Commit graph

4010 commits

Author SHA1 Message Date
Monster Iestyn
db7d217199 Merge branch 'exit-your-server-desync-fix' into 'master'
Exit your server desync fix

This branch fixes this bug: https://mb.srb2.org/showthread.php?t=41811

Here's an explanation of how this bug is actually happening:
* Pressing Enter with "Start" selected calls `D_MapChange` to make a map change command. (Note that part of it is actually delayed because it's called from a menu, but this isn't relevant to the bug)
* `D_MapChange` itself calls `SV_StartServer` to start up the server, if you're supposed to be the server player.
* Pressing Esc while starting up a server like shown in the gif causes the server to be cancelled, and go back to the title screen.
* HOWEVER, since we're still in `D_MapChange`, the map change command is made anyway.
* Since we're not in a netgame as expected, no random seed is added, and the map change command has a total size of 9 bytes (1 for the id + 8 for the actual data). This is then stored in the net command buffer. The random seed would add another 4 bytes if it WAS added, note.
* Next time you start a server, and let it work as normal, two more net commands are added: an "add player" command (size 3), and a map change command (size 13, including random seed this time). Your net command buffer now has a total size of **25** (9 + 3 + 13).
* This net command buffer is then sent as a `PT_TEXTCMD` packet to yourself.
* SRB2 recieves the `PT_TEXTCMD` you sent to yourself, and some time later, tries to run the net commands stored in it.
* The first map command is read as if it's size 13 rather than size 9 (it isn't), mistakeningly thinking a 4-byte random seed is stored (since you ARE in a netgame now). This means your buffer reads what it thinks is the next net command starting from position **[14]** in the buffer.
* Bad luck, it gets an id of **0**, which doesn't correspond with any net command!
* This message is then printed in the console:

> WARNING: Got unknown net command [14]=0 (max 25)

* The game subsequently kicks you out the server with a "desynch" message.

tl;dr The game doesn't correctly cancel the map change when you cancel the server with Esc like that. This is not a good thing.

See merge request !188
2017-05-25 20:19:27 -04:00
Monster Iestyn
437780d503 Add missing break for MT_EGGCAPSULE's switch case in P_SpawnMobj (sorry!) 2017-05-25 18:40:48 +01:00
Monster Iestyn
a23e9bc32b Fix size_t printing 2017-05-25 18:10:20 +01:00
toasterbabe
fe9ebbd1ca Files which are unimportant are not sent in the fileneeded list.
This removes exactly one feature in exchange for a greater number of files to be loaded. I asked around, and people don't actually like what it currently does (load unimportant server-side files if you can findfile it locally), so.

Can be reverted if requested.
2017-05-25 17:58:32 +01:00
Monster Iestyn
58236af6f7 Tweak to D_MapChange: if you failed to start a server, DON'T send a map change command 2017-05-25 16:55:59 +01:00
toasterbabe
47e171250f * Prevent joining a server if you have too many files loaded to add the remainder.
* Made the check in Got_RequestAddfilecmd more comprehensive, since I might as well. Just something to tweak a little later in internal.
2017-05-25 16:06:39 +01:00
toasterbabe
5c302d7ffc Partial implementation of fix for Got_RequestAddfilecmd (the other half - the limitation on the size of the filesneeded section of the serverinfo packet - will be applied in internal.) 2017-05-25 15:34:21 +01:00
toasterbabe
2aa1215716 Fix a thing where the host adds a WAD and you have too many WADs loaded to add it. 2017-05-25 15:22:32 +01:00
toasterbabe
9fcf40b5a3 woops 2017-05-24 22:36:59 +01:00
toasterbabe
e2f6366292 Minimum fill for temperature gauge if you have any wads loaded at all. 2017-05-24 22:05:41 +01:00
toasterbabe
c2705b4662 * Make wads that don't modify the game not count towards the internal packet size tally in w_wad.c, like they wouldn't in d_netfil.c.
* Now that it's consistent, removed the redundant packet size tally check in d_netfil.c.
2017-05-24 21:45:03 +01:00
toasterbabe
74c4ad4bad Minor refactor, plus cutting out a potential bug of setting dir_on[] to -1. 2017-05-24 20:55:35 +01:00
toasterbabe
e41976fb05 https://cdn.discordapp.com/attachments/293238104096112641/317023848493088779/srb20016.png
* Visual Indicator of modifiedgame/savemoddata.
* Other assorted improvements to layout.
2017-05-24 20:55:05 +01:00
Monster Iestyn
c92aaa74a4 I think PT_CLIENT2MIS was meant to do this too, probably 2017-05-23 18:46:34 +01:00
Monster Iestyn
7147e0fcaf Improve previous PT_TEXTCMD/PT_TEXTCMD2 check, add another one to ignore zero size textcmds! 2017-05-23 17:35:32 +01:00
Monster Iestyn
0b0b738a6f Don't allow non-servers to receive PT_RESYNCHGET 2017-05-23 16:34:56 +01:00
Monster Iestyn
ff1cc07a1d Implemented toaster's suggestion to make a macro here 2017-05-23 16:29:09 +01:00
Monster Iestyn
9e3bdc5ee2 Prevent bad PT_TEXTCMD/PT_TEXTCMD2 textcmd[0] sizes? 2017-05-23 16:13:42 +01:00
Monster Iestyn
fdfd6e1c0b Whoops don't do that for PT_REQUESTFILE 2017-05-22 22:20:08 +01:00
Monster Iestyn
c70839334e Added a bunch of missing checks to prevent non-server players from sending other non-server players stuff 2017-05-22 22:17:51 +01:00
Monster Iestyn
3784d765e4 Remove extra whitespace I added 2017-05-22 20:47:38 +01:00
Monster Iestyn
28444c12dd Some more things I overlooked in this fix 2017-05-22 16:54:39 +01:00
Monster Iestyn
247ed56e17 Don't allow nonsense PT_ASKINFOVIAMS packets, nor any at all if offline 2017-05-22 16:44:50 +01:00
Monster Iestyn
6062223eb5 Merge branch 'master' into next 2017-05-22 13:23:46 +01:00
Monster Iestyn
c23cf25f65 Merge branch 'dedicated-bonuses-fix' into 'master'
Dedicated bonuses fix

