This guide is part of the project GX-TXT Script Debugger. The screens come from real executions in GX-TXT dev112-P011-bis, already made for the debugger course; examples use synthetic data. The D codes identify the corresponding exercises of the full course on the debugger.

Download the six practical guides’ scripts (ZIP). To repeat this guide, extract the guide_02/scripts folder files.

D13 — Step Over in a sequence

Step Over performs current education and, in the ordinary case of a function call, returns to the next row of the caller without stopping within the called function.

Prepare the example

Import D13_nested.py to Python Scripts, save in the session broom and set a breakpoint to adjusted = adjust(sample). The example distributed with GX-TXT sums the values 1, 2, 3, 4 and 5: the expected result is NESTED_RESULT 15. The D13_nested.m correspondent uses the same calculation in Octave.

Start View → Debugger → Start Debug. If the initial breakpoint is before the breakpoint, press Continue to reach it. In the first iteration sample is 0 and total is 0. Press Step Over on the adjust(sample) call.

Python debugger after Step Over, with adjusted equal to 1 and the next line set to update total.
The Python debugger after Step Over: adjusted 1 and the next line updates total.

The arrow moves to the total += adjusted line. The panel shows adjusted = 1, while total = 0: the update of the total has yet to be executed. An additional Step Over performs the sum. The red point on the call remains and can stop the subsequent iterations again.

Exercise

To the first iteration which will be total after performing total += adjusted?

Worked solution

Vale 1: the previous total is 0 and adjusted worth 1. The line highlighted in the screenshot has not yet been executed.

D14 — Step Into a function

Step Into allows you to follow a call inside a language function. It is the useful command when the hypothesis concerns the internal calculation, rather than its only result.

Enter to adjust

Use D13_nested.py recorded at Session. Set a breakpoint on the adjusted = adjust(sample) line and start the debug. On the call break, press Step Into. If you leave from the first break of the program, use Continue to reach the breakpoint.

In the illustrated test we find ourselves at the second iteration: sample is 1. The debugger enters the adjust function, shows the increment = value + 1 line and the local value = 1 variable.

Python debugger inside adjust with value equal to 1 and the call stack visible.
Debugger Python in adjust with value equal to 1 and call stack.

The Call Stack panel shows adjust, accumulate and the script level. value is the local topic of the function called; sample instead belongs to the caller. At the first iteration the corresponding values are 0, because range(5) begins with 0.

Exercise

If sample is worth 1, what value will adjust return after performing increment = value + 1?

Worked solution

It will remain 2. Before the assignment, the debugger is on the calculation line and value is 1.

D15 — Step Out and return to the caller

After inspecting a function, Step Out continues its execution until returning to the caller. In this way you do not need to manually follow each instruction left in the function.

Exit to adjust

Open D13_nested.py, put a breakpoint on the adjust(sample) call and use Step Into as in the D14. When you’re stationed in adjust, press Step Out.

Back in the caller after Step Out, with adjusted equal to 2 in the second iteration.
Return to the caller after Step Out: adjusted 2 in the second iteration.

In the test, at the second iteration, the debugger returns to accumulate on the total += adjusted line. The panel shows sample = 1, adjusted = 2 and total = 1: the returned value has been assigned, but the upgrade of total is still to be executed. In the first iteration adjusted would be 1 and the previous total 0.

Any breakpoint encountered in the path can affect the next break: always check arrow, stack and state, instead of assuming the point reached.

Exercise

After the screenshot, how much will the total pass when the highlighted line will be executed?

Worked solution

3: the previous total is 1 and the value returned from the second call is 2.

D16 — The Call Stack and the selected frame

The call stack describes how the execution came to the pause point. Each row represents a frame, with the function, file and location associated.

Compare function and caller

With D13_nested.py, stop at adjust using the breakpoint on the call and Step Into, as in the D14. The stack contains adjust, its caller accumulate and the script level.

Select accumulate in the Call Stack panel to examine the caller. In the frame of the called function you find value; in the caller you can compare sample, total and other local variables present at that time.

Caller frame selected in the stack while execution is paused.
Caller frame selected in the stack during break.

Selecting a stack row does not run the program and does not move its true execution point. Change the context you are looking at; to resume with the execution commands, check the active frame and the arrow shown. An external frame can refer to another file: always read the path and the line number.

Exercise

If you select the caller while adjust is on break, is the adjust function already back?

Worked solution

No. The selection is used to observe another frame. To continue the function you need an execution command, for example Step Out.

Index of the six guides

Previous: Guide 01 · Next: Guide 03

#debugger #GNU Octave #GX-TXT #Python

Other steps of this project

Leave a Reply

Your email address will not be published. Required fields are marked *