[BUG] Hoonymode should disable powerups, but does not

[hemostx]

Describe the bug When setting /hoonymode, powerups remain on the map or respawn if coming from no powerups.

This suggests that it should toggle k_pow:

const char _1on1hm_um_init[] =
    "coop 0\n"                      // no coop
    "maxclients 2\n"                // duel = two players
    "k_maxclients 2\n"              // duel = two players
    "fraglimit 1\n"                 // hoonymode - every 1 frag we toggle spawns
    "timelimit 0\n"                 // hoonymode - timelimit 0
    "k_hoonymode 1\n"
    "k_hoonyrounds 12\n"            // first to seven
    "teamplay 0\n"                  // hurt yourself, no teammates here
    "deathmatch 3\n"                // weapons stay
    "k_overtime 1\n"                // overtime type = time based
    "k_exttime 3\n"                 // overtime 3mins
    "k_pow 0\n"                     // powerups
    "k_membercount 0\n"             // no efect in duel
    "k_lockmin 0\n"                 // no efect in duel
    "k_lockmax 0\n"                 // no efect in duel
    "k_mode 1\n"
;

However, k_pow 0 changes to k_pow 1 after /hoonymode and the powerups are enabled.

All PUs remains enabled on any map.

To Reproduce Caveat: Setting add_q_aerowalk 0 ensures the quad is removed when you do /aerowalk /hoonymode, so I would be careful with this special case when testing. In general, I'd go with other maps for repro first.

This works on any map with powerup(s):

1. /dm3 (it has so many powerups)
2. /1on1
3. Run around - no powerups!
4. /hoonymode
5. all powerups are back. check with rcon - k_pow is 1

Expected behavior Powerups should disappear in hoonymode - this appears to be the intended behaviour and it's a 1on1 mode, after all.

Environment(please complete the following information):

• Operation System: repro on linux/win
• Processor Architecture: most likely x86_64 and not relevant
• MVDSV Version: 0.34 to 0.36-dev
• KTX Version: reproduced across versions at least as old as 1.39 and as recent as 1.41-reki or 1.43-dev

[mikraj01]

I'm able to reproduce this issue with the nQuakesv package.

server log when calling /1on1. k_pow is 0 as expected

1 on 1 settings enabled by FU-rusti
execing configs/usermodes/default.cfg
execing configs/usermodes/1on1/default.cfg

matchtag not set
] k_pow
"k_pow" is "0"

and then /hoonymode. Note: 1on1/default.cfg is not executed. k_pow is 1

HoonyMode settings enabled by FU-rusti
execing configs/usermodes/default.cfg

matchtag not set
] k_pow
"k_pow" is "1"

When I look at default configs I can see following k_pow values. usermodes/default.cfg

set k_pow 1

usermodes/1on1/default.cfg

set k_pow 0

[mushis]

when is this getting attention? its annoying and should be an easy fix.

[mushis]

I guess this means its not a server config issue, but rather a code issue.

[tcsabina]

Upon activating /hooneymode, first we apply the content of '_1on1hm_um_init', followed by executing ktx/configs/usermodes/default.cfg.

This default.cfg in the nquakesv installer contains 'k_pow 1' (see https://github.com/nQuake/distfiles/blob/master/sv-configs/ktx/configs/usermodes/default.cfg).

I think the easiest fix is to make the change in the nquakesv, and remove k_pow from the default.cfg, instead of diving into the execution order of these config files upon executing a usermode command like /hooneymode, or /1on1. I guess every mode where powerups are required (2on2, 4on4, etc) have 'k_pow 1' defined in the dedicated 'default.cfg' file, so removing it from the top-level default.cfg will do no harm.

[tcsabina]

I've just made a PR to update the default.cfg in the nquakesv package. Let's see...