|
VW_Red_Jetta wrote:
Is pointers a headache for you?
No. I can't dream of a fairly large program without pointers.
VW_Red_Jetta wrote:
And is it useful?
See above.
VW_Red_Jetta wrote:
If you don't use pointers, what alternatives do you have?
I dunno.
Actually, I had to do a couple of small programs (one in VB, one in Java). I got really pissed off  - with VB for the language and Java for lack of pointers. C++ is for me. I'm gonna learn C#, though. 
Vikram.
-----------------------------
My site due for a massive update.
"Good lord you yanks are getting worse than a defensive VB developer at a C++ conference." - Paul Watson in the Lounge, 31 May 2003.
"Do not give redundant error messages again and again." - A classmate of mine, while giving a class talk on error detection in compiler design.
|
|
|
|
|
Pointers are one of the most useful parts of C++/C. Used well, they become one of the most powerful tools in a programmers toolkit - used poorly and they can make your life a living hell.
When I first started wiht pointers in the late 80's (in C), it took me a couple of months to understand the concept but now they are second nature.
Michael
'War is at best barbarism...Its glory is all moonshine. It is only those who have neither fired a shot nor heard the shrieks and groans of the wounded who cry aloud for blood, more vengeance, more desolation. War is hell.' - General William Sherman, 1879
|
|
|
|
|
As long as java exists, there's and endless discussion between C/C++ and Java programmers about which language to use.
I have some experience with java. I learn it at school, from my point of view, Java programmers are lazy thinkers(no offence) and C/C++ programmers are lazy writing!
Back to pointers, I use pointers as much as I can. They keep the mind sharp!
In Java they didn't get rid of pointers at all, they just hide them from the programmer. (Ever had a null pointer exception??)
|
|
|
|
|
First of all, my first misconception about java, was that there were no pointers. That misconception from a C++ point of view led me to believe that you just used copied objects everywhere.
Java is all about references -- which really are just smarter pointers. Smart would be pushing it, but smarter... I can't tell you how many hours I've spent debugging java code because it wouldn't release objects. {how smart can it be if you can still screw it up?}
I think of Java like I think of PASCAL. Remember PASCAL? It tries real hard to protect you from shooting yourself in the foot, but if you attempt anything too lofty, you will end up shooting yourself in the foot, but not realize it, and then you die a slow but somewhat painless death. Java makes a great teaching language -- as long as you don't really want to understand what's happening under the hood.
References live in C++ too, and but they don't do the auto-delete-referenc-counter thingy that java does. MS came out with this thing called COM. COM implements some java-like stuff in C++. Reference counting and smarter casting stuff...
My answer is: that I'd much rather use C++. I think it's the control-freak in me, but I know what I am getting with C++. {knowing assembly helps too} Java leaves you greatly at the whim of the person that developed the runtime...
The learning curve with C++ is higher. And java has some constructs that are missing in C++ (like "super" -- I love "super"). But I'd still pick C++ any day!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~<br />
Peter Weyzen<br />
Staff Engineer<br />
Santa Cruz Networks
|
|
|
|
|
Well, I remember OBJECT PASCAL. They try to prevent the use of pointers. I couldn't even find information about pointers in the delphi help files!
The 'auto-delete-reference-counter thingy' from Java is called the garbage collector. This is one of the many things that make Java a slow language!
Some people also tried to implement a garbage collector in C++, but failed at some aspects. The reference counting is done by so-called 'smart pointers'. A topic about C++ garbage collecting and 'smart pointers' is around here somewhere.
Almost everyone in my class didn't start their programming experience with C/C++. Almost everyone (with experience) was aquainted with (object) PASCAL, Java, Visual Basic. They all did well, until we got tje subject 'C'. Almost everyone failed at understanding pointers. They didn't know with what the were dealing with (maybe they're just stupid).
So my opinion is: If you want to learn how to program, start with C/C++. Why? You don't get used to 'easy' programming and when you know something about C/C++, no other language is hard to learn. They are all easier than C/C++. If you do it the other way around, you still got a hard language to learn!
