + i + w;
product = product * i;
}
write(sum);
write(product);
```

The static executable slice for criteria (`write(sum)`, sum) is the new program shown below.

```int i;
int sum = 0;
int w = 7;
for(i = 1; i < N; ++i) {
sum = sum + i + w;

}
write(sum);
```

In fact, most static slicing techniques, including Weiser's own technique, will also remove the `write(sum)` statement. Since, at the statement `write(sum)`, the value of `sum` is not dependent on the statement itself. Often a slice for a particular statement x will include more than one variable. If V is a set of variables in a statement x, then the slice for (x, V) is the union of all slices with criteria (x, v) where v is a variable in the set V.

## Lightweight forward static slicing approach

A very fast and scalable, yet slightly less accurate, slicing approach is extremely useful for a number of reasons. Developers will have a very low cost and practical means to estimate the impact of a change within minutes versus days. This is very important for planning the implementation of new features and understanding how a change is related to other parts of the system. It will also provide an inexpensive test to determine if a full, more expensive, analysis of the system is warranted. A fast slicing approach will open up new avenues of research in metrics and the mining of histories based on slicing. That is, slicing can now be conducted on very large systems and on entire version histories in very practical time frames. This opens the door to a number of experiments and empirical investigations previously too costly to undertake.

## Dynamic slicing

Makes use of information about a particular execution of a program. A dynamic slice contains all statements that actually affect the value of a variable at a program point for a particular execution of the program rather than all statements that may have affected the value of a variable at a program point for any arbitrary execution of the program.

An example to clarify the difference between static and dynamic slicing. Consider a small piece of a program unit, in which there is an iteration block containing an if-else block. There are a few statements in both the `if` and `else` blocks that have an effect on a variable. In the case of static slicing, since the whole program unit is looked at irrespective of a particular execution of the program, the affected statements in both blocks would be included in the slice. But, in the case of dynamic slicing we consider a particular execution of the program, wherein the `if` block gets executed and the affected statements in the `else` block do not get executed. So, that is why in this particular execution case, the dynamic slice would contain only the statements in the `if` block.