Thread: Structure pass-by-reference - help?

C has no pass-by-reference it is purly pass-by-value. However your problem appears to be related to an invalid setting in your Project.

TheFunc ( &valAddr )
That depends on how one defines pass-by-reference I suppose, a value does still get pushed onto the stack, but it's the effective address.

>In C, you can pass by reference.
The C standard makes no mention of passing by reference. You can fake it by passing a pointer, which is what the Bullschildt you've been reading was poorly explaining, but that's not true pass-by-reference, it's pass-by-value using a pointer as the value.

>This is mentioned in C++: The Complete Reference in the C section.
Take everything Herbie says with a grain of salt. He's usually wrong.

If it does the trick, then it does the job. It is still pass-by-reference because you're pointing to an address, not a value. By the way, I always thought he said by reference, but in the book, he's discussing how to pass the address, which he describes as structure pointers. The code is working now after making a new project, so you guys can chat all you want. Thanks for the help.

Based on what - your reading of a book which isn't about C, by an author who doesn't know C?

He was a member of the ANSI/ISO committee that standardized C++. After selling millions of books internationally, I think he deserves respect. He teaches the language, not feed you bull........ that you see every day on message boards.

You guys need to study more. When a pointer to a structure is passed to a function, only the address of the structure is pushed on the stack. This makes for very fast function calls. A second advantage, in some cases, is when a function needs to reference the actual structure used as the arguement, instead of a copy. By passing a pointer, the function can modify the contents of the structure used in the call. Same as a reference!

You guys argue over the dumbest things on message boards. No wonder why professionals I know say to stay away from them. No one in the field cares about listening to the most misinformed BS coming from a young generation online. Bye!

Originally Posted by dxfoo

He was a member of the ANSI/ISO committee that standardized C++. After selling millions of books internationally, I think he deserves respect. He teaches the language, not feed you bull........ that you see every day on message boards.

You guys need to study more. When a pointer to a structure is passed to a function, only the address of the structure is pushed on the stack. This makes for very fast function calls. A second advantage, in some cases, is when a function needs to reference the actual structure used as the arguement, instead of a copy. By passing a pointer, the function can modify the contents of the structure used in the call. Same as a reference! You guys argue over the dumbest things on message boards. No wonder why professionals stay away from them. There's no educational value in it.

1) Schildt is a horrible author. Don't get your feelings all hurt because we don't like an author of a book you own. It's not an attack on you. No one cares if you bought a book and who wrote it. People mention the fact that he's a horrible author because he is in fact a horrible author. Go do a search on your favourite search engine for "schildt" and "bad book".

2) You need to study more. A pointer is a variable which passes the value of the address to a function. It is not a reference. It is the value of an address. It is a copy of the address. That's why it's still "passed by value". The value of the address is passed. Not a reference. There is a difference. You need to study an author who knows what they're talking about.

It is mistakenly referred to as "by reference", however, C has no "by reference" feature. It is simulated using pointers. However, it is by value, and no matter how much you wish it wasn't, it isn't going to change the fact.


Quzah.

>He was a member of the ANSI/ISO committee that standardized C++.
There's a difference between an active member and an observer. Schildt was an observer, and not a very good one judging from his mistakes.

>After selling millions of books internationally, I think he deserves respect.
He deserves a slap in the face. It doesn't matter how many books you sell if they're all crap. Herbert Schildt is well known for being able to explain incorrect concepts in an easy to understand way. We've been correcting his students for years, so don't expect any respect for him around here.

>He teaches the language, not feed you bull........ that you see every day on message boards.
We're not forcing you to post here. If you trust your book more than us, then go read your book and leave us alone.

>You guys need to study more.
You're mixed up, it's the other way around.

>You guys argue over the dumbest things on message boards.
The difference between an expert and everyone else is that the expert understands the nuances while everyone else just settles for "close enough".

>No wonder why professionals stay away from them.
Quite a few of us are professionals.

EDIT EDIT EDIT EDIT EDITE DITE ESJREGJHGWKHWGWG: I'm stupid.

Okay. I was also interested in this, so I did a little debugging.

void a(int& x) { x += 2; }
int main() { int b = 2; a(b); }

Disassembly:

Code:

// main()
00401088  |. C745 FC 020000>MOV DWORD PTR SS:[EBP-4],2
0040108F  |. 8D45 FC        LEA EAX,DWORD PTR SS:[EBP-4]
00401092  |. 50             PUSH EAX
00401093  |. E8 77FFFFFF    CALL test.0040100F
// a()
00401038  |. 8B45 08        MOV EAX,DWORD PTR SS:[EBP+8]
0040103B  |. 8B08           MOV ECX,DWORD PTR DS:[EAX]
0040103D  |. 83C1 02        ADD ECX,2
00401040  |. 8B55 08        MOV EDX,DWORD PTR SS:[EBP+8]
00401043  |. 890A           MOV DWORD PTR DS:[EDX],ECX

void a(int* x) { *x += 2; }
int main() { int b = 2; a(&b); }

Disassembly:

Code:

// main()
00401078  |. C745 FC 020000>MOV DWORD PTR SS:[EBP-4],2
0040107F  |. 8D45 FC        LEA EAX,DWORD PTR SS:[EBP-4]
00401082  |. 50             PUSH EAX
00401083  |. E8 87FFFFFF    CALL test.0040100F
// a()
00401038  |. 8B45 08        MOV EAX,DWORD PTR SS:[EBP+8]
0040103B  |. 8B08           MOV ECX,DWORD PTR DS:[EAX]
0040103D  |. 83C1 02        ADD ECX,2
00401040  |. 8B55 08        MOV EDX,DWORD PTR SS:[EBP+8]
00401043  |. 890A           MOV DWORD PTR DS:[EDX],ECX

So, my findings conclude that they are exactly the same, and that I didn't accidentally copy and paste the same thing twice. Both simply push a "reference" (the effective address) of our number 2 and modify that accordingly. The pass-by-refeference vs. faked pointer stuff seems to be only a higher level implementation to let you use '.' and not derenference the pointer on every use. I could and probably am wrong in saying this, and my compile (VC++) might have made bad assumptions. I don't know.

Edit: Dearly sorry for the edit's but gcc gives the same results. And compiles the pass by reference. "meh"

I think we should just let the microsoft apologist get on with being a bullschilter.

> He was a member of the ANSI/ISO committee that standardized C++
*looks at forum - it says C*
Nope - not impressed by that argument at all - same amount of credibility as if he'd been on say the pascal committee.

> You guys argue over the dumbest things on message boards.
Because the detail always bites you in the ass eventually.

> misinformed BS coming from a young generation online. Bye!
LOL - there are quite a few people here who've got a lot of years over you. If you want to extrapolate the universe from one bad book, then that's your problem.
Oh, you're leaving - nevermind then.

> So, my findings conclude that they are exactly the same
Well that's what you get when you mix up language concepts (pointers and references) with what an implementation does with them.
It's the same argument as saying pointers and arrays are the same thing - they're not, though they do have some similarities. Sure you can get away with thinking they're the same, but sooner or later you'll come unstuck (for example, most peoples first attempt at passing a 2D array to a function usually ends up with ** in there somewhere).