Simple point script

[Fraysa]

You shouldn't use ternary for such a long thing. Thats when arrays (as mentioned above), ifs, switch-case come to the script.

Ternary in my opinion should be used for short terms as its very ineffeicient and very hard for the reader to get whats going on when its huge.

Fraysa's method is what I would use for an instance.

I believe the compiler would optimize the same for ternary or if statements. The only disadvantage is readability, but that wouldn't really matter because the majority of this community doesn't care about that.

Please base your statements with proofs or show why are you stating that.

[NmZero]

I believe the compiler would optimize the same for ternary or if statements. The only disadvantage is readability, but that wouldn't really matter because the majority of this community doesn't care about that.

Please base your statements with proofs or show why are you stating that.

I think he meant to say:
"I dont personally like using ternary
EXCEPT
when using long sentecnes (or w/e)
for example:"

options += "\r\n#L"+x+"##i"+ITEMS[s][x][0]+"# "+cr()+"#t"+ITEMS[s][x][0]+"##k for "+(cm.getPlayer().isGM() === false ? ""+ITEMS[s][x][1]+" EP" : ""+cr()+"FREE")+"#l";

He never said "ternary is bad"

[Fraysa]

i think he meant to say:
"i dont personally like using ternary
except
when using long sentecnes (or w/e)
for example:"

options += "\r\n#l"+x+"##i"+items[s][x][0]+"# "+cr()+"#t"+items[s][x][0]+"##k for "+(cm.getplayer().isgm() === false ? ""+items[s][x][1]+" ep" : ""+cr()+"free")+"#l";

he never said "ternary is bad"

it's very inefficient

#tooshort

[Eric]

I believe the compiler would optimize the same for ternary or if statements. The only disadvantage is readability, but that wouldn't really matter because the majority of this community doesn't care about that.

Please base your statements with proofs or show why are you stating that.

Oh, good one. I completely missed that part but yes, I had researched this myself and it has been said numerous times that ternary operations don't differ in performance from an if-else statement. Which is why in my eyes, if a ternary operation and a if-else statement are equal in performance, and a ternary is shorter than doing several if-else-if-else statements, I will result in using a ternary over if statements and even switch (however, I have not checked performance between switch and if-else, though I have heard that switch is faster than if-else).

[xStr0nGx]

#tooshort

If one want to become a greater coder, his codes should function and be readable

as the end-users (the developers who may receive jobs from others on resuming this development) will find it easier

and be much more thankful than the codes that werent coded this way.


But yeah, every developer got his own conventions. However if he is to release something, I personally would

expect a readable code and not a complex one (massive lines of code without side-comments).

[Sunnyday]

Good job m8

[ALotOfPosts]

Oh, good one. I completely missed that part but yes, I had researched this myself and it has been said numerous times that ternary operations don't differ in performance from an if-else statement. Which is why in my eyes, if a ternary operation and a if-else statement are equal in performance, and a ternary is shorter than doing several if-else-if-else statements, I will result in using a ternary over if statements and even switch (however, I have not checked performance between switch and if-else, though I have heard that switch is faster than if-else).

Switch is the same, at least in old Javas. At least the compiler is supposed to implement it the same. There have been some tests that show that switch is faster, but they seem pretty inconclusive. There's some interesting ways to set up tests that it would give you data that you want. See this for some weirdness.

As a side note about readable code and shorter code:
Please stop thinking shorter code is better and ternary operations are that good.

