OCCT3D
DiscussionsIssue archiveOCCT:Visualization

Archived issue #0023813

TKOpenGl, OpenGl_Texture ignores stride image property

Open CASCADEOCCT:Visualizationclosed4 public notes

Search issues

Description

Texture could be loaded incorrectly if rows in image are packed with padding bytes for alignment (Image_PixMap::SizeRowBytes()) or their order was specified from top-to-down rather than bottom-to-top (Image_PixMap::IsTopDown()).

Steps to reproduce

# This scenario reproduces problem with wrong dimensions
pload MODELING
pload VISUALIZATION
box b 1 2 3
vinit
vsetdispmode 1
vdisplay b
vfit
vdump texture_409.png rgb 409 409
vdump texture_412.png rgb 412 412

# texture loaded correctly
vtexture b texture_412.png

# texture is corrupted during uploading to OpenGL
vtexture b texture_409.png

Public activity

4 archived notes

Participants are labeled by their role within this record.

01Author
Patch is ready for review in CR23813 branch (based on CR24392 branch).
02Commenter 2
Dear Commenter 1,

The branch is reviewed without remarks.

Please test.
03Commenter 2
Dear Commenter 1,

Branch CR23813 (and products from GIT master) was compiled on Linux and Windows platforms and tested.
[revision removed]

Number of compiler warnings:

occt component :
Linux: 320 (320 on master)
Windows: 0 (0 on master)

products component :
Linux: 188 (188 on master)
Windows: 286 (286 on master)

Regressions/Differences:
No regressions/differences

Testing cases:
bugs vis bug23813 - OK

Testing on Linux:
Total MEMORY difference: 358409872 / 358543144
Total CPU difference: 43218.369999999966 / 43977.48000000039

Testing on Windows:
Total MEMORY difference: 413033320 / 412727724
Total CPU difference: 32455.953125 / 32189.140625

There is difference in images found by testdiff:
IMAGE bugs vis bug9517: bug9517_Driver1_Viewer1_View1.png differs
http://occt-tests/CR23813-master-occt/Debian60-64/diff-Debian60-64.html
http://occt-tests/CR23813-master-occt/Windows-32-VC9/diff-Windows-32-VC9.html

04Author
Dear Commenter 1

> IMAGE bugs vis bug9517: bug9517_Driver1_Viewer1_View1.png differs
This is an improvement (for regression which has not been reported) - see original issue #0009517, image now looks correctly.

Related records