Raven RawScreen Aspect Fix

[andrikpowell]

So a thing that's bothered me for quite a while is that the "RawScreen"s that Raven uses have only been rendered using the "Not Adjusted" stretch setting... regardless of the stretch settings.

This PR fixes that and the issue mentioned here (#467). They are both related to the same issue in different ways. The problem with the specific issue, was that the "RawScreen" under "Fit to Width" was supposed to fit to width (imagine that), but code was set to prevent that.

Now the "RawScreen"s will look as they should for all three settings: "Not Adjusted", "Doom Format", and "Fit to Width". I will mention that the "RawScreen" code seems a bit WIP, so eventually there may be a better way to approach this further down the line.

One of the reasons the format was previously forced was probably due to the Raven Statusbar not being rendered properly under "Not Adjusted"... But the "RawScreen"s don't have that issue.

I haven't been able to fix the Raven Statusbar for "Not Adjusted", so I have it using the "Doom Format" stretch settings (just as DSDA Doom currently does).

[andrikpowell]

Is this better?

[fabiangreffrath]

Most of these changes appear redundant with code that is alrady there. It would be easier to implement a function (or macro) that returns (raven && stretch == patch_stretch_not_adjusted) ? patch_stretch_doom_format : stretch and use this in the switch statements instead of redefining in a later paragraph what was already checked before. This would also hold up the "temporary" aspect of these changes, as that function could later be changed to just return stretch (or get removed entirely).

[andrikpowell]

Most of these changes appear redundant with code that is alrady there. It would be easier to implement a function (or macro) that returns (raven && stretch == patch_stretch_not_adjusted) ? patch_stretch_doom_format : stretch and use this in the switch statements instead of redefining in a later paragraph what was already checked before. This would also hold up the "temporary" aspect of these changes, as that function could later be changed to just return stretch (or get removed entirely).

I think this is close to what you were suggesting? (check force push)

[fabiangreffrath]

I think this is close to what you were suggesting? (check force push)

Yes, except that it now uses the global render_stretch_hud in InitStretchParam() instead of the passed stetch parameter.

[andrikpowell]

Yes, except that it now uses the global render_stretch_hud in InitStretchParam() instead of the passed stetch parameter.

How do you wanna get around that?

Because the only way to force that switch statement to use patch_stretch_doom_format is to not use the passed stretch parameter.

I guess we could have a if (stretch_hud != stretch) then stretch == stretch_hud above?

[fabiangreffrath]

How do you wanna get around that?

#define stretch_hud(a) ((raven && (a) == patch_stretch_not_adjusted) ? patch_stretch_doom_format : (a))

[andrikpowell]

How do you wanna get around that?

#define stretch_hud(a) ((raven && (a) == patch_stretch_not_adjusted) ? patch_stretch_doom_format : (a))

oh fascinating! This is actually the first time I've seen a #define statement with arguments! That's actually really cool.

Force pushed new version with that included.

[fabiangreffrath]

Looks good now, thank you!

[Pedro-Beirao]

Thank you both!