Fixes the issue reported here: https://mb.srb2.org/showthread.php?t=42530

What it turns out is going on is that making the game bail out in the middle of Y_StartIntermission if you're a dedicated server's host prevents the game from awarding the players bonuses in coop mode. Therefore it's not just special stage bonuses, but ANY kind of bonuses that can cause desyncs if someone loses all their lives.

This can be merged to master since it's a change only for dedicated server hosts, and should otherwise be compatible with 2.1.18.

See merge request !186
2017-05-22 07:15:50 -04:00
Monster Iestyn
5aeb4b5919 Revert "Loop through rebound packets until you found a good one or ran out of them"
On second thought, this was probably unnecessary anyway.

This reverts commit 1078a642df.
2017-05-20 21:29:05 +01:00
Monster Iestyn
448014097e Merge branch 'endgame-code-refactor' into 'master'
Endgame code refactor

This is mostly just some code cleanup and moving about:
* `Y_EndGame` was moved from from y_inter.c/h to g_game.c/h, and renamed to `G_EndGame`. Frankly the old place that function lived didn't make sense, since it has nothing to do with tally screens directly.
* Most of `Y_FollowIntermission`'s code was removed, because `G_AfterIntermission` does it all already.
* And lastly, `G_ExitLevel` uses `HU_ClearCEcho()` instead of `HU_CEcho("")` to clear CEchos from the screen.

The only visible changes in-game from the above should be just that log.txt stops adding empty new lines when you complete a level, and CEchos will always be cleared after tally screens when a level goes to credits/evaluation next.

See merge request !98
2017-05-20 16:23:21 -04:00
Monster Iestyn
284e539c66 Don't bail out in Y_StartIntermission in dedicated mode, this causes the game not to add on score bonuses for players from the server's view of things! 2017-05-19 16:31:23 +01:00
Monster Iestyn
1078a642df Loop through rebound packets until you found a good one or ran out of them 2017-05-19 11:52:50 +01:00
Monster Iestyn
5a1fc6098e Attempts to prevent stupid stuff from happening 2017-05-19 11:25:28 +01:00
Nevur
19c46f3732 More PK3 work.
-Removed code duplicity on map resource reading.
-Fixed all known PK3-related bugs so far except for the ones mentioned below.
Issues:
-Blockmap and reject matrix still aren't loaded yet when using a map WAD.
-Palettes and colormaps aren't loaded (by extension, TRANS tables might not work either).
2017-05-18 21:13:18 +02:00
Monster Iestyn
229e75ab8b Removed all code in Y_FollowIntermission that's already handled in G_AfterIntermission
Only real difference here is that CEcho messages will always be cleared when going to credits/evaluation, but that's hardly a loss tbh.
2017-05-18 17:23:19 +01:00
Monster Iestyn
d2313ea32b G_ExitLevel tweak: Use HU_ClearCEcho() instead of HU_DoCEcho(""), the latter causes an empty line to appear in log.txt 2017-05-18 17:10:44 +01:00
Monster Iestyn
bc79365cf4 Moved Y_EndGame from y_inter.c/h to g_game.c/h, renamed it to G_EndGame 2017-05-18 16:54:58 +01:00
Monster Iestyn
fd1efff338 Added the META_ACTION userdata type to Lua, created getActionName() to return string name
SRB2's built-in action types will now all be pushed as META_ACTION, state_t.action can be assigned META_ACTION and return it

