What changed in EDuke32, build by build

The attribute is set per tile from DEF: either texhitscanrange <begintile> <endtile> or tilefromtexture <tile> { ... texhitscan ... } (As a special case, the list may only contain "texhitscan", in which case the texture is not changed.) In passing, do some cleanup for "tilefromtexture" DEF parsing: the list tokens should now be accepted in any order, and errors don't appear in the middle of the list parsing.

Submersion only happens when the upper sector has lotag 1 and its vertical neighbor has lotag 2 (this can be useful to create non-submergible above-water sectors). Currently, only the player is handled. On the implementation side, the water handling code is split into P_Submerge() and P_Emerge() preparing the "teleportation", and P_FinishWaterChange() for the finalization.

sizeof(_prlight) differs across compilers, and besides, Polymer lights were not loaded anyway (they're restored live in-game later). The savegame version was not bumped, but only savegames made with builds where sizeof(_prlight)==261 (like synthesis) will stay compatible! Report and helpful debugging information by LeoD.

Instead of having 'noalreadyrefd' as a 'corruptcheck' subcommand. This way, TAB completion can be had.

This means that loadoldboard() now should also initialize Polymer map info after loading a v5/v6 map. Make the variables holding a file descriptor an int32_t instead of int16_t.

Note: CLASSIC_NONPOW2_YSIZE_WALLS is still not enabled in svn.

Previously, we used them in our builds for sprites, but MSVC builds don't have the ASM ported to MASM, and thus the workaround of ignoring the linking errors was not safe.

Also, add convenience wrapper script findmaps.sh for quickly searching for sprites/walls/sectors satisfying a certain condition in all map files under a given directory.

Requested auto-correction will place such sprites at their sector's first point.

This trims four 4k pages of code from the release EDuke32 build and one page from Mapster32.

When this happens, parallaxmodes other than 0 will be unavailable and silently draw as parallaxmode 0. This usually happens with extremely wide resolutions.

For reference, they are the following: - cache1d.c: suckcache() - build.c: compare_wall_coords() - make switch-invisible heuristic - Mapster32: old sprite search - Mapster32: manual z range - m32script: read/writearray, qgetsysstr - menus.c: savetemp()

1. EVENT_PREGAME 2. hardcoded stuff 3. EVENT_GAME One potential use is to examine values which are clobbered by commands such as ifhitweapon/ifwasweapon.

The spawn-time SE 17 setup uses nextsectorneighborz() which can fail (return -1) on some circumstances and would cause an OOB sector[] access.

This is mostly for debugging, since currently, Mapster32 restores sprites not in the same order as the original sprite index order. Also, expose this option from map2text.lua and mapdiff.sh.

This is really an artifact of overloading the sector fields for different uses. For copy-pasting TROR'd portions, the bug still persists.

A debug macro to test tints being stuck at the least intense non-zero value is provided in engine.c.

Noted by Nukey. See Duke3D 1.5 GAME.C line 4220.

Because rotating sectors "smoothly" by 1 BUILD ang will quickly accumulate roundoff error, this lets the user first do that as a preview and then use the manual angle rotation do carry it out for real.

This means that classic and the GL modes now look the same as far as e.g. hud_bgstretch or HUD weapons are concerned.

Use integer math; both args to setaspect are constant.

Instead of setting and resetting the [xy]dim globals, use them as locals in a block (shadowing the globals). Also, do some common subexpression elimination for clarity.

When shrunk, the player position is below the floor for some weird reason. Updatesectorz would set the sector to -1, and the view would not be drawn.

That is, use the view boundries instead of the whole screen.

This makes the code rather more readable in some places. Unlike the two preceding commits, this one is actually purely textual replacement.

The idiv instruction also signals an FPE when the resulting *signed* quotient overflows, so we simply use div instead.