Comments by H.Brydon (Top 200 by date)

The cynic in me wants to suggest looking at the Fisher-Price website... sorry, I am not a Windows 8 fan...

I would recommend the slightly different:

fwrite(FileOut, sizeof(FileOut[0]), ArraySize, file2);

Note that OP flagged Win32 and C++.

OP submitted around 10 questions on the forum all at the same time. All were pretty basic. This was perhaps the most complicated question. I would not have replied as I did if it was the only question. My comments here reflect the bigger picture.

... although I think OP is being lazy and stand by that comment.

Good answer from you (+5 for that)...

This is really basic stuff. You need to do a little more effort on your part...

If you want MFC, look at how to use CFormView. There should be a bazillion examples out there.

Okay, or look at documentation on ATLTrace2() and TRACE() or OutputDebugString()... These all display info in the output window when you run your program in the IDE.

Write an API that looks at the address of the stack, and takes the difference from a call to the same API at program start...

Come on now, if you put this search into Google, you get a multitude of good to great hits on useful information.

You are being really lazy.

Your questions don't belong in this forum...

You seem to have several questions in the Q&A section that are likely answered by looking at the basic Microsoft documentation on how to do exactly what you want...

I learned how to create a Windows app tool bar in 1992 using the most basic compiler/IDE documentation available at the time.

The obvious answer is to write a short program that will dump out a trace message for each of the above messages... and change the locale on the machine. [without looking] I'm sure that Google might give some useful info too... (or try Bing)

This looks a lot like homework, which is not the purpose of this forum. I am surprised that you already got a couple of responses.

You didn't say what kind of problems you are having with it... more detail?

Do some research on your own. Learn how to use Google. Don't ask other people to do your work for you. Write something, try a few things. Ask for explanations about things that are not clear. If something doesn't work, try several different approaches before asking here.

Don't ask here for someone to do your work for you. Don't ask for someone to debug your program. Don't say it is urgent, and don't ask to have any requests done quickly.

FYI - Requests of "please debug this for me" or "please write a program for me" will not get much traction in this forum...

Yes it is still part of VS2010. I use it. It comes with the VS2010 product but you have to select it as a feature to install during the installation process.

Sorry, but this site is not for doing homework, "fix my program" or writing code for you. You need to try a little harder...

That does not look like C or C++ to me...

Ewwww for the chalkboard statement!
+5 for the great answer... :-)

We don't do that kind of thing on this forum. Specifically, requests for "send me code", "fix this for me" and the like are not typically met with friendly responses.

I think you meant compile time error but +5 anyhow...

+5.

Ummm ... F5?

What you are describing here is a little different. Your code captures the OnNotify message in the view/dialog class, and holds a tooltip control there, relaying the event to the control when a hit is detected. OP wants to handle this in the control (without subclassing)...

If you hit a web server every few milliseconds, the service provider will shut you down. Torturing somebody else's machine like that is bad juju and the provider will notice this almost immediately.

Anything more than once a minute is in the range of denial of service attack.

I don't think you want that...

More correctly, it is not part of the project, so it won't get compiled or linked. The errors in question are linker errors...

"Please send code" is not a valid request for this forum and is likely to get you a lot of well-earned abuse.

...or in the same directory as the executable if the app is deployed or otherwise run outside the IDE.

I somewhat agree with the code suggestion but use of enumerations would be better than int...

I also agree with RM's comments about doing homework...

Somewhat reminds me of someone wanting to ask a question but doesn't know the protocol. :-)

+5 for OG of course.

+5 for "monster" aka "pattern" reference. :-)

+5 for obscure Edgar Allen Poe reference. :-)

My first guess would be https://www.google.com/#q=how+to+link+to+jni

You seem to be using JNI in a way that means you are calling java from C++. I don't have expertise in this area but it seems you need to link to the JNI libraries as well as VOCE. Google should be helpful for this...

The debug assertion will (likely) tell you what line is failing. Look at that line in the code.

... except that I hate seeing uninitialized values - anywhere. My preference would be to combine the two statements into one:

int result = SomeFunc();

...or the place I used to work hired some new programmers. :-)

Additional one from me instead...

Your first sentence I think you meant "authentication and encryption" but +5 anyhow...

+5 for both of you for the effort (watch for it...)

You could hire a boiler room to do it like Google does:

http://web.archive.org/web/20110403054337/http://www.google.com/intl/en/jobs/uslocations/mountain-view/autocompleter/index.html

You submitted this as a solution, but it is an extension of your problem statement. I think you want to edit your question to include this text and delete the non-solution "Solution".

