Posts by Dogtree

Well...you can't. Time to upgrade to Office 2003? ;)

> But is it really worth building up ur own instead of buying a pre-configured one ?
It depends. First, are you comfortable building your own? Can you find a pre-build system that gives you everything you want for a reasonable price? With your desired configuration, would it cost more to build or buy? Do reasearch, my friend. :)

You know we're not going to do it for you, right?

#1) I would recommend that you reformat the drive and reinstall. But, since that's #2, I'll just mention that there are Linux installations on live CD that you can run and have access to your drives. For example, in bad situations like this I would use my Knoppix disk for troubleshooting and recovery.

#2) It sounds like somebody went into the CMOS and changed the boot order so that the primary hard drive is at the top of the list. You need to go in and change that so that, for example, the floppy drive is first, then the CD-ROM drive, then the primary hard drive.

#3) Removing the CMOS password is pretty easy. First, to avoid searching high and low for jumper settings, remove the CMOS battery from your motherboard for a few seconds. That should reset the password. Otherwise, you'll probably need to flip some jumper settings to reset the password, which requires a somewhat advanced knowledge about the board you're using. If it's a brand name then a quick call to tech support will get you the jumper number and setting to kill the password.

When it comes to pre-configured retail systems, I've always stuck with PowerSpec. Then again, it's almost always better to build your own. That way you have complete control over price, scalability, and power.

> Write down your favorite cp commands and what they do!
ping! :)

> With my ADSL i can go to the config page and change some internet stuff on
> there but i just dont know what to change.
Look through the options for some mention of DMZ. Let me know and I'll walk you through what to do about it and give you sufficient warnings. :D

Okay, Windows 2000 isn't as easy to get into as 95 or 98. You'll probably have to break the admin password and change it to suit your needs. Start here.

I don't think this is the right forum for your question, unless you plan on writing this program in C or C++.

Okay, what do you have so far?

Jedi Knight goes straight through a software or router based firewall, I've played every game in the series online and it never failed to work. But it does sound like you aren't getting out. Are you on a router or other hardware firewall?

Symantec can be a brute sometimes. Were you running a scheduled background scan or update at the time? Those two operations are the only ones that should take up so much of your resources, and it shouldn't last more than an hour on average, though that depends heavily on your setup.

A small portion of the OS would need to be written in assembly, beyond that it's theoretically possible to use Java, though I'm not sure if it's been attempted.

$60 will get you a four port switch. Even less for a hub. You don't need a full blown router to do the job.

>> my ISP is telling me that it wasn't their fault
It's never their fault. ;) Be aware that just about everyone will try to dodge blame when something goes wrong. Like I said, if it happens again, call them and force them to walk you through some troubleshooting steps. At the very least, they can help you pinpoint the problem wherever it may be.

It's probably your ISP. If it happens again for any length of time, give them a call and see if there are any outages in your area. Otherwise, I wouldn't worry about it.

If it's an older PC then it's probably Windows 95 or 98. In that case, you really don't need the password. When the dialog pops up asking for it, just hit cancel and it should let you in.

Are we talking about a CMOS password or an operating system password?

twm, just because it was the first one I got a feel for and I never bothered to switch to another. ;)

If this person has physical access to your computer, you're basically SOL. Remote attacks can be hindered with software and hardware protection, but if the attacker can walk up to your computer and turn it on then no amount of protection will keep them out if they know what they're doing.

Barring physical protection like a lockbox, of course. ;)

Generally, if your description of the functionality relies on the word 'if', you probably need some form of conditional, which CSS doesn't natively provide. You're looking at a PHP/CSS or Javascript hack as your best option.

Try opening the image using the software most likely to view it. If it looks great in photoshop, but like crap in a browser or other image viewer than you you should be the first to know. :)

>> Lets test out your wit
Translation: Who can search google better? ;)

I've already posted a review in the Geek's Lounge.

>> but it was cheaper for me to make it ratheer than hire a pro
Clearly you're looking at the wrong pros. ;) My own Dogtree Studios provides high quality design for pocket change. Compared to just about any other development firm, we're ridiculously underpriced. In reality, everyone charges something different. If you shop around you can find good designers for a reasonable cost. For a business web site, that's the way to go unless you consider yourself a hobbyist professional who can do the work just as well, but doesn't make a living from it.

