This bug report was migrated from our old Bugzilla tracker.

These attachments are available in the static archive:

• Shows incorrect behavior in SDL 2.0.10 (Windows) (bugxxxx_2.0.10.png, image/png, 2020-01-24 10:47:31 +0000, 8760 bytes)
• Shows incorect behavior in SDL 2.0.10 (MacOS) (bugxxxx_macos.png, image/png, 2020-01-24 10:48:50 +0000, 8670 bytes)
• Shows correct behavior in SDL 2.0.9 (Windows) (bugxxxx_2.0.9.png, image/png, 2020-01-24 10:49:55 +0000, 8711 bytes)
• Beginning and end points swapped (2.0.10) (bug4958.png, image/png, 2020-01-24 17:26:14 +0000, 8770 bytes)

Reported in version: 2.0.10
Reported for operating system, platform: Windows 10, x86

Comments on the original bug report:

On 2020-01-24 10:47:31 +0000, Richard Russell wrote:

Created attachment 4173
Shows incorrect behavior in SDL 2.0.10 (Windows)

SDL_RenderDrawLine draws the final point at coordinates (0,0) rather than in the correct place. Superficially it appears that the line is shorter than it should be (the docs state "SDL_RenderDrawLine() draws the line to include both end points") but in fact the last pixel is drawn, just in the wrong place!

This is a regression in SDL 2.0.10, so I don't think it is a duplicate of any of the other SDL_RenderDrawLine bugs. I have attached images showing what happens, enlarged for clarity, in 2.0.9 and 2.0.10 on Windows and in 2.0.10 on MacOS.

On 2020-01-24 10:48:50 +0000, Richard Russell wrote:

Created attachment 4174
Shows incorect behavior in SDL 2.0.10 (MacOS)

On 2020-01-24 10:49:55 +0000, Richard Russell wrote:

Created attachment 4175
Shows correct behavior in SDL 2.0.9 (Windows)

On 2020-01-24 17:26:14 +0000, Richard Russell wrote:

Created attachment 4176
Beginning and end points swapped (2.0.10)

This is what happens if you swap the beginning and end of the line, confirming that it's the last point (here 3,3) that gets plotted incorrectly at 0,0.

On 2020-01-25 02:50:24 +0000, Sam Lantinga wrote:

Ryan, can you look at this? This code seems suspicious...
data->glVertex2f(verts[(count-1)2], verts[(count2)-1]);

On 2020-01-25 10:14:26 +0000, Richard Russell wrote:

(In reply to Sam Lantinga from comment # 4)

This code seems suspicious...
data->glVertex2f(verts[(count-1)2], verts[(count2)-1]);

Sorry to interject, but it could be right when you consider it's equivalent to:

data->glVertex2f(verts[(count2)-2], verts[(count2)-1]);

On 2020-01-25 15:34:35 +0000, Richard Russell wrote:

Somebody kindly sent me a DLL of SDL 2.0.11 to try (see the related thread at the forum) and the issue seems to be fixed there. So although a search of Bugzilla didn't find any reference to this specific bug, it seems that it has been addressed.

On 2020-01-25 20:05:15 +0000, Sam Lantinga wrote:

Okay, thanks for letting us know!