Thread: [Dev-C++] About member functions
Open Source C & C++ IDE for Windows
Brought to you by:
claplace
Menu
▾
▴
dev-cpp-users
|
From: Ioannis V. <no...@ya...> - 2001-10-31 20:11:23
|
This subject has been discussed on th e past, but now new evidence came up. :) =20 =20 In "The C++ Programming Language" Special Edition on page 419 it says that each object has its own member function as i was claiming in the beginning. =20 =20 It says many things but lets concentrate at the end of that page: =20 "Just like ordinary pointers to functions, pointers to member functions are used when we need to refer to a function without having to know its name. However, a pointer to member isn't a pointer to a piece of memory the way a pointer to a variable or a pointer to a function is. It is more like an offfset into a structure or an index into an array. When a pointer to member is combined with a pointer to an object of the right type, it yields something that identifies a particular member of a particular object". =20 =20 i think we were discussing it with Jason Hardman. Jason if you see it, please tell me your opinion (since you seem you are a pro). =20 =20 Ioannis =20 * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com <http://www.noicys.f2s.com/>=20 * Alternative URL: http://run.to/noicys =20 |
|
From: Jason H. <jas...@bt...> - 2001-11-01 00:58:39
|
MessageI am not a pro for a start, but because of my background I tend = to understand some of the concepts in Classes and Objects from a = different perspective. My background being as a hardware engineer and = using assembler (68000 mainly). The section you are refering to is about pointers to member functions. = Something that a class can do is provide a table of pointers to it's = member functions in it's internal representation. There will still be = only one instance of the member funstion, but each object will contain a = pointer to that fuction. This is specifically needed in the case of = virtual member functions where the caller doesn't necessarily know what = class it is using. By saying that a particular pointer to the member is = at such and such a position in an object then you can indirectly call = the correct function for the object. Pointers to pointers are a useful = form of indirection. In essence what that paragraph is saying is that if you call method 5 = (for example) of an object the compiler will know that it actualy refers = to a particular function in a particular place in memory, or at least a = pointer to that function in memory. I will qoute from my edition (third) on the same page you quote: "C++ offers a facility for indirectly referring to member of a class. A = pointer to a member is a *value* that identifies a member of a class. = You can think of it as the position of the member in an object of the = class, but of course an implementation takes into account the = differences between data members, virtual functions, non-virtual = functions, etc." Does that seem reasonable to you? It seems that we are looking at the = same paragraphs and getting two totally different answers from it. Jason. ----- Original Message -----=20 From: Ioannis Vranos=20 To: Dev-C++=20 Sent: Wednesday, October 31, 2001 8:11 PM Subject: [Dev-C++] About member functions This subject has been discussed on th e past, but now new evidence = came up. :) In "The C++ Programming Language" Special Edition on page 419 it says = that each object has its own member function as i was claiming in the = beginning. It says many things but lets concentrate at the end of that page: "Just like ordinary pointers to functions, pointers to member = functions are used when we need to refer to a function without having to = know its name. However, a pointer to member isn't a pointer to a piece = of memory the way a pointer to a variable or a pointer to a function is. = It is more like an offfset into a structure or an index into an array. = When a pointer to member is combined with a pointer to an object of the = right type, it yields something that identifies a particular member of a = particular object". i think we were discussing it with Jason Hardman. Jason if you see it, = please tell me your opinion (since you seem you are a pro). Ioannis * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com * Alternative URL: http://run.to/noicys |
|
From: Ioannis V. <no...@ya...> - 2001-11-01 01:26:42
|
Yes for virtual functions the pointer is used an an array index (he has a picture). For non virtual functions it acts like an offset. If it helps, he says that we can pass a pointer to a virtual function between address space, but we can not exchange a pointer to a non-virtual function between address spaces. =20 Special Edition and 3rd edition *are* the same. If your printing is not recent it simply lacks only some appendixes at the end of the book which you can download online as pdfs from Stroustrup's pages, so go to his site and check. =20 I found your quote. What i say is on page 420 the draw and the first paragraph. =20 =20 Ioannis =20 * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com <http://www.noicys.f2s.com/>=20 * Alternative URL: http://run.to/noicys -----Original Message----- From: dev...@li... [mailto:dev...@li...] On Behalf Of Jason Hardman Sent: Thursday, November 01, 2001 2:58 AM To: Dev-C++ Subject: Re: [Dev-C++] About member functions I am not a pro for a start, but because of my background I tend to understand some of the concepts in Classes and Objects from a different perspective. My background being as a hardware engineer and using assembler (68000 mainly). =20 The section you are refering to is about pointers to member functions. Something that a class can do is provide a table of pointers to it's member functions in it's internal representation. There will still be only one instance of the member funstion, but each object will contain a pointer to that fuction. This is specifically needed in the case of virtual member functions where the caller doesn't necessarily know what class it is using. By saying that a particular pointer to the member is at such and such a position in an object then you can indirectly call the correct function for the object. Pointers to pointers are a useful form of indirection. =20 In essence what that paragraph is saying is that if you call method 5 (for example) of an object the compiler will know that it actualy refers to a particular function in a particular place in memory, or at least a pointer to that function in memory. =20 I will qoute from my edition (third) on the same page you quote: "C++ offers a facility for indirectly referring to member of a class. A pointer to a member is a *value* that identifies a member of a class. You can think of it as the position of the member in an object of the class, but of course an implementation takes into account the differences between data members, virtual functions, non-virtual functions, etc." =20 Does that seem reasonable to you? It seems that we are looking at the same paragraphs and getting two totally different answers from it. =20 Jason. ----- Original Message -----=20 From: Ioannis Vranos <mailto:no...