You need to do a bit more debugging. Try printing out intermediate values and (for sure) identify which line generates the error.

Go hang out around a meat packing plant.

I'm not an expert in this area but my guess is that you should be using rootNode.SelectSingleNode() instead of bookResult.SelectSingleNode().

Having said that, your xml segment is not well formed, and it is not clear if the entire xml content will give you anything sensible.

Not sure why you were downvoted. Your solution is correct (except that DEBUG_NEW is misspelled). More details are provided in Solution #4.

One thing to watch out for is to make sure you match new/delete and new[]/delete[] as this can upset the leak detection report.

+5 to help set the universe straight again.

This isn't a solution. If you have details to add to your question, you should edit the question.

You really aren't asking a coherent question. It is coming across like "my brother married a woman who has a daughter. What is her name?"

Null pointer is one of the easiest errors to identify. Run the program with the debugger and see the line number where it barfs.

You should use the "Question or Comment" mechanism to make this response instead of providing a "Solution" that is not a solution. Pranay likely didn't see the non-solution "solution".

Your solution will also likely be downvoted; you probably want to delete this...

Minor point but when you make the code recursive, I'd suggest "if(x<= 0)" instead of "if(x==0)"...

A java null pointer exception is the same on any platform - you have an uninitialized pointer.

You should remove your 2 "solutions" here or you will find that people will downvote your misuse of the mechanism. This info belongs either in the comment section or as an update to your question...

If the two apps are always on the same machine, I'd suggest shared memory.

Heh heh. New woman in my life.

See you in a few more weeks. :-)

This is clearly homework, which we don't do here. Your best choice is to consult your class notes. Google will also be more helpful than this forum.

[without checking] Google will probably tell you how to do this if you search for "open file append"

I think Google is going to be more helpful for this kind of question than anything you will see here...

You need to make far more progress before asking that kind of question here. This site is for answering fairly specific programming questions, not how to program or "please write a program for me".

Rather impressive question.

Before you fix it up, let me just say that brushing up on Google skills will probably get you what you want.

Learning so fast does not help.

The only way to do this kind of thing is to use a multiplatform language such as java (ie. which has multiplatform UI features). C++ is not going to help you here.

You could write your non-UI code in C++ and interface it to java for the UI.

Oh yeah and +5 for Griff.

I don't know if it was you that downvoted the solution (3 out of 5) but since the second bullet point in the list seems to have identified the problem exactly, a +5 would be a good thing to do. :-)

Thanks for your comments.

The Wikipedia page at https://en.wikipedia.org/wiki/Packet_analyzer has a reasonable explanation of packet analyzers in general, and some good leads on free software and "how to" links.

Gone to the dark side you have.

I am out of ideas but it still goes back to file system protection on the share, on the host machine directory/directories and the file. Perhaps place a file on the remote machine with "everyone" write access, same again for its parent directory and work backwards from there.

When I displayed the images, my browser went to la-la land, and I had to close it and restart. Not a good sign.

I don't changing a hosting site will change much.

You should get to know Google, and it will reward you with lots of documentation and examples of your choosing.

Changed from providing advice to doing homework now?

Yeah, my thoughts exactly. +5

Security alert for readers - these links give some problems with the browser, might be malware.

This is not a "give me the code" site. If you have a programming question, we can probably help with that.

For what you want, Google will likely grant you more than 3 wishes.

What are you specifying for parameters 2, 3 and 6?
What operating system for your local machine and the remote machine?
Are there any firewall issues between your machine and the remote machine?

That kind of question is what Google is for.

Sorry but this forum is not for writing code or doing homework for you. You need to research this and do it yourself.

+5 to compensate for whomever downvoted...

Your question is not clear and your code is nonsense. You are not doing any totalling (ie. accumulation) and most code optimizers would just throw all your code away and return to the caller.

I think you want to send the resize event notification from the parent window to the splitter window. (I don't have the coding details in front of me so I send you this as a comment not a solution)

+5 for the effort

My mother went to a restaurant last Thursday and ordered a tuna salad sandwich, hold the mayo. When she bit a mouthful of said sandwich, one of her friends walked by.

What was her name?

I <3 your C. :-)

Why are you answering a question from 2011?

This forum is not for doing your homework.

Perhaps show the definition for the VlcControl class.

[going offline for about 12 hours shortly...]

Hmmm ... warnings bad juju.

Okay, +5 for the effort!

It seems to be complaining about vlc.dotnet.forms. Where did you initialize vlc.dotnet and likewise vlc.dotnet.forms? You need to show what VlcControl consists of.