In the end, it's a trade-off. Do you think a pro would do a better job? If not, is the time spent in developing your own site worth the monetary savings?

Okay, rule #1 about dynamic content: Audio that you can't turn off is EVIL! Always give the client an option when it comes to potentially annoying features. The first thing I did when the page loaded was kill the browser. The second thing I did when the page loaded again was stop the flash movie from playing. Those aren't reactions that you want to cultivate. A better option would be a flash video with no audio, and a link that gives the client a chance to listen to the music if they so choose.

This site is best viewed with Netscape.

The days where this message would be acceptable are long gone. First, you're assuming a browser, which probably means that the site will be netcrap for other browsers. Second, you're not even assuming the most likely browser, which would be IE 5 or 6. You even have a button that links to a place where Netscape can be downloaded! Are you trying to sell your band, or Netscape? I can't really tell.

The link to get into the site isn't blindingly obvious. To be perfectly honest, you would be better off losing the splash page entirely because it's more of an annoyance than anything.

Moving into the 'real' site, the first thing I noticed was that it looks like someone actually bothered to do design work. That's good, and another reason to lose the splash page, because it doesn't give off a level of professionalism that …

That's because it's not allowed there. :rolleyes: Using static to qualify functions is deprecated anyway, use an unnamed namespace instead:

namespace {
  template <> double ToDouble( double source )
  {
    return source;
  }
}

template <> double ToDouble( double source )
{
  return source;
}

>> So I guess to answer my own question, 'yes, I missed something'
Yes, but it's an easy mistake to make. Since string literals are of type const char *, and it's not possible to overload built-in data types, "F" + "F" is trying to add two addresses, which is an ambiguous, and thus, illegal, operation.

>> correct?
That would work, but a somewhat more concise method is:

std::string("F") + "F"

or

"F" + std::string("F");

since the + operator allows the string object on either side of the expression.

>> Temperary variable to be passed in recursive function
No.

>> temparary variable are temparary to scope of function
Yea.

>> and in recursive function we use global variable sort of things
What are you babbling about? It's possible and perfectly reasonable to use a non-static local variable in a recursive function. This function doesn't need a static variable or a global variable. Here's your current function:

list *reverse (list *head)
{
  if (head->next == NULL)
  {
    tail = head;
    return;
  }else{
    reverse (head->next);
    head->next->next = head;
    head->next = NULL;
  }
}

Here's my modification using a local temporary:

link *reverse_list(link *head)
{
  link *temp;

  if (head->next == 0) temp = head;
  else {
    temp = reverse_list(head->next);
    head->next->next = head;
    head->next = 0;
  }

  return temp;
}

The algorithm is identical, but my function is completely self-contained and works properly whereas yours would have unnecessary coupling with that global variable if you kept following the same line of thinking.

You're on the right track. Try incorporating a temporary link into your function rather than trying to work just with head and its next links.

>> (certainly not C or C++).
It looks like Managed C++. But I agree that this question would be better asked in a .NET forum.

You need to include <iomanip> for setprecision().

This is a classic trade-off, and unfortunately, one that only you can decide.

Anytime you need to split off into one or more separate paths, you're looking at a switch or if chain.

>> is <fstream.h> different than <fstream>?
Yes, very different. <fstream> is a standard modification of the prestandard <fstream.h>, where all of the classes are turned into template classes and a bunch of other relatively minor changes are made. Unless you're forced to write legacy code on a legacy system, always use the standard headers. You can tell them apart because the standard headers don't have a .h extension.

Where to start depends on what you want to do. To have a fairly broad range of marketable skills, you should be comfortable with C, C++, Java, Perl, SQL, PHP, and Visual Basic. My personal belief is that everyone should know C, and that it should be the first, or one of the first, language learned.

What bay area? There are a lot of bay areas on this planet. ;)

>> #include <fstream.h>
Mixing old and new headers will bite you every time.

#include <fstream>

using std::ifstream;

That's why it's a good idea to cut and paste the code you have from the start rather re-type it in the post.

>> Here is the compile log from Dev-C++:
That's assuming that Zackery doesn't have Allegro installed, which he probably does. Almost all of those errors and warnings have to do with you not compiling and linking correctly with a third party library.

>> i dont understand how Its not wiorking
How is it not working? Do you get compiler errors? Linker errors? Run-time errors? Is the output just not what you expected? Your question is too vague.