Ternary is pretty terrible if they consist of long clauses. Ternary is extremely bad looking if you chain them together. I'm not saying I don't use them, because I have used them plenty of times. It does the same thing as an if statement but at least if statements are more clear and readable. Take the example earlier. It takes a much longer time to figure out what's going on in

options += "\r\n#l"+x+"##i"+items[s][x][0]+"# "+cr()+"#t"+items[s][x][0]+"##k for "+(cm.getplayer().isgm() === false ? ""+items[s][x][1]+" ep" : ""+cr()+"free")+"#l";

than if you did

options += "\r\n#l"+x+"##i"+items[s][x][0]+"# "+cr()+"#t"+items[s][x][0]+"##k for ";
if (!c.getPlayer().isGM()) {
     options += items[s][x][1]+" ep";
} else {
     options += cr()+"free")+"#l";
}

There's no performance difference in either. One of them is a lot easier to understand and debug. Long ternary operations are among the worst to debug along with undescriptive variable names and uncommented code chunks. Personally I've never seen it used in any production/deployed code because it wouldn't pass any reviews. There are many times when shorter isn't better. Compilers these days do very well in optimization and so the user should never try to optimize by him/herself.

You can do things like x << 1 instead of x * 2, to change a function call into a bit manipulation to "optimize" your code, but the compiler will already do that. Even things like int x = 5 + 5; The compiler will probably evaluate that during compile time.

Anyway going back to the branch issue from the url above, as you noticed that sorting the array makes it a lot faster, but sorting the array adds code lines. Obviously sometimes adding additional code lines makes the code run faster. Shorter code almost never really implies better code, especially when algorithms are around Sometimes a trivial method could have a runtime complexity of n^2 (pure for loops), but then when you think about clever ways to approach a problem, you can bring a problem down to log n, exponentially better.

I know @xStr0nGx is hated around here, but I'd have to agree with him. Sometimes when you release things you want to make things readable for everyone. You've all seen XxOsirisxX's one line for loops. They're cool in concept, but not really for practical use.

[NmZero]

I know @xStr0nGx is hated around here, but I'd have to agree with him. Sometimes when you release things you want to make things readable for everyone. You've all seen XxOsirisxX's one line for loops. They're cool in concept, but not really for practical use.

I have to agree as well, but when you are making new things yourself you tend to see them more clearly; since you have written the line/code yourself. This can sometimes make it hard to detect what is normally readable for the public in my opinion.
I don't know if the following example also applies for higher 'skill level' but here is what I mean:

this is how somebody who never heard of loops would write a npc script that sells things:

function action(m, t, s) {
//mode blabla
if (status === 0) {
  cm.sendSimple("#L1# selection 1#l\r\n#L2#selection 2#l etc etc etc");
} else if (status === 1) {
  if (selection === 1) {
    cm.warp(91000001);
  else if (selection === 2) {
    cm.warp(91000002);
// etc etc etc
}
} 

here is how anybody who heard of loops would make this:

function action(m, t, s) {
//mode blabla
if (status === 0) {
  var text = "blabla\r\n";
  for (var i = 1; i < 10; i++)
    text += "#L"+i+"#selection "+i+"#l";
} else if (status === 1) {
  var teleport = 91000000 + i;
  cm.warp(teleport);
} 

If a person who never heard of loops opened both scripts, which one would be more readable for him?
script 1 ofcourse

anyway, I get that:

options += "\r\n#l"+x+"##i"+items[s][x][0]+"# "+cr()+"#t"+items[s][x][0]+"##k for "+(cm.getplayer().isgm() === false ? ""+items[s][x][1]+" ep" : ""+cr()+"free")+"#l"; 

is a little bit unreadable for everybody on first sight.
All I'm saying is that wether its readable is also an opinion.

[ALotOfPosts]

That is true but a better comparison is doing:

for (var i = 1; i < 10; i++, text += "#L"+i+"#selection "+i+"#l");

Also starting at 1 is an interesting choice if you're doing a less than in the second part. Some people may be confused because they wouldn't be sure if you want to go from 0-10, 1-10, 1-9, but that's not really an issue.

For the original for loop, I'm pretty sure anyone would be ok with it if there was a comment added or if there was an example of what was supposed to be there. The code is clear enough. There really is just no reason to do cond1 ? a : cond2? b : cond3 ? c : cond4 ? d : cond5? e : f.

