Add optional parameter for *gettimestr

[AnnieRuru]

Pull Request Prelude

• I have followed proper Hercules code styling.
• I have read and understood the contribution guidelines before making this PR.
• I am aware that this PR will be closed if the above-mentioned criteria are not fulfilled.

Issues addressed

we have *getcalendartime script command, but why our *gettimestr isn't update for it?

Changes Proposed

Add optional parameter for *gettimestr

Tested with

prontera,155,185,5	script	sdkfjhsdkf	1_F_MARIA,{
	mes(gettimestr("%Y-%m/%d %H:%M:%S", 21));
	mes(gettimestr("%Y-%m/%d %H:%M:%S", 21, getcalendartime(0,0)));
	close;
}

which is stated in the documentation

Affected Branches

• Master

Known Issues and TODO List

none
in fact this is mostly a copy-paste from rathena repo

[HerculesWSAPI]

This change is 

[AnnieRuru]

curious, isn't that after using Malloc should always use Free ?
in this script command, has aMalloc but doesn't have aFree ... isn't that will cause memory leak ?

[4144]

if string next pushed to script engine as string and not cont, script engine will free it if need remove from memory

[AnnieRuru]

hmm ... for some reason this isn't merge ...

[Emistry]

btw, cant we make the script command auto calculate the max length parameter? by default its just the same length of the format parameter. Aren't this max length parameter currently look redundant and not necessary?

[AnnieRuru]

hmm ... the person who made this script command probably has die hard habit from C language
char message[CHAT_SIZE_MAX];

then how do you want to modify it ? this is actually a copy paste from rathena, literally
https://github.com/rathena/rathena/blob/78eed02273414a4332f9c088ad7e795a2903ea5f/doc/script_commands.txt#L3155

if we make the <max length> optional parameter, that would become ...
*gettimestr(<format string>{, <max length>{, <timestamp>})
means still have to input the max length ...
or ...
*gettimestr(<format string>{, <max length>})
*gettimestr(<format string>{, <timestamp>})
both fields are INT type, its not like integer/string that we could switch it easily

changing this behavior silently is big no-no ...
try do a search in npc folder ... some scripts already using it

[Emistry]

some script already using it...

that doesn't meant we aren't allow to change it or update it? Both rA/Herc and any other emulators has done it countless time. So its not a restriction.

what I meant in previous post are totally remove the max length field.
End Result: *gettimestr(<format string>{, <timestamp>})

For example: changing the code to auto determine the length ?

fmtstr=script_getstr(st,2);
maxlen=script_getnum(st,3);

into

fmtstr=script_getstr(st,2);
maxlen=(int)strlen(fmtstr);

if worry about the length aren't enough for that specific format, then just use the max length allowed.

changing this behavior silently is big no-no ...
try do a search in npc folder ... some scripts already using it

So far only 2 scripts are using it. That's why changelog/commit history exists for a reason. For user to read/keep track of what has been changed. Every milestone release come with a changelog for each release.

[AnnieRuru]

I'm not doing that, because your format won't throw error on old scripts
old script that still uses gettimestr("%H%M%S",7); this will print 080007 on your format

[AnnieRuru]

just noticed the documentation says strfmtime
when I ask google, says -> Do you mean strftime ?
LOL