Printable View
Hi,
I am looking for websites for objectives questions for C/C++ like we have exams for objectives.
It will be good to have list of free objective tests websites. But I am happy to pay if somebody
know that perticular website is good for objective tests and worth paying.
Thanks,
You mean a website that has some sample test questions?
Assignments designed for an ancient compiler from the previous millenium is hardly relevant..Quote:
Originally Posted by phantomotap
yes pleaseQuote:
cpjust asked:
You mean a website that has some sample test questions?
Posting opinions derived from a cursory glance is relevant how? Or, if your prefer the direct approach, read the questions offered before you behave as an imbecile.
Soma
I did read some of the questions in there, the indentation is inconsistent and the code was designed to run in DOS. Also, they use void main several places, and global variables where it isn't necessary, and old header files. That site is a bad place to learn C++ from...Quote:
Originally Posted by phantomotap
Question 9 is definitely using "iostream.h" which is "old style".
Question 31:
My understanding of an incomplete type is different.Quote:
Incomplete types refers to pointers in which there is non availability of the implementation of the referenced location or it points to some location whose value is not available for modification.
Example:
int *i=0x400 // i points to address 400 *i=0; //set the value of memory location pointed by i.
Incomplete types are otherwise called uninitialized pointers.
An incomplete type is something like this:
Otherwise, I don't see anything directly wrong (although some questions where quite obviously targetting someone who had read some particular book/coursework).Code:struct blah; // We have not described what is INSIDE blah - just that it's some sort of struct - thus it is incomplete.
void func(blah *p)
{
// We can use p here, as long as we don't try to access it's content.
}
--
Mats
Well I did notice this right at the top:
Which doesn't instill a lot of confidence in me, and the code formatting really is horrible, but I saw some things that would be useful while skimming over it.Quote:
Note : All the C sample programs are tested under Turbo C/C++ compilers.
It is assumed that,
Programs run under DOS environment,
The underlying machine is an x86 system,
Program is compiled using Turbo C/C++ compiler.
O_o
Yes, and that shouldn't bother you. The note is offered to suggest that one may safely ignore the problems associated with threads/locking/etc..Quote:
Programs run under DOS environment
Would assuming a TI 8533 be more appropriate? Indeed, some questions are hardware dependent; the logic being that understanding the results of undefined behavior can be bloody useful.Quote:
The underlying machine is an x86 system
It didn't actually say what version... but in the context I would assume version 4.0; a compiler that outperforms virtually every compiler from virtually every company, produced over a decade later, in many areas--most notably any mechanics involving conformance of templates. It still outperforms the majority of developer owned compilers for small devices.Quote:
Program is compiled using Turbo C/C++ compiler.
Would it instill more confidence I wonder if the environment listed was Windows and the compiler an unlisted version of MSVC?
Soma
Well like I said, there were some useful things I saw there. I just wouldn't use it as my only source for studying. Surely there must have been a lot of assignments and notes passed out in the class that would help narrow down what areas would be best to focus on. It wouldn't really be fair for the teacher to ask questions about stuff that wasn't taught.
But the answer to the first question is wrong!
A null pointer assignment cannot occur in that case. The behavior is undefined, but null is not involved. "delete ptr" will not set ptr to null. It will only invalidate the call to *ptr. Odds are the program will still work as intended.
A decade after it's release is still 5 years ago, no matter how good it was back in 1993 it doesn't comply to the standards anymore. Even some of the answers are wrong, as matsp and king mir pointed out, so i maintain that this guide is of little relevance, but you can just continue calling me an imbecile if you wish...Quote:
Originally Posted by phantomotap
On the contrary, it should bother you. We don't live in a DOS age anymore.Quote:
Originally Posted by phantomotap
As others kindly point out, Turbo C++ is very old, and has poor standards compliance today. It's faster because it's less complex than today's standards require. And therefore, due to bad standards compliance, it shouldn't be used. Not to mention it produced DOS binaries, which adds further limits, such as limited to 640k or so of memory.Quote:
It didn't actually say what version... but in the context I would assume version 4.0; a compiler that outperforms virtually every compiler from virtually every company, produced over a decade later, in many areas--most notably any mechanics involving conformance of templates. It still outperforms the majority of developer owned compilers for small devices.
Would it instill more confidence I wonder if the environment listed was Windows and the compiler an unlisted version of MSVC?
Plus the solution given is not a proper fix to the problem. The solution is to implement the copy constructor and copy assignment operator to perform deep copying, as per the rule of three.Quote:
But the answer to the first question is wrong!
ok. that site is olde' and crappe'.
now please provide the OP with an alternate, better site, instead of nit picking :(
It doesn't. Discussing issues of threads/locking/etc. can't be done in the realm of standard C++ because the standard allows reordering of instructions and provides no primitives for such constructs. You may as well discuss GUI constructs for all of their validity in the realm of standard C++.Quote:
On the contrary, it should bother you.