getUserdataType returns "action" for META_ACTION variables
2017-05-17 20:36:40 +01:00
Monster Iestyn
21d29c8550 Merge branch 'public_next' 2017-05-17 16:19:01 +01:00
Monster Iestyn
0cf42d8bb8 Merge branch 'next' into public_next 2017-05-17 16:18:42 +01:00
Monster Iestyn
1781ccf4ca Merge branch 'master' into next 2017-05-17 16:16:50 +01:00
Monster Iestyn
b564c8ab9a Merge branch 'sdl-fixes' into 'master'
SDL fixes and cleanup

Mostly cleanup tbh. I noticed a few things that bugged me when examining SDL's code recently, thought I might as well fix them up a bit.

These changes definitely does not affect netplay, so this can be merged to master just fine.

See merge request !183
2017-05-16 23:09:23 -04:00
toasterbabe
593416328a String addition to the addfile menu suggesting to visit the modifications page! No change to actual functionality, just secondary utility. 2017-05-16 23:43:21 +01:00
Monster Iestyn
551ed797db Merge branch 'jumping_fixes' into 'master'
Nojumpspin fix

MI can attest to this being a problem that's now solved.

To test, check out root/!LatestSRB2Files/srb2win_branch_jumpfixes.exe and root/MonsterIestyn/robohood.wad.

See merge request !95
2017-05-16 17:02:20 -04:00
Nevur
d4c324eb30 PK3 stuff again.
-Now can load map WAD files inside Maps/ directory, but they must only contain map data explicitly!

Known problems:
-There's an issue causing a crash with palettes and colormaps in PK3s.
-SS_START and SS_END don't work now. Will check later.
-Standalone lumps for maps in the Maps/ folder don't work anymore; perhaps I should keep that functionality?

Notes:
-It's now a mashup of something dumb that I wanted to do and yet piling hacks again.
-A lot of code duplicity with map lump loading functions.
2017-05-16 21:10:02 +02:00
toasterbabe
5f8cfb47e4 Merge branch 'master' of http://git.magicalgirl.moe/STJr/SRB2Internal.git into addfile_menu 2017-05-16 19:54:54 +01:00
toasterbabe
87f32abc24 i have no idea whether this will fix the garbage or whether it'll get broken again, but here's an attempted fix for VAda's name 2017-05-16 14:11:10 +01:00
Monster Iestyn
8454e75e3c Merge branch 'extracolormaps-fix' into 'next'
R_InitExtraColormaps fix

This fixes the `R_InitExtraColormaps` function so that it correctly realises there are no "extra colormaps" actually in SRB2's files, rather than telling you there are 6 extra colormaps on game startup (or however many WAD files total you have added).

For those who don't know what the function does, it searches for C_START/C_END markers, and if they exist, adds any lumps inbetween to a list of extra colormaps internally. They are never used by the game though, since we last supported colormaps of that kind so long ago I don't think anyone remembers when anymore (definitely more than a decade at least).

Merging to next since I don't know if this would cause any netplay issues or not tbh.

See merge request !182
2017-05-15 17:14:47 -04:00
Monster Iestyn
17a06dd6c4 I_GetConsoleEvents: Split KEY_EVENT code into a function of its own, like with I_GetEvent's event types
One benefit of this is that event_t data need only be created if KEY_EVENT is found, since the other event types never do anything anyway
2017-05-15 18:39:59 +01:00
Monster Iestyn
87085f2475 SDLSetMode: merge wasfullscreen/windowed mode code into one block 2017-05-15 16:35:32 +01:00
Monster Iestyn
a340f2c8ed Impl_CreateWindow: re-use "flags" for SDL_RENDERER_* flags, remove unnecessary curly braces 2017-05-15 16:19:28 +01:00
Monster Iestyn
4979ab6b8e Not really important or anything, but checking render_soft == rendermode rather than rendermode == render_soft always bugged me. And it's not consistent with the rest of the source code (or at least most of it) anyway. 2017-05-15 15:43:31 +01:00
Monster Iestyn
d1bbd1261e VID_SetMode: SDLSetMode should use vid.width/vid.height, not windowedModes[modeNum].
If modenum was < 0 or >= MAXWINMODES, that would make windowedModes[modeNum] be out of bounds and possibly crash the game.
2017-05-15 15:36:51 +01:00