There's also the SuperLol's status changer, which is status = 2 * mode - 1; for status++ if mode = 1, status-- if mode = 0. This just gets interesting when someone packet edits mode to be something else. You can probably skip a few statuses and if key variables are set during those times, it might be exploitable.


function cr() {
var colors = ["b", "d", "g", "r", "k"];
return "#" + colors[(Math.floor(Math.random() * 5))];
}

The way you are using cr() is a bit excessive. You may want to look into making a function that takes a string as an argument and returns a randomized color string. Probably a syntax error somewhere here but you get the idea.


function cr(strings) {
var text = "";
var colors = ["b", "d", "g", "r", "k"];
for (var i = 0; i < strings.length(); i++) {
text += "#" + colors[(Math.floor(Math.random() * 5))] + strings.charAt(i) + "#";
}
return text;
}

[NmZero]

That is true but a better comparison is doing:

for (var i = 1; i < 10; i++, text += "#L"+i+"#selection "+i+"#l");

I didn't even know this was allowed/possible.

Also starting at 1 is an interesting choice if you're doing a less than in the second part. Some people may be confused because they wouldn't be sure if you want to go from 0-10, 1-10, 1-9, but that's not really an issue.

The id 91000000 is main free market map, the 1 to 24 are free market rooms, so basically this could be used to warp to a room without getting to main fm first. But like you said, thats not really the issue here.

function cr() {
var colors = ["b", "d", "g", "r", "k"];
return "#" + colors[(Math.floor(Math.random() * 5))];
}

The way you are using cr() is a bit excessive. You may want to look into making a function that takes a string as an argument and returns a randomized color string. Probably a syntax error somewhere here but you get the idea.


function cr(strings) {
var text = "";
var colors = ["b", "d", "g", "r", "k"];
for (var i = 0; i < strings.length(); i++) {
text += "#" + colors[(Math.floor(Math.random() * 5))] + strings.charAt(i) + "#";
}
return text;
}

function caw(string) { //caw = color all words ^_^
    var colors = ["b", "d", "g", "r", "k"];
    var color = string.replace(/ /g, "_");
    var amount = color.split("_").length - 1;
    for (var i = 0; i < amount; i++)
        color = color.replace("_", " #" + colors[(Math.floor(Math.random() * 5))]);
    return color;
} 

[AristoCat]

wouldnt it be easier to

function sendNext_color(str) {
    var colored = "";
    var splitted = str.split(" ");
    for each (var i in splitted)
        colored += cr() + i;
    sendNext(colored);
} 

not sure if works didnt test

[NmZero]

meh, idk :p

[xStr0nGx]

meh, idk :p

function sendNext_color(preText) {
    var coloredText = "";
    var words = preText.split(" ");
    for each (var word in words)
        coloredText += cr() + word;
    sendNext(coloredText);
} 

Is what he meant.

[NmZero]

Its getting a bit off topic here, but why is strong so hated. He made a good point.
more half offtopic things:

function color(str) {
    var colored = "";
    var splitted = str.split(" ");
    for each (var i in splitted)
        colored += cr() + " "+i;
    return colored;
} 

correct code + can be used for any cm. text

I've been trying to do this as well

function cal(string) { //cal = color all letters ^_^
    var text = "";
    var colors = ["b", "d", "g", "r", "k"];
    //text += string.charAt(17-2);
    for (var i = 0; i < string.length(); i++)
        // if (string.charAt(i - 1) !== "")
            text += "#" + colors[(Math.floor(Math.random() * 5))] + string.charAt(i);
        // else
            // text += string.charAt(i);
    return text;
    // var colors = ["b", "d", "g", "r", "k"];
    // var color = string.replace(/""/g, "_");
    // var amount = color.split("_").length - 1;
    // for (var i = 0; i < amount; i++)
        // color = color.replace("_", "#" + colors[(Math.floor(Math.random() * 5))]);
    // return color;
} 

Color each letter differently, the only problem here is that it uses the standard string which includes \r\n multiple times, the result is this:


The commented things are the things I've tried, but so far I've had no succes :c

on topic:
updated script with aristocat's color method

[Xenoverz]

Looks nice !