Give me a horror :)

Philip Laureano

Hi Guys, I'm going to write an article for beginners on how to refactor deeply nested "if" statements, and I was wondering if anyone here had any VB.NET or C# code that they could share that had a few "if" statements at least five levels deep that I could use as a refactoring example...do I have any takers? :)

Do you know...LinFu?

jhwurmbach

Philip Laureano wrote:

article for beginners on how to refactor deeply nested "if" statements

You won't endorse the use of multiple return in one function, no? And surly not even name the dreaded word GOTO? :-D

Let's think the unthinkable, let's do the undoable, let's prepare to grapple with the ineffable itself, and see if we may not eff it after all.
Douglas Adams, "Dirk Gently's Holistic Detective Agency"

Vasudevan Deepak Kumar

Isn't if-else if-else if-else itself horrendous? A decent switch-case (C#) or a select-case (VB) should be cool and elegant!

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson

Paul Conrad

I'll see what I can dredge up :)

"I guess it's what separates the professionals from the drag and drop, girly wirly, namby pamby, wishy washy, can't code for crap types." - Pete O'Hanlon

leppie

I could give you some code, but you are not going to to be able to refactor it :) (think (hand)unrolled loop)

xacc.ide - now with IronScheme support
IronScheme - 1.0 alpha 2 out now

Philip Laureano

jhwurmbach wrote:

You won't endorse the use of multiple return in one function, no? And surly not even name the dreaded word GOTO?

I'd never endorse using GOTOs, but I'd definitely endorse using multiple returns from a single function if it's easier to read. If you look at some of the code that I've done with LinFu, you'll notice that most of my "if" statements don't go over two levels deep, and you'll never find a single switch statement inside LinFu. My goal is to help people write better code, and hopefully the article I'll write will be able to do that. :) You never know if the code you teach someone how to write will be the code you have to maintain someday...

Do you know...LinFu?

Philip Laureano

leppie wrote:

I could give you some code, but you are not going to to be able to refactor it (think (hand)unrolled loop)

Try me. :) This should be interesting...

Do you know...LinFu?

Philip Laureano

Vasudevan Deepak K wrote:

Isn't if-else if-else if-else itself horrendous? A decent switch-case (C#) or a select-case (VB) should be cool and elegant!

Hopefully, I can show the neophytes how to get rid of switches and elses altogether. Personally, I prefer replacing switch statements with polymorphism, and replacing explicit else logic with implicit else logic. It makes the code so much easier to read.

Do you know...LinFu?

leppie

Now that I was thinking of it, it wasnt really nested! :-O Due to me formatting behavior of always having braces around if / while/ do/ foreach, VS's formatting makes the code look nested :doh: I will keep an eye out for some. I am sure I have a few uglies ;P

xacc.ide - now with IronScheme support
IronScheme - 1.0 alpha 2 out now

Bevan C Bird

I've got a couple snippets from what I wrote about a year and a half ago, as a C newbie. Sorry about the formatting. The first one is a line clipping function for displaying lines on a calculator screen, as part of a rudimentary zoomable vector diagram program.

void Geom_LineOn_inWindow(Geom_Line *lines, unsigned short int n)
{
	
		Geom_Line W_N = {{W_Min_x, W_Min_y}, {W_Max_x, W_Min_y}};
		Geom_Line W_E = {{W_Max_x, W_Min_y}, {W_Max_x, W_Max_y}};
		Geom_Line W_S = {{W_Min_x, W_Max_y}, {W_Max_x, W_Max_y}};
		Geom_Line W_W = {{W_Min_x, W_Min_y}, {W_Min_x, W_Max_y}};
		
		unsigned char p1_in;
		unsigned char p2_in;
		
		unsigned short int k = 0;
		for (; k < n; ++k)
		{
				Geom_Line l = lines[k];
				
		    p1_in = ( l.p1.x >= W_Min_x && l.p1.x <= W_Max_x && l.p1.y >= W_Min_y && l.p1.y <= W_Max_y );
		    p2_in = ( l.p2.x >= W_Min_x && l.p2.x <= W_Max_x && l.p2.y >= W_Min_y && l.p2.y <= W_Max_y );
		
		    if ( p1_in && p2_in )
		    {
		    	  // Both Points are in window...
		        DrawFastLine(LCD_MEM, l.p1.x, l.p1.y, l.p2.x, l.p2.y, A_REPLACE);
		    }
		    else if ( (!p1_in) && (!p2_in) )
		    {
		