The 'super' keyword is indeed a handy thing, but C++ allows a similar construction!
exemple:
<br />
class A<br />
{<br />
private:<br />
int n;<br />
public:<br />
A();<br />
int GetInt();<br />
};<br />
A::A()<br />
{ n = 1; }<br />
int A::GetInt()<br />
{ return n; }<br />
<br />
class B : public A<br />
{<br />
private:<br />
double b;<br />
public:<br />
B();<br />
double GetDouble();<br />
}<br />
<br />
B::B() : A()<br />
{<br />
d = 1.1; <br />
}<br />
<br />
B::B()<br />
{<br />
d = 1.1;<br />
A::A();
}<br />
Did you know that there is a way to interface assembly with C/C++? This can be achieved by using the __asm keyword. Consult your help or the online msdn library for more details
|
|
|
|
|
Well, languages that hide pointers, they are similar for me to having some car with engine capable to run 300kmph, but the car vendor decided for you , that it is dangerous and you cannot do it and forced some limitation to the car, that you can now drive max. 130kmph.
Yes, it is safe, and it will protect some people from being hurt, but it also hides some part of power inside the thing and you are not able to use it even on a racing polygon.
|
|
|
|
|
I need to get the Thread ID of another thread of the same application. The situation is like this, IEXPLORE.EXE loads my DLL (a thread) and it also has some other threads . I (my DLL) want to get the Thread ID of one of those threads (lets say an applet). Can I?
Thanx in advance,
Mahesh
|
|
|
|
|
One solution is GetThreadId().
Kuphryn
|
|
|
|
|
Thanx.
kuphryn wrote:
One solution is GetThreadId().
  I have never seen such an API function....
Please get me more info pls...
Thanx.
Mahesh
|
|
|
|
|
I have written application which connects to MYSQL
database using recordset. I have written the code to
update a record from the table , which gives me error 'No
cursor name available'
Code is as follows.
CMyRecordSet set;
set.Open(CRecordSet::snapshot, "select * from
customer", 0);
set.m_CustNoParam = 1;
set.Edit();
set.m_CustNo = 2;
set.Update();
set.Requery();
I got the resolution for the problem in MSDN help as
follows.
--------------------------
The version 2.0 "ODBC Specification" requires that
implicit cursor names must remain until statement handles
have been dropped. The drivers provided with the ODBC
Driver pack, version 2.0, have been changed to conform to
the 2.0 specification. These drivers may be used with the
MFC library database classes without running into the
above problem.
----------------------------
I checked MYSQL driver version. It is 2.0 and still I am
getting the problem.
|
|
|
|
|
|
Is there is freeware on WIN 98 that display the content of memory in use by a program?
Thanks.
|
|
|
|
|
By a single program- not any that I know of. 
But if you wanna display the amt of memory in use by the entire system ,there are lots of freeware programs like Rambooster.
Vikram.
-----------------------------
My site due for a massive update.
"Good lord you yanks are getting worse than a defensive VB developer at a C++ conference." - Paul Watson in the Lounge, 31 May 2003.
"Do not give redundant error messages again and again." - A classmate of mine, while giving a class talk on error detection in compiler design.
|
|
|
|
|
hi, all,
we have a project, a search engine to search pictures over internet.
before start, i have to know if it is possible.
main idea of the project is: the app doesn't "download" web pages (as normal engine does), instead, it sends a small app to each site to run the small app there to analyze if the site has pictures we need (i.e. pictures which have red color as main backgroud).
the reason to run the small app in each site is: downloading pictures is to slow (size of a picture is over 500K normally), sending a mall app is faster (size is 30K).
if no security protection, can i send a small app to each site to run it there silently (sites don't known about this)?
if yes, how to do that and any links or hints will be great.
thx
includeh10
|
|
|
|
|
Basically, you're asking if you can execute arbitrary code on a remote server without the permission of the admin.
The short answer is: No.
Long answer: Such a system could only work if it were voluntary (that is, the admin of the server decides to participate by downloading your app and running it) and you provided ports of the app for a number of different platforms. Many admins would want the source to check that it is indeed doing what you say it does (and nothing else). Also, the app would have to not use much CPU power (image analysis can be very CPU expensive). After all that, you'd have to ask yourself if it would really produce better results than GIS: http://images.google.com/[^]
- Mike
|
|
|
|
|
Dear all,
I am developing an application for Windows NT/2000/XP in which I want to create a folder whose size should not exceed 5MB. And should promt for error if data exceeds 5MB
Is there any way to fix the size of folder ?
At same time i want that folder should not be deleted.
Please comment!
Rohit
|
|
|
|
|
No. You cannot change the size of a folder without removing files.