@ya...> =20 To: Dev-C++ <mailto:dev...@li...> =20 Sent: Wednesday, October 31, 2001 8:11 PM Subject: [Dev-C++] About member functions This subject has been discussed on th e past, but now new evidence came up. :) =20 =20 In "The C++ Programming Language" Special Edition on page 419 it says that each object has its own member function as i was claiming in the beginning. =20 =20 It says many things but lets concentrate at the end of that page: =20 "Just like ordinary pointers to functions, pointers to member functions are used when we need to refer to a function without having to know its name. However, a pointer to member isn't a pointer to a piece of memory the way a pointer to a variable or a pointer to a function is. It is more like an offfset into a structure or an index into an array. When a pointer to member is combined with a pointer to an object of the right type, it yields something that identifies a particular member of a particular object". =20 =20 i think we were discussing it with Jason Hardman. Jason if you see it, please tell me your opinion (since you seem you are a pro). =20 =20 Ioannis =20 * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com <http://www.noicys.f2s.com/>=20 * Alternative URL: http://run.to/noicys =20 |
|
From: Jason H. <jas...@bt...> - 2001-11-01 01:43:28
|
MessageI think in general that it being called a pointer to a member is = quite misleading. I'm not sure in what you say whether I have convinced = you of my point of view or that you are sticking to the idea of a = function to each object. Of course everything we have talked about is = pretty implementation specific, but they will all tend towards the same = systems, with only minor differences. You will note that he says it is = like an offset he doesn't say it is an offset. I believe he is using the = idea of offsets and such as an analogy, which is misleading in it's own = way. He does say that it is up to the compiler to interpret the pointer = in the correct way depending upon what is being referenced. Jason.=20 ----- Original Message -----=20 From: Ioannis Vranos=20 To: Dev-C++=20 Sent: Thursday, November 01, 2001 1:26 AM Subject: RE: [Dev-C++] About member functions Yes for virtual functions the pointer is used an an array index (he = has a picture). For non virtual functions it acts like an offset. If it = helps, he says that we can pass a pointer to a virtual function between = address space, but we can not exchange a pointer to a non-virtual = function between address spaces. Special Edition and 3rd edition *are* the same. If your printing is = not recent it simply lacks only some appendixes at the end of the book = which you can download online as pdfs from Stroustrup's pages, so go to = his site and check. I found your quote. What i say is on page 420 the draw and the first = paragraph. Ioannis * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com * Alternative URL: http://run.to/noicys -----Original Message----- From: dev...@li... = [mailto:dev...@li...] On Behalf Of Jason = Hardman Sent: Thursday, November 01, 2001 2:58 AM To: Dev-C++ Subject: Re: [Dev-C++] About member functions I am not a pro for a start, but because of my background I tend to = understand some of the concepts in Classes and Objects from a different = perspective. My background being as a hardware engineer and using = assembler (68000 mainly). The section you are refering to is about pointers to member = functions. Something that a class can do is provide a table of pointers = to it's member functions in it's internal representation. There will = still be only one instance of the member funstion, but each object will = contain a pointer to that fuction. This is specifically needed in the = case of virtual member functions where the caller doesn't necessarily = know what class it is using. By saying that a particular pointer to the = member is at such and such a position in an object then you can = indirectly call the correct function for the object. Pointers to = pointers are a useful form of indirection. In essence what that paragraph is saying is that if you call method = 5 (for example) of an object the compiler will know that it actualy = refers to a particular function in a particular place in memory, or at = least a pointer to that function in memory. I will qoute from my edition (third) on the same page you quote: "C++ offers a facility for indirectly referring to member of a = class. A pointer to a member is a *value* that identifies a member of a = class. You can think of it as the position of the member in an object of = the class, but of course an implementation takes into account the = differences between data members, virtual functions, non-virtual = functions, etc." Does that seem reasonable to you? It seems that we are looking at = the same paragraphs and getting two totally different answers from it. Jason. ----- Original Message -----=20 From: Ioannis Vranos=20 To: Dev-C++=20 Sent: Wednesday, October 31, 2001 8:11 PM Subject: [Dev-C++] About member functions This subject has been discussed on th e past, but now new evidence = came up. :) In "The C++ Programming Language" Special Edition on page 419 it = says that each object has its own member function as i was claiming in = the beginning. It says many things but lets concentrate at the end of that page: "Just like ordinary pointers to functions, pointers to member = functions are used when we need to refer to a function without having to = know its name. However, a pointer to member isn't a pointer to a piece = of memory the way a pointer to a variable or a pointer to a function is. = It is more like an offfset into a structure or an index into an array. = When a pointer to member is combined with a pointer to an object of the = right type, it yields something that identifies a particular member of a = particular object". i think we were discussing it with Jason Hardman. Jason if you see = it, please tell me your opinion (since you seem you are a pro). Ioannis * Ioannis Vranos * Programming pages: http://www.noicys.f2s.com * Alternative URL: http://run.to/noicys |