		        // Both Points are outside the window...
		    		// Find all Intersections of the line we'll draw with the four window boundaries:
		    		
		    		//DrawFastLine(LCD_MEM, l.p1.x, l.p1.y, l.p2.x, l.p2.y, A_REPLACE);
		   		  if (Geom_intersect(l, W_N));
		    		else if (Geom_intersect(l, W_E));
		    		else if (Geom_intersect(l, W_S));
		    		else if (Geom_intersect(l, W_W));
		        else
		            goto skipLine;
		            
				    signed short int dx = l.p2.x - l.p1.x;
						signed short int dy = l.p2.y - l.p1.y;
						float m;
						
						Geom_Point int_pt, int_pt_other; // intersection point, starts out as the outside point, then is brought in to the boundary line.
						short int i = 0;
						
				    int_pt = l.p1;
				    while (i < 2)
				    {
						    if (dx == 0) // vertical
						    {
										if (int_pt.y < W_Min_y)
										    int_pt.y = W_Min_y;
										else
										    int_pt.y = W_Max_y;
						    }
						    else if (dy == 0) // Horizontal
						    {
										if (int_pt.x < W_Min_x)
										    int_pt.x = W_Min_x;
										else
										    int_pt.x = W_Max_x;
							  }
						    else 
						    {
							    	m = dy / dx;
							    	if (int_pt.x < W_Min_x)
							    	{

Philip Laureano

Ok, this is definitely one I can work with, but do you have any other horror samples in the .NET languages? I still need to be able to compile those samples...

Do you know...LinFu?

MarkB777

Just nest heaps of if/else if/else statements without using brackets where possible, and making terrible use of white space :) Mark.

jhwurmbach

Philip Laureano wrote:

I'd definitely endorse using multiple returns from a single function

Myself, I DO multiple return, but I definitely try to limit them. I mostly use them in the first lines of a function, while validating the input. But after that, I most often use nested if or even a bool marker. There are quite a few company coding style guides disallowing (or strongly discouraging) multiple returns.

Let's think the unthinkable, let's do the undoable, let's prepare to grapple with the ineffable itself, and see if we may not eff it after all.
Douglas Adams, "Dirk Gently's Holistic Detective Agency"

Tristan Rhodes

Well, if i still worked in my last place, i could provide you with TONS of sample material. However, not to inundate you, i would give you the gem of the entire system. A single 750 line PageLoad function that was 12 nested If / Case statements deep at places. Unfortunately, i don't have it :( Still, it should keep you occupied for a month or five.

------------------------------- Carrier Bags - 21st Century Tumbleweed.

Ri Qen Sin

For VB.NET, I just use ElseIfs. I haven't found myself in a situation where I had to use a lot of nested If statements. If I do, I end up moving that code to a new method anyway.

So the creationist says: Everything must have a designer. God designed everything. I say: Why is God the only exception? Why not make the "designs" (like man) exceptions and make God a creation of man?

Derek Bartram

OI, some of use like GOTO :) (although use sparingly granted)

Lost User

Heres a good solution: do{ if(not some state) break; if(not some other state) break; ... etc... }while(FALSE)

Morality is indistinguishable from social proscription

CTF66

I too use the 'do', also Back in my VB6 days, I would often (amongst other things) start an app up using a Select. Select Case False Case DoThis() Case DoThat() Case Else //Launch Main Form End Select Each method is called in turn, the first method that fails (returns false) aborts the startup. Much cleaner than endless if-else's, and although it's probably not correct to compare Select and switch, switch is no match.

modified on Thursday, April 3, 2008 3:58 AM