+5 - Exactly. The same page I searched for and was going to put in my own solution.

Answering a question from 2011?

Thanks. I haven't heard that title for a while but I suppose it fits. :-)

I was trying to say that console output can't (easily) be saved to an external file.

This is pretty basic knowledge. If you can't figure that out, you need somebody local that can teach you with some hands-on training.

Google can come up with coding examples. For fopen_s() for example, Google gives https://www.google.com/#q=c%2B%2B+code+examples+fopen_s

+5 good answer. :-)

Whew! +5

Agreed. Some of the laundry items I can identify:
- #include/#import rules we just kicked around here...
- lack of module support (like Pascal or better)
- lack of concurrent memory model
- poor support for concurrency (C++11 is on the way but hasn't arrived at the station yet)
- precompiled headers (defined, and how they are used)
- lack of "negative declaration" identifiers (if there is a virtual, there should be a non-virtual)
- overloaded identifiers (static means several things depending on where it is used)
- there should be 'interface', 'default', 'inherit', 'disabled' and others (such as in C++/CLI)
- 2 types of whitespace

There are more...

The SO page just seems to say to use angle brackets for system headers without saying why. At my company we had a build system that (I think correctly) required all includes to use angle bracket syntax.

Suppose you (stupidly) added a file called "windows.h" to your local directory and used quote syntax. That would break a lot of things. Other more subtle examples could be determined with the same principle...

No, (at least for the Microsoft compiler) - use of the quoted version uses an arguably defective search path depending on essentially cached info (which can have bizarre header dependencies that are impossible to debug). Use of the angle bracket version, the search path is unambiguous.

Check out Microsoft's rules at http://msdn.microsoft.com/en-us/library/36k2cdd4.aspx

I don't know if this is Microsoft specific or not, so I don't know the rules for other compilers.

Thanks very much.

Thanks - that info is actually solution-worthy by itself.

That's not really a workaround - that is the right way to do it.

Thanks Sergey (I think?). I think we belong to the same religious brotherhood.

+5 - not sure why you were downvoted; your answer looked reasonable to me...

I agree - that would work out to about one question asked per 10 CP members. Yup, that's about right. :-)

+5 although I'd say 10 years.

+5 for your perfect English. Oh yeah the answer looks good too. :-)

You said "why" - did you perhaps mean "how"?

If you couldn't use the main method in java, it would always have to be started from another language API or some other entry point convention.

You said SMTP but I think you probably meant POP3 or IMAP (and correspondingly different RFCs).

Your best bet is to invoke the great search god Google: https://www.google.com/search?q=wininet+binary+file+uploading+in+c%2B%2B

+5 (although you are doing all the work for OP who should have done this himself.)

I count 15. :-)

I think your command line in the .bat file should perhaps be

start /d "C:\Program Files\Microsoft Visual Studio 10.0\Common7\IDE\VCExpress.exe" readxmlresou.exe /debugexe

and your command line in java should be

r.exec(new String("C:\\users\\abdehalim\\desktop\\abs.bat"), null, new File("C:\\users\\abdelhalim\\desktop\\"));

[edit]
Oh yeah, and you want to adjust the .bat file until it does the right thing when you execute it with a double click from Explorer. It should not expect parameters. Once you get that working, then go to the java IDE and get that part going...

I'm not sure what you are looking at. The coding examples don't include attachment examples but every library listed at this Google page supports attachments (MIME or otherwise).

I don't have VS2010 Express edition (I have the full version). The help for devenv shows the following:

d:\vs2010\VC>devenv /?

Microsoft (R) Visual Studio Version 10.0.40219.1.
Copyright (C) Microsoft Corp. All rights reserved.

Use:
devenv [solutionfile | projectfile | anyfile.ext] [switches]

The first argument for devenv is usually a solution file or project file.
You can also use any other file as the first argument if you want to have the
file open automatically in an editor. When you enter a project file, the IDE
looks for an .sln file with the same base name as the project file in the
parent directory for the project file. If no such .sln file exists, then the
IDE looks for a single .sln file that references the project. If no such single
.sln file exists, then the IDE creates an unsaved solution with a default .sln
file name that has the same base name as the project file.

Command line builds:
devenv solutionfile.sln /build [ solutionconfig ] [ /project projectnameorfile [ /projectconfig name ] ]
Available command line switches:
[...]
Product-specific switches:

/debugexe Open the specified executable to be debugged. The
remainder of the command line is passed to this
executable as its arguments.
/useenv Use PATH, INCLUDE, LIBPATH, and LIB environment variables
instead of IDE paths for VC++ builds.

