Time  Nick      Message
08:18 ANAND     Trivial bug-fix to blocker #9022: #9025
08:18 ShadowBot https://github.com/minetest/minetest/issues/9022 -- Using 3d_armor mod with "Enable Damage" inactive makes the world mostly refuse to load
08:18 ShadowBot https://github.com/minetest/minetest/issues/9025 -- Revert Client::sendPlayerPos optimization (part of 81c7f0a) by ClobberXD
12:12 nrzkt     oh nice we are in feature freeze, yeah i'm sleeping
12:12 nrzkt     we are releasing 5.1.0 ?
12:13 sfan5     there's 2 remaining issues, then 5.1 can be released yes
12:23 nrzkt     oh it's nice to hear, the quality process is nice
12:23 nrzkt     what amazing feature is present on this release ?
12:24 sfan5     many formspec changes i didn't keep track of
12:26 sfan5     knockback, tool wear for pvp
13:09 ANAND     Approval required for trivial blocker PR #9025
13:09 ShadowBot https://github.com/minetest/minetest/issues/9025 -- Revert Client::sendPlayerPos optimization (part of 81c7f0a) by ClobberXD
13:09 ANAND     It has been tested by two people, and has been confirmed to work
16:45 sfan5     ANAND: do you think its viable to fix the sendplayerpos/autoforward disable stuff after release?
16:45 sfan5     because if not I'd rather just revert the whole thing (and leave autoforward "broken")
16:47 Krock     turn off autoforward on death
16:51 sfan5     that sounds like the better solution but I haven't looked into whether its really that simple
16:55 Krock     would be doable on HP updates, or content_cao on_step if necessary
16:57 Krock     if (m_activeobjects_received && player->isDead() && !player->isImmortal())
16:58 Krock     weird. isDead already checks for that
16:58 sfan5     that would suffer from the same problem
16:58 Krock     indeed. caused by removing the immortal armor group ...
16:58 Krock     how about armor doesn't do such stuff?
17:00 sfan5     the api could ensure that the immortal group can not be removed
17:00 Krock     well, that's not quite the idea of this armor group. players should be set freely to immortal and back
17:01 Krock     and so that immortal defaults to whether damage is enabled or disabled on the server
17:01 sfan5     I was under the impression that enabling damage does not work per-player
17:01 Krock     now it should work
17:02 Krock     didn't that that through. *sets up a testing mod*
17:03 sfan5     that leaves the initial problem though, existing mods break the entire game in hard-to-debug ways if they are not careful
17:03 sfan5     (the problem here is not that mods *can* do this, it's that 3d armor causes this *by default*)
17:04 Krock     question now is who decides about such modding changes - mods or Minetest
17:07 Krock     per-player damage works, but the builtin statbars aren't updated
17:08 Krock     * they are only updated when damage is enabled server-wide
17:08 sfan5     there are probably a few places where the engine checks enable_damage
17:08 sfan5     autoforward is a toggleable key right?
17:08 Krock     yes
17:09 sfan5     then why not this: if a hp update is received and the player dies as result (and isn't immortal), toggle the autofoward off
17:09 Krock     that's the idea
17:09 sfan5     no "optimization" in SendPlayerPos that could break, and the "sent active objects after init2" is not needed either
17:09 sfan5     oh, I thought you just wanted to modify the existing condition
17:10 Krock     I wonder why this wasn't done in first place. Maybe paramat mentioned something, but I'm not sure
17:11 sfan5     if we want to pursue this simpler approach after release, it'd make sense to revert all of the related commits for now (instead of just commenting a check out)
17:11 sfan5     do you agree?
17:12 Krock     I don't see a need for revert. Updating the autoforward move and uncommenting those lines should suffice
17:12 Krock     will propose a patch in an hour or so
17:21 paramat   as long as autoforward is still usable when damage is disabled i'm happy. dead players creeping forwards if they were using autoforward seems not a significant bug
17:25 * nepugia wonders whether dead players should even be allowed to move in any case
17:54 Krock     Minetest should actually send death messages when hp == 0 and immortal flag isn't set
17:55 Krock     would solve this error too
18:28 sfan5     so for 5.1 we'll be merging #9025 then, right?
18:28 ShadowBot https://github.com/minetest/minetest/issues/9025 -- Revert Client::sendPlayerPos optimization (part of 81c7f0a) by ClobberXD