Welcome to the ZDoom Forums!

I realize you are trying to be helpful, but running a code analyzer is anything but. We have access to such tools, too, and will use them as needed. No one asked you for your PVS-studio report, and if they wanted it they would have asked.

For the most part I still feel like a lot of this stuff here is nit-picky - some of the code is the way it is for a reason, and while it is good to be aware that it is that way, it does not do us a lot of good with a project that has so many contributors. Believe it or not, some redundancy that you pointed out is actually needed.

In the future I would ask you bear this in mind - if you are going to run PVS-studio over GZDoom then you need to keep 3 things in mind:
a) First, you must be willing to change the code yourself.
b) Second, you must understand the reason why the code is the way it is. If changing it defeats the purpose, then we will not accept the change.
c) Lastly, you must also ensure that any changes made would not ultimately break anything. When running PVS-studio or something else that big which makes such a massive report, the room for error is just too significant in order to solve all the problems. It would take a combined effort of the 4 of us to fix all such issues, and - to put it lightly - "ain't nobody got time for that".

We welcome assistance, but any sort of "here's 2000 problems, gotta go bye!" is not what we're looking for. If you think you can whittle these problems down, you're welcome to make a pull request. But otherwise, I feel this report has been addressed to a satisfactory degree, and I am going to close it.

Thanks for trying to help.

Unfortunately we'd rather not have people spamming us with output from random tools unless it serves a very specific purpose. I quote from the [url=https://forum.zdoom.org/viewtopic.php?f=48&t=49500]code submission guidelines[/url]:

[quote]Code sanitizer/static analyzer is not an excuse to contribute

If you don't know what you're doing, do not use these tools. We typically find that people who use static analyzers as an excuse to contribute apply the wrong fix. Similarly reporting the errors to us isn't especially helpful as it can become a cat and mouse game. We're not saying they're bad tools, but use them in your own projects first.[/quote]

Good job guys!
I checked Gzdoom project in PVS-studio, here results, maybe it's useful info
[url]https://www.dropbox.com/s/g8d3dgjiyznnnw2/GZDoom.plog.html?dl=0[/url]

[quote="dpJudas"]The one in r_draw.cpp should be changed to the following:

static int next_random = 0;

I must have been blind when coding the upscaled fuzz demon version for not noticing this bug.[/quote]
[url=https://github.com/coelckers/gzdoom/commit/495fc79fe93b0ea6a92e8889dbe14e426449ce17]Done[/url]

[quote="Graf Zahl"]Looking at those 3, I think you did one wrong and one not optimally.

If you want to set a double to the maximum, use DBL_MAX, not some hard coded constant.
And the third one should check height instead of the second 'width'.

The rest: Most should not be changed, unless the message indicates a missing pointer check outside the always true/false condition.[/quote]

[url=https://github.com/coelckers/gzdoom/commit/47e9975e9068e4c2f6020937122c9d2f83601994]Fixed[/url]

The one in r_draw.cpp should be changed to the following:

static int next_random = 0;

I must have been blind when coding the upscaled fuzz demon version for not noticing this bug.

Looking at those 3, I think you did one wrong and one not optimally.

If you want to set a double to the maximum, use DBL_MAX, not some hard coded constant.
And the third one should check height instead of the second 'width'.

The rest: Most should not be changed, unless the message indicates a missing pointer check outside the always true/false condition.

[url=https://github.com/coelckers/gzdoom/commit/2bc86b52c8e8cf2005e8ab3d894f7a62625d1caf]Fixed a few[/url]

I am not touching the ones I either cannot follow, or seem to be safety bars, or seem to be compatibility hacks. For any of the latter two, I think it should be Graf, exclusively, who deals with those.

For example, the PCX thing, we could simplify the statement quite a bit by removing the check, but at the same time it may be there for a reason despite the check always being false.

Those are commits when corresponding files were changed the last time.
And I wouldn’t say that it’s just a nitpicking. Most of them seem to be logic errors.

Why are you using different commit numbers for each and every line here? Most of these are more nitpicks than they are real problems. What is the purpose of this report?

I can check with PVS-studio and send Log file
This ten errors i found randomly tring adding IMGUI

Did you use a tool for this?

1. [url]https://github.com/coelckers/gzdoom/blob/30087f7cd7d64dde75c4f6e67ece85f333eaa929/src/gl/renderer/gl_postprocess.cpp#L941[/url]
Uninitialized variable fbestdist

2. [url]https://github.com/coelckers/gzdoom/blob/afdeed305bccc786fd019c191896426c8c1c87f6/src/vectors.h#L651[/url]
Non-void function should return a value

3. [url]https://github.com/coelckers/gzdoom/blob/78061f1cf70f294244fafd1cc8b25ac5ea8ac1f7/src/win32/fb_d3d9.cpp#L1226[/url]
Copypast ? box.right > 0 && box.right > 0

4. [url]https://github.com/coelckers/gzdoom/blob/f34ededdef02c4588a2e457633ca68b4f5a827f9/src/swrenderer/drawers/r_draw.cpp#L194[/url]
if (next_random == FUZZ_RANDOM_X_SIZE) always false

5. [url]https://github.com/coelckers/gzdoom/blob/61c91f05fc028868c27b77b191a90aea0ffd54b5/src/textures/pcxtexture.cpp#L589[/url]
Expression always false

6. [url]https://github.com/coelckers/gzdoom/blob/61c91f05fc028868c27b77b191a90aea0ffd54b5/src/textures/pngtexture.cpp#L113[/url]
Expression always true

7. [url]https://github.com/coelckers/gzdoom/blob/4beefb70075df5cfa3a923ef7c19fb9d6926b6de/src/scripting/backend/codegen.cpp#L9028[/url]
Expression always false

8.[url]https://github.com/coelckers/gzdoom/blob/61c91f05fc028868c27b77b191a90aea0ffd54b5/src/sfmt/SFMT.cpp#L526[/url]
if (size >= 623), else if (size >= 68), else if (size >= 39)
Expression always false

9. [url]https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/am_map.cpp#L2836[/url]
Expression always true

10. https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L198 and
https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L208
Expressions always true

11. [url]https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L5515[/url] and
[url]https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L6314[/url]
Expressions always true

12.[url]https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L6348[/url]
Some part of expression always true: mobj

13.[url]https://github.com/coelckers/gzdoom/blob/6a340927426c0ac049663821e59bb80bbff5c965/src/p_actionfunctions.cpp#L6527[/url]
Some part of expression always true: !checker