To attach the debugger from the command line, use:
VsJITDebugger.exe -p <pid>

d:\vs2010\VC>

... so you need to specify /debugexe if you want the program to be started when vcexpress.exe starts (assuming it has the same command line switches).

Also, for VCExpress.exe to run correctly it needs to have the proper environment initialized. If this is not the case for a plain command prompt on your computer, you need to use a different strategy. I would suggest building a batch file "abc.bat" that initializes the environment, and runs the programs you have described above. This means putting "abc.bat" in your r.exec() call instead of VCExpres.exe. This also means that you can do a little more debugging by calling your batch file from the command line to see if it works correctly.

Yeah, you are out of time and want someone else to do your homework for you.

If you graduate, I wouldn't hire you. Sorry...

How do you know it does not execute? I haven't used this activation method but I suspect that it starts a program without providing either a UI or an output console. Is it reasonable that the program starts and runs but does not create any detectable output? Perhaps put a MessageBox() call in the startup sequence of your app...

If that isn't helpful, I don't have any further expertise with this way of activating; I haven't used this method.

Posting homework answers here is a bad idea. It only encourages lazy and/or dishonest students to abuse this forum.

I've used this technique and it works fine (+5). This requires setting the address manually, and it disappears at the end of the debug session. To make this more persistent, establish a reference or non-void pointer at the memory address in question (programmatically or manually) and set a watch point on the variable.

Messages can be deceiving but the computer does not lie. If it says that it can't find a file, then that is the case.

+5. I didn't try it but it looks good.

+5 back... I didn't realize you posted a solution just before mine. :-)

Sadly, yes. I just can't tear myself away from the keyboard. :-(

+5 for your endurance. :-)

I agree, I like your code better, although there are still some other issues. The values ee and Pee should be class variables (not static) and the cleanup stuff should really be in destructors instead of being reused.

What we have ended up with here is a C program written using some C++ features.

Thank you Sergey. Agreed, I think that there are probably other issues to address, too.

If you ask Google, you will get several hits. Have a look at http://msdn.microsoft.com/en-us/vstudio/ff459609.aspx
Beware of some of the 'lessons' that teach some rather advanced features such as IDE extensions, interacting with MS Office and other such complicated things. Stick to the basic concepts. There is a wealth of sample code that you can use and/or study to develop your skills. Microsoft.com and MSDN are quite helpful.

No, I didn't laugh. I was too busy stepping out of the way in case of a bucket split or an overflow of some kind.

(+5)

Your url seems to be truncated. Perhaps you meant http://en.wikipedia.org/wiki/Calling_convention (?)

Another link that might be better is http://en.wikipedia.org/wiki/X86_calling_conventions

Also:
http://msdn.microsoft.com/en-us/library/zxk0tw93%28vs.71%29.aspx
http://msdn.microsoft.com/en-us/library/984x0h58%28v=vs.71%29.aspx

This won't work if the nodes are a non-contiguous list. For different reasons, it also won't work if the nodes are a contiguous list.

If the node pointers are truly pointers, you will end up with a botched list in both cases.

The choices you have are:
- scan the entire list each time you delete a node and update all index values
- scan to the node to be deleted, and (as you say) abandon the in-node index value

The choice is up to you. If you don't use the index value any more, it should probably be removed from the node struct definition or somebody else will come along and try to use it and it will find out the hard way that it has incorrect info. [This is where big bugs come from.]

Oh great poobah
Named 'nv3'
You get a +5
And more respect from me.

Apologies to the Vogons. :-)

Be careful about mixing debug and nodebug dlls within an image. The official answer from Microsoft is that you can't do it. In fact you can do it in some cases but it is unsupported. Notably, memory management is different between the 2 build types, and if you create an object with new() in the debug world and delete it with delete() in nodebug (or vice versa), you will trash your heap.

Bad juju.

My +5 also for providing codeless algorithm. Wise craftsman you are.

Still still almost almost :-) :-)
It is not dependent on compiler or platform. This is defined in the C standard, and brought forward to C++ and beyond. It is common for all compliant compilers and platforms.

There is no difference (to generated code or internal representation between values on each of the following lines:
0. 0.0 0.000 00.0000
0.f 0.0f 0.0F 0.000f 00.000F
The first line represents a double zero, and the second line, each represents a float zero.

Ack - that is exactly correct. I need a vacation. :-)

Exactly. Too short I might add. :-(

That's a very good one. Prepare to be quoted. :-)

Yo another +5 for nv3.

This sounds like a restatement of solution #2...

