What changed in EDuke32, build by build

Pass types via ffi.typeof() instead of declaring them in the global namespace when possible.

Use a new flag to mark hard-coded enemies, not SPRITE_BADGUY. This fixes an issue where in E1L4, a pigcop would appear in the area you have to crawl under shrunk. Thanks to LLCoolDave1 for pinpointing the range of relevant revisions.

... when 'make' is invoked like "make ... STRIP=some_strip".

Whenever it should be not, STRIP is set to the empty string. This fixes the Lunatic RELEASE=1 build.

The pixel doubling now only applies to the area where the world scene is drawn, i.e. it may be smaller than the physical screen / WM window size. The optimized version is slightly faster than for non-doubled pixels for me (optimized build), but see code for caveats. Some other minor issues: - won't work when the world is drawn from demo cameras (and offscreen, but that matters less) - will leave a few pixels empty when running with x resolutions not evenly divisible by 4

... to have a little moew overview.

- more predefined vars - fix recursive states and ones with a stray "else" before the end

- Fix "gv" access. - camera - translator: comparison predicates, predefs...

Whether a sprite is considered for texel-hitscan is determined on the base tile number, not the individual animated tile numbers.

* Make the "classic status bar fullscreen viewport" mode accessible in classic too. * Make range of status bar scale 36..100 * Update ud.statusbarmode when executing OSD command r_size (ud.screen_size). NOTE: ud.statusbarmode is considered internal. Don't use from CON! * Make sure 1) loading any configuration and 2) menu bars work correctly. (The "classic status bar fullscreen viewport" mode will never be restored though because ud.statusbarmode isn't handled by the OSD var system).

One use was in determining the ray vector for the mouse-aiming hitscan in the editor. Unfortunately, the change doesn't make it any less broken in Polymost, even if the two instances were out of sync.

If enabled, dragging wall vertices will correct the xrepeat after the mouse button is released, so that the pre-drag absolute stretching is restored.

- Run it twice, since the first one is wrong. - Warn when attempting to align based on a top-oriented wall. When the sequence of walls to align has "windows", only the bottom parts will be correct. - Make the modifiers actually useful: * Pressing SHIFT aligns at most one wall, remove the old CTRL modifier. * The rest is as before: ALT makes the walls have (approximately) equal texture stretching, ['] (quote) aligns the immediate TROR-nextwalls.

Also, factor out 2x dup'd code of insertsprite() into do_insertsprite() and add searchwall-displaying code into package/samples/a.m32.

Also, add another flag, signifying that from the editor, also the "lastwall"s (i.e. the CCW-linked points) should get collected. This is to signal the editor that their wall lengths should be displayed, too.

This is done by introducing an additional internal bit, meaning "play only one instance". It is set for all sounds which have bit 1 ("repeat in the sound system") set at definesound time, but not those that set bit 1 temporarily (see r3336).

- weapon data - changing ones... (locals for now) - operators

Also, rewrite comparison to the non-tint in astub.c to be even less hackish.

- Stop using memcache on some other failure paths, particularly when failing to read from the on-disk texcache. - Factor out cache reading code (3x). - Fix endianness issues affecting big-enadian systems (one introduced by r3382, one existing before). Comment each B_LITTLE32 with whether we're converting from native to on-disk (little) endianness or back.

This obviously won't help performance as the scene has still to be drawn at the original resolution, but it's better than the draw-to-tile hack.

The SDL and Windows layers had slightly different code: the latter would dereference a NULL pointer if stdout.txt failed being write-opened.

When reaching wall limits, it is possible that only some circle points will be inserted and the result is left unfinished.

- fix error handing with recursing events - wallsofsect iterator, sector[]:contains() - more codegen; mangle function names - m32: compinside

On the C side, zrange, angrange and autoaimang are represented as DukePlayer_t members then.

Reproduced as follows (assuming all tiles have texel-hitscan for simplicity): In E2L5, shoot the opening switch with the shotgun, aiming for the border. The crash occurs because the *other*, depressed switch tile isn't yet loaded when we index into its tile storage. Dereferencing 0+small number == BAD!

Bump BYTEVERSION.

New local functions: Proj_DoRandDecalSize, SectorContainsSE13, HandleHitWall, Proj_CheckBlood, Proj_HandleKnee.

The diff may look daunting, but it's clear what is changed with git diff (...) --color-words='[a-zA-Z0-9_]+|[^[:space:]]' -b

Factoring out 2x almost duplicated code into {P,A}_PostFireHitscan().

The code is duplicated with small changes for the hardcoded and custom projectiles. Adding local functions P_PreFireHitscan(), A_PreFireHitscan() and Proj_MaybeAddSpread().

In the normal game, these arrays are conceptually [MAX_WEAPONS][MAXPLAYERS], allocated as CON per-player gamevars (e.g. WEAPONx_WORKSLIKE). For Lunatic, they are replaced with weapondata_t g_playerWeapon[MAXPLAYERS][MAX_WEAPONS].

Specifically, have a weapondata_t type mimicking the aplWeapon* arrays. Keep a list weapondefaults[] which undergoes some static->dynamic tweaks and then makes its way to the WEAPONx_XXX per-player gamevars.

- Rewrite the "clear background" routine in a no-brainer way instead of juggling around with rotatesprite(). Make it common to game+editor. Expose glRectd to glbuild. - Don't stop OSD text line drawing when encountering a non-printable char. Instead, treat it as space. - In OSD_SetTextMode(), don't use swaplong (which really swaps 32-bit ints) to swap pointers. Write an analogous "swapptr" instead. - When changing from/to OSD, don't inject a pause key. This *might* have been the cause of the reported pausing problems. - clean up the code... (Yes, this commit throws together too much stuff. I suck sometimes. :P)

Rotation-fixing happens for a couple of hard-coded statnums that presumably never move (DEFAULT, STANDABLE, FX, FALLER, LIGHT), but for actors it wouldn't make sense since the common case is that they do move. For this reason, bit 4 was introduced in r1934. The position of such useractors will not diverge due to error roundoff accumulation in rotating sectors (SE0, train).