split pawn codes

[redex]

hi.
i am working on a sa-mp gamemode project, its going to take about +50k code line, i was wondering is there any way to split the code, for exp put the JOBS related codes to another file and include it , so if i need to edit something about jobs i dont need to check hell a lot code lines to fix a problem!

how can i do that? is it good?
can it cause bad effects like lags or some other bad effects to server?
if u know anything please let me know, thanks!

[Kwarde]

Well off course! And you absolutely should do that! You're also already doing it in fact when including includes.. Thus use #include to include different files.
#include <someInclude> would attempt to include "PATH_TO_INCLUDES_DIRECTORY/someInclude.inc"
#include "someInclude.inc" or #include "someInclude" would attempt to include "PATH_RELATIVE_FILE/someInclude.inc" (you could use different extensions. Eg. gamemode sf-cnr uses .pwn for all includes -- don't, though. ".inc" is the correct file extension)

For example:

./gamemodes/gamemode.pwn

Code:

#include <a_samp>

#include "src/assembly.inc"

public OnGameModeInit()
{
    print("Final OnGameModeInit() call!");
    return 1;
}

./gamemodes/src/assembly.inc

Code:

#include "src/init.inc"
#include "src/server/mysql.inc"

./gamemodes/src/init.inc

Code:

#include <YSI_Coding\y_hooks>

hook OnGameModeInit()
{
    print("First OnGameModeInit() call! Start loading gamemode");
   return 1;
}

./gamemodes/src/server/mysql.inc

Code:

#include <YSI_Coding\y_hooks>

hook OnGameModeInit()
{
    print("Second OnGameModeInit() call! Setting up MySQL connection");
    return 1;
}

Etc..

There are probably 'rules' around as how to name files (eg. using 'header.inc' and 'impl.inc' files) but I'm too noob for that :-)
Here are a few gamemodes that do use multiple files in case you need examples. 
https://github.com/PatrickGTR/gta-open
https://github.com/MaxAndolini/NG-RP
https://github.com/zeelorenc/sf-cnr

Ultimately this will all be combined into one file (by the pre-processor) - including the plugins. It has no effect on the performance of the server.
If you're curious and want to see that for yourself, compile with option "-l" (either put #pragma option -l in your script (do use the latest compiler) or append it to pawncc args).
This will show the output of the stage after the pre-processor's output and before compiling it to assembler code/binary file (amx) in a .asm file. This for example (note that this doesn't make much sense):

./script.pwn

Code:

#include "myInclude.inc"

#pragma option -l

#define MY_CONSTANT 5

main()
{
    print("MY_CONSTANT IS "#MY_CONSTANT);
}

./myInclude.inc

Code:

native print(const string[]);

Would output something like this: (without "#file" and "#line" outputs: when using -l it adds that to the script to show what's where)

Code:

native print(const string[]);

main()
{
    print("MY_CONSTANT IS "#5);
}

[Pinch]

https://github.com/Southclaws/sampctl/wiki/Modern-Pawn
https://github.com/TradeWars/gamemode/blob/master/docs/Style-Guide.md

[redex]

thank u guys it was really helpful +rep