This solution is not correct. If you delete anything "at or before" a forward iterator, the iterator becomes invalid. Making a second copy does not help you. It is invalid as well.

I agree, solution #1 and #2 are both correct, but #2 addresses the immediate problem better.

No, it works because reverse iterators start from the end of the list, and if you delete the last element, it is at the end of the list, and the reverse iterator is still valid.

If you delete the first element with a (forward) iterator, all the remaining items in the iterator's context juggle down and the iterator becomes invalid.

Interesting... Most of the questions I see like this are people who mix up ansi and unicode and get Chinese when they are expecting text. :-)

I'll give you another +5. :-)

[The only answer I give a +5, for the text "never call virtual functions from constructors..."]. IMHO a constructor should never call them, and any articles about doing it are really academic. My recommendation is to not call anything in the same class hierarchy except a private non-virtual method or (C++11) a delegating constructor.

+5 for the answer, but in defense of OP, the answer to this question is not easily findable unless you know the answer already.

It is considered bad form to do lazy student's homework for them on this forum.

Me too. Please post your code.

Interesting answer but why are you answering a 3 year old question?

+5 (real) for the answer
+5 (virtual) for the BG reference. :-)

The industry answer is controversial. I like your answer the best. +5

I read your answer and gave it a +5 (obscure reference to your other comment). :-)

Your answer is so good that I figured out what the question was by reading the answer. Is this Jeopardy?

+5

Snowden, but yeah. +5

Well too bad; I +5d you about 5 times today (so far). See if you can reach integer overflow... You have good advice; don't give up on it.

I'll pass on the +5 :-)

(mine is probably more points for you...)

Right, and if it needs to be accessed in multiple .cpp files then declare it in a common .h file and use 'extern' in the .h file...

Not a good idea to do homework for people here, especially for really basic questions. By doing this, you are inspiring other lazy students to bombard this forum with silly questions that they should be doing themselves.

I think we are going way out on a limb here. OP's question was about aborting threads and getting threads to exit gracefully (see original question title and question text). My position is that threads should always exit gracefully whenever possible and not aborted externally (ie. external to the thread). We have strayed a long way from that original question. You said that you disagreed with me but most of what you said here agrees with my original text in solution #2.

I think we beat this issue to death. OP should be able to write his code.

Likewise you seem to be disagreeing with me but you are saying the same thing that I did. A thread should not be killed externally but should be signalled to die and then wait for it to notice the request, handle it and die of its own accord.

External thread termination is bad juju.

You said that you disagree, but you pretty much said what I said. My major point is to not use TerminateThread() from an external thread/process, but to somehow tell the thread to die and let it handle its own hospice details.

You're on fire today. I would have +5'd that. :-)

Exactly, float and double should not be used for any representation of currency (including stock prices).

+5

Posting an elementary question with no code and saying it is urgent is a recipe for getting no respect, no help and/or no response. I'm feeling nice today and am going out of my way by being helpful and redundant here.

Probably the easiest +5 you've ever received... :-)

To expand on your thought ... Compilers are usually either written in a hardware specific language such as assembler for the target machine, or (after a brief minimal bootstrap process) in the same programming language as that being compiled. For example, the first FORTRAN compiler was written in FORTRAN. I wrote a C++ compiler in C++. et cetera... I would not expect a C++ compiler to be written in VB.NET or C#.

Roses are red
Violets are blue
Your answer gets
a +5 for you

Wrong forum for that. We don't write code or produce samples. You need to ask Google for ideas.

You beat me to it. That's pretty close to the answer I was going to give.

+5

Good info, +5... I will likely use it myself.

Yup. +5

+5 for answering the OP question with his own code.
+5 for the Klingon reference and bombastic style. :-)
0 for the robustness of the answer (see Ian Davidson comment)

Average: +5 :-P :-)

+5 for the answer. I owe you +5 for the effort. :-)

... probably the easiest +5 you'll get today. :-)

+5 for being correct, yet refined.

More to come I'm sure...

+5 for being correct, yet bombastic.

No, that is not a realistic way to do that.

I'm not suggesting that this is an answer but where you have '&#xA;' I think it needs to be '&#x0A;'.

Ha ha - I owe you a +5 for that pseudo-answer. Watch for it... :-)

Good question (with perhaps confusing details per the comments and solution text).

+5 to offset whomever downvoted you.

Compare the following 2 sets of statements:

if(list == 1);
{ ... }

if(list == 1)
{ ... }

Why do you have a semicolon at the end of the if() statement?

I assume that you use a domain (as stated). Start with a sample machine that has not seen activity from 'dragon' before, and login to the domain