>> Are they the same?
Kinda sorta, but not really. A switch is a control flow structure that directs execution through a certain block of code depending on an integral value. In other words, it's like door #1, door #2, and door #3 in the old gameshows. Those three doors together would be the body of a switch, 1, 2, and 3 would be the cases, and the condition is the very confused contestant trying to pick which one to open.

A function is a named block of code that can optionally accept data to work with and optionally return data when it's done. Think of a function like a drive through carwash. The carwash is the function, a dirty car going in is an argument, the body of the function cleans the car, and a clean car coming out is the return value.

>> What are yours?
Daniweb and newsgroups. My most frequently visited are comp.lang.c, comp.lang.c++, comp.lang.c++.moderated, and comp.std.c++. From what I've seen, the same people post to multiple forums, so there's really no need to visit more than one on a regular basis.

I have no idea what your question is. Can you state it in a different way?

Why even bother? A hash_map has better performance properties, so it makes sense to choose it over a map. If the user needs something sorted, they can list to a stringstream, then to a vector, then sort it themselves at only minor cost. You're talking about having detrimental performance for everything to support a convenience that probably wouldn't be used often enough to justify it.

>> I was hoping to not force the user to know that I was using map or hash_map
That's dangerous since the operations of a map and typical hash_map implementations are wildly different. If you're going to go the hidden implementation route, you need to specify an interface in your wrapper that conforms to both the map and hash_map. Otherwise you'll just confuse the users of your class when a bool changes everything in seemingly unpredictable ways.

I think a strong separation between a sorted map and an unsorted map is a good one, and I personally would consider a wrapper for both to be a design flaw. Then there's the question of how much you want to hide from your users at the cost of flexibility?

Is that the exact error or your rendition of the error?

>> can a context belong to more than one process?
Why would you want it to? The context of a process is it's state at any given time. If two processes shared the same context then they would effectively be duplicates. That's not as useful as it may sound at first because the overhead of sharing context doesn't really buy you anything compared to creating new context for a new process, and there are only a small number of times where sharing the state of a process would be useful.

>> ccin.open(data.dat);
I think you mean

ccin.open("data.dat");

>> getline(ccin,str);//the problem :(
Looks okay to me. What compiler are you using? Can you give us a short and complete program to compile that gives you the error? Can you post the error?

There's not a common base class for map and hash_map since hash_map isn't even a standard container and they aren't commonly implemented in remotely the same way. set and map might have a common base class, but that's an implementation detail that you shouldn't rely on because the standard doesn't require it. A better way to design your map wrapper would be with policies:

#include <iostream>
#include <map>
#include <set>

template <typename MapType>
class MapWrap {
  MapType m;
public:
  typedef MapType value_type;
  MapType& get() { return m; }
};

int main()
{
  MapWrap<std::map<int, int> > wrap1;
  MapWrap<std::set<int> > wrap2;

  for (int i = 0; i < 10; i++)
    wrap1.get()[i] = i * i;

  for (int i = 0; i < 10; i++)
    wrap2.get().insert(i);

  MapWrap<std::map<int, int> >::value_type::iterator it1 = wrap1.get().begin();

  while (it1 != wrap1.get().end()) {
    std::cout << it1->first << ' ' << it1->second << '\n';
    ++it1;
  }

  std::cout << '\n';

  MapWrap<std::set<int> >::value_type::iterator it2 = wrap2.get().begin();

  while (it2 != wrap2.get().end()) {
    std::cout << *it2 << '\n';
    ++it2;
  }
}

You could also inherit from MapType:

#include <iostream>
#include <map>
#include <set>

template <typename MapType>
class MapWrap: public MapType {
};

int main()
{
  MapWrap<std::map<int, int> > wrap1;
  MapWrap<std::set<int> > wrap2;

  for (int i = 0; i < 10; i++)
    wrap1[i] = i * i;

  for (int i = 0; i < 10; i++)
    wrap2.insert(i);

  MapWrap<std::map<int, int> >::iterator it1 = wrap1.begin();

  while (it1 != wrap1.end()) {
    std::cout << it1->first << ' ' << it1->second << '\n';
    ++it1;
  }

  std::cout << '\n'; …

Um, can you be more specific? It sounds like you want us to give you the full code for an FTP server for Linux, written in C. That's not a request that will ingratiate you to us. ;)

I'm there regularly. Maybe you're just there are a slow time. :)