View source for Template talk:O

== Empirically determining !activator and !caller  ==

It's easy enough to guess oftentimes, but I was wondering if there was a more error-proof way of determining them (be it reading source code or setting up an I/O chain).<br/>— [[User:SirYodaJedi|SirYodaJedi]] ([[User_talk:SirYodaJedi|talk]]) 09:23, 24 April 2025 (PDT)
:Reading the source code is mostly error proof by just checking what parameters is <code>.FireOutput</code> or <code>.Set</code> called with for example. Checking out activator and caller in IO can be done like this

:Example with checking activator and caller of OnTakeDamage output of prop_physics. I aimed at it and used !picker to target it and addoutput to it first in this case. Then paused the IO. Shot the prop_physics and checked with {{cmd|dumpeventqueue}} command.
<dd><pre>
] ent_fire !picker addoutput "OnTakeDamage whatever"
] ent_pause 
Pausing entity I/O events
] dumpeventqueue 
Dumping event queue. Current time is: 941.23
   (940.83) Target: 'whatever', Input: 'Use', Parameter ''. Activator: 'Nescius', Caller 'prop_physics'.  
</pre></dd>

:Or also in games with vscript can use something like <code>ent_fire !picker AddOutput "OnSomething !caller:RunScriptCode:printl(self)"</code> which will print the caller

:Example of checking OnUser1 output.
<dd><pre>

] ent_fire !picker addoutput "OnUser1 whatever"
] ent_pause 
Pausing entity I/O events
] ent_fire !picker fireuser1
] dumpeventqueue 
Dumping event queue. Current time is: 955.63
   (955.63) Target: '!picker', Input: 'fireuser1', Parameter ''. Activator: 'Nescius', Caller 'Nescius'.  
Finished dump.
] ent_step 1
] dumpeventqueue 
Dumping event queue. Current time is: 955.63
   (955.63) Target: 'whatever', Input: 'Use', Parameter ''. Activator: 'Nescius', Caller 'prop_physics'.  
Finished dump.
</pre>
:--[[User:Nescius|Nescius]] ([[User talk:Nescius|talk]]) 09:55, 24 April 2025 (PDT)

::{{quote|Reading the source code is mostly error proof by just checking what parameters is .FireOutput or .Set called with for example.}}
::Okay, if I'm reading CBaseEntityOutput::FireOutput() correctly, it looks like it's .FireOutput(value,!activator,!caller,delay), with value and delay being optional. If I remember C correctly, "this" should be !self.<br/>— [[User:SirYodaJedi|SirYodaJedi]] ([[User_talk:SirYodaJedi|talk]]) 11:08, 24 April 2025 (PDT)
:::This code for example
:::<syntaxhighlight lang=c++>
void CBaseEntity::InputFireUser1( inputdata_t& inputdata )
{
	m_OnUser1.FireOutput( inputdata.pActivator, this );
}
</syntaxhighlight>
:::Will cause !activator to be the activator of the FireUser1 input that was sent and !caller and !self to be equal to the this entity (the one that FireUser1 was sent to). Check also stuff I just mentioned here [[Talk:Targetname#!caller_is_weirdly_incosistent]] --[[User:Nescius|Nescius]] ([[User talk:Nescius|talk]]) 11:34, 24 April 2025 (PDT)
::::Yeah, I saw that post. A lot of entities seem to use <code>m_Whatever.FireOutput(this,this)</code> even when they shouldn't.<br/>— [[User:SirYodaJedi|SirYodaJedi]] ([[User_talk:SirYodaJedi|talk]]) 12:06, 24 April 2025 (PDT)

== woifurawiufiuwaofiwunaw ==

This still feels kinda confusing at the moment. Because the !activator/!caller/!self currently only specifies how 'Target entity' behaves. I am not sure how to make this clear and ensure that someone won't think it also applies to some random input's parameters. --[[User:Nescius|Nescius]] ([[User talk:Nescius|talk]]) 14:03, 25 April 2025 (PDT)