[VEN]

Player's current weapon information (pPlayer->current) not always updated properly.

For exmaple if player is used a player_weaponstrip entity:

stock fm_strip_user_weapons(index) {
    new ent = fm_create_entity("player_weaponstrip")
    if (!pev_valid(ent))
        return 0

    dllfunc(DLLFunc_Spawn, ent)
    dllfunc(DLLFunc_Use, ent, index)
    engfunc(EngFunc_RemoveEntity, ent)

    return 1
}

pPlayer->current isn't updated, so get_user_weapon returned result is equal to previous pPlayer->current.

I'm not sure but i think that the problem is in emsg.cpp:

void Client_CurWeapon(void* mValue)
{
    static int iState;
    static int iId;
    
    switch (mState++)
    {
        case 0:
            iState = *(int*)mValue;
            break;
        case 1:
            if (!iState) break;
            iId = *(int*)mValue;
            break;
        case 2:
            if (!mPlayer) return;
            if (!iState || (iId < 1 || iId >= MAX_WEAPONS)) break;
            mPlayer->weapons[iId].clip = *(int*)mValue;
            mPlayer->current = iId;
            mPlayer->lastHit = mPlayer->lastTrace;
    }
}

I think the "case 2" code block should be changed to something like this:

            if (!mPlayer) return;
            if (!iState || (iId < 0 || iId >= MAX_WEAPONS)) break;
            mPlayer->current = iId;
            if (iId == 0) break;
            mPlayer->weapons[iId].clip = *(int*)mValue;
            mPlayer->lastHit = mPlayer->lastTrace;

[Freecode]

dont see anything wrong nor does your change make any sense to me

[VEN]

On every weapon change CurWeapon message is sent to the client.
It's even sent on player_weaponstrip entity use:

    MESSAGE_BEGIN( MSG_ONE, gmsgCurWeapon, NULL, pev );
        WRITE_BYTE(0);
        WRITE_BYTE(0);
        WRITE_BYTE(0);
    MESSAGE_END();

Since weapon index is 0, that event is skipped:

if (!iState || (iId < 1 || iId >= MAX_WEAPONS)) break;

and hence mPlayer->current isn't updated:

mPlayer->current = iId

[Freecode]

weapon 0 doesnt exist

[VEN]

Of course i know. But the message listed above can be intercepted and then mPlayer->current can be set to 0. The practical point of that is that this will prevent the incorrect get_user_weapon result.

If you still didn't get what i mean, my question is why get_user_weapon returns incorrect result after player_weaponstrip entity use?
Answer: because pPlayer->current isn't updated to 0.
Why? Probably because we igrone CurWeapon WeaponIndex == 0

if (!iState || (iId < 1 || iId >= MAX_WEAPONS)) break;

Probably your point that we can't set pPlayer->current to 0 for some reason? Well, the fact is still that there is a bug and it should be fixed. If i offered the incorrect solution or pointed to the incorrect amxmodx code section i apologize.

[Freecode]

i think its suppose to be like that. I dont think its ignoring, current is only set to your CURRENT weapon and since there is no weapon 0 current should not be set to 0. Otherwise your most likely breaking the whole point of current.

[VEN]

Yes, i see what you mean, but i hope we can have a certain workaround for get_user_weapon anyhow.

[Freecode]

get_user_weapon returns what you hold, and you cant hold weapon 0...because it doesnt exist. This is the way engine is suppose to work. I doubt there is anything to do about your problem, as you are not following engine standards.

[karlos]

if player has no weapon what value should be returned in your eyes ?
the ID of last used weapon ? then you need to rename the function to get_user_last_weapon and NOT get_user_weapon

[VEN]

fun's strip_user_weapons uses:

    void *_wpns = MF_PlayerPropAddr(params[1], Player_Weapons);
    *reinterpret_cast<int *>(MF_PlayerPropAddr(params[1], Player_CurrentWeapon)) = 0;

i believe that sets current to 0.

the ID of last used weapon ? then you need to rename the function to get_user_last_weapon and NOT get_user_weapon

I think you overlooked my point. Vice versa: get_user_weapon acts as get_user_last_weapon on player_weaponstrip entity use. Is that logical?