O_oQuote:
It's faster because it's less complex than today's standards require.
Hah. Who said anything about it being faster? It is actually more advanced in many areas than most compilers distrusted today as modern, conforming, C++ compilers. Do any of you know the history of the compiler or the C++ standard? Most of the techniques today called "modern C++" was invented/discovered only because of the robust support for templates in that compiler. Microsoft first managed a compiler capable of compiling conforming source, that version 4.0 of Turbo C++ can easily compile, in 2005.
Get over yourselves; the compiler still outperforms, in many areas, most compilers from most distributors for any platform.
Soma
This is Windows specific, and there's no need to discuss about that just yet. Nevertheless, such things are required in today's world, so it is a good idea to talk about them.Quote:
Originally Posted by phantomotap
It doesn't. It's old; it's horrible and limited to DOS environments and all of its disadvantages. Writing a compiler is a difficult thing. When did Turbo C++ start vs Visual C++?Quote:
Hah. Who said anything about it being faster? It is actually more advanced in many areas than most compilers distrusted today as modern, conforming, C++ compilers. Do any of you know the history of the compiler or the C++ standard? Most of the techniques today called "modern C++" was invented/discovered only because of the robust support for templates in that compiler. Microsoft first managed a compiler capable of compiling conforming source, that version 4.0 of Turbo C++ can easily compile, in 2005.
Get over yourselves; the compiler still outperforms, in many areas, most compilers from most distributors for any platform.
Microsoft may not be the best company at creating software, but that is irrelevant. Microsoft's compiler adheres to the newest C++ standard available, Turbo C++ does not and that, aside from its DOS limitation, is reason enough to get rid of it.
I think that it does not really matter how important the test compiler was historically, or even how relatively good it is now, but how good the technical questions, answers and explanations are.Quote:
It is actually more advanced in many areas than most compilers distrusted today as modern, conforming, C++ compilers. Do any of you know the history of the compiler or the C++ standard? Most of the techniques today called "modern C++" was invented/discovered only because of the robust support for templates in that compiler. Microsoft first managed a compiler capable of compiling conforming source, that version 4.0 of Turbo C++ can easily compile, in 2005.
Get over yourselves; the compiler still outperforms, in many areas, most compilers from most distributors for any platform.
Several of the code examples given obviously do not conform to the C++ standard despite using what should be standard constructs, but that matters less than that some of the examples are bad examples, that the code examples are not consistently formatted, and that the decision to use special characters makes for an impediment to reading and testing out the code. On the other hand, many of the questions that I looked at are generally not bad at all, and many examples and explanations are accurate.
My conclusion is that we are looking at something that needs to be updated and polished, but other than that it is fine if prayami already has a reasonable grasp of standard C++.
Off topic:
You're Zero Valintine, formerly a user from the Dev-C++ forums on Sourceforge (if I remember correctly)?
Do you have any specific evidence such as links to credible websites that can back up those claims or are you just pulling those opinions out of your ass?Quote:
Originally Posted by phantomotap
Turbo C++ might be great for compiling DOS programs, but if DOS is extinct, then who cares?
Exactly. You'll notice I haven't argued with anyone that said the answers were wrong or misleading. I haven't read the questions in a very long time. Laying the malformation of the questions/answers on a compiler/platform is... well, actually rather funny!Quote:
I think that it does not really matter how important the test compiler was historically, or even how relatively good it is now, but how good the technical questions, answers and explanations are.
That said, I do disagree that having all the questions following some specific formating should be desired. If a person could only find a flaw if the formatting was what they personally used then they need that experience.
"Visitor" would be more correct; just as I am here. "User", I feel, would imply I get something out of it.Quote:
You're Zero Valintine, formerly a user from the Dev-C++ forums on Sourceforge (if I remember correctly)?
No. They certainly exist. They are very easy to find. I just have no intention of wasting my time providing such links.Quote:
Do you have any specific evidence such as links to credible websites that can back up those claims or are you just pulling those opinions out of your ass?
Soma
Not specific, but consistent (and reasonable, of course).Quote:
That said, I do disagree that having all the questions following some specific formating should be desired. If a person could only find a flaw if the formatting was what they personally used then they need that experience.
As a visitor, you (and I) used the Sourceforge forum functionality, so that makes you a user. But yes, good to see you again :)Quote:
"Visitor" would be more correct; just as I am here. "User", I feel, would imply I get something out of it.
It hasn't been as long as you apparently think. I've visited this forum for a few weeks from time to time since 2003... when you told me this is where you were going.
Still, yes, good to see you as well.
Soma
(Nope, not this username. I couldn't convince anyone that I was me when I forgot my password and no longer had access to my Hotmail account. ;_;)