Kuphryn
|
|
|
|
|
I donot want to change the size of folder, i just want to limit its size
|
|
|
|
|
I ran into a FAQ article written by Bjanrne Stroustrup, the
creator of C++. One question is about if void main() is a
legal statement and he replied this:
<quote>
The definition
void main() { /* ... */ }
is not and never has been C++, nor has it even been C. See the
ISO C++ standard 3.6.1[2] or the ISO C standard 5.1.2.2.1. A
conforming implementation accepts
int main() { /* ... */ }
and
int main(int argc, char* argv[]) { /* ... */ }
</quote>
This is link of his article:
http://www.research.att.com/~bs/bs_faq2.html#void-main
So what do you think? Do you do void main()? I think most
of the programmers do, and because this syntax is so popular
that eventually Java and C# has the statement like
public static void main(string[] args)
{
// blah blah blah
}
they say void main() !!
|
|
|
|
|
Most compilers will accept void main(), but that doesn't mean you should use it. In fact, on gcc 3.2.2, attempting to use void main() results in:
<br />
void.c: In function `main':<br />
void.c:2: warning: return type of `main' is not `int'<br />
VW_Red_Jetta wrote:
I think most
of the programmers do, and because this syntax is so popular that eventually Java and C# has the statement like ...
In Java, the reason main is void is because the standard way for dealing with errors is exceptions, not return codes.
Really, I just can't get worked up about the entry point of an app. However, if you're really into the theory behind the syntax and semantics of languages, I recommend writing your own script language interpreter. Writing a very simple language and interpreter is relatively easy and serves as a good introduction to abstraction and working with strings (and lets you get creative with the language). 
- Mike
|
|
|
|
|
main() has always had a return type of int. The return value from main() is used as the exit code for the process.
Now, the reason for why some people write void main() is that historically, C programmers wrote:
main(int argc, char** argv)
{
}There are two things happening there. First, earlier versions of C let you omit the return type if it was int. Second, some compilers let you omit the return statement from main() and implicitly returned 0 or 1 or whatever their default was.
So later on, new C programmers who don't know those rules come along and assume that since there is no return type listed, that's the same as void.
--Mike--
Latest blog entry: *drool* (Alyson) [May 10]
Ericahist | Homepage | RightClick-Encrypt | 1ClickPicGrabber
"You have Erica on the brain" - Jon Sagara to me
|
|
|
|
|
|
hello,
i've a strange problem, i've written a program where i use a lot of keybd_event functions to simulate keyboard events, it seems like that this messes up the whole windows system. drag&drop is not working, and therefore starting programs with a doubleclick on the desktop is also not working.
this happens only on a few systems, i've searched for days but found no solution, can anybody help me?
bye
Mario
|
|
|
|
|
Are you also sending key up events? If you send
keybd_event ( VK_ESCAPE, 0, 0, 0 ); you'd better also send
keybd_event ( VK_ESCAPE, 0, KEYEVENTF_KEYUP, 0 ); or else the system will think the ESC key is still pressed.
--Mike--
Latest blog entry: *drool* (Alyson) [May 10]
Ericahist | Homepage | RightClick-Encrypt | 1ClickPicGrabber
"You have Erica on the brain" - Jon Sagara to me
|
|
|
|
|
I'm not at all sure whether this is the correct forum for this question but here goes:
I'm trying to investigate an existing signal processing algorithm that has been written in C. I do have the source code so I'm NOT reverse engineering! Instead I want to assess its "performance" using some "time series" data that I have generated in MATLAB.
The problem is that MATLABs data storage format (MAT files) does not allow you to extract records (i.e. data at specific timeshots in the time series)- instead you have to store an MxArray variable representing the whole time span of the data. This is inefficient in terms of memory space required - what if I wanted a huge timespan of data points?!
So I've looked at the possible use of CDF - Common Data Format - created by NASA for storing multidimensional data together with descriptive metadata. It is freely available on the Web (nssdc.gsfc.nasa.gov/cdf/html/FAQ.html) and comes with a C API. NOW ... MATLAB has functions "cdfread", and "cdfwrite" so ought to be able to create CDF files and read them back BUT ... the cdfwrite does not work when you try to append a new record to the CDF file. I've sent example files to MathWorks an they concede that the cdfwrite function doesn't work to spec!
Has anyone out there had a similar need to store time series data from MATLAB to a file which can later be accessed from C, one timeshot at a time? How did you do it and was it an efficient process? I'd be very grateful to anyone who can help me. ![Rose | [Rose]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/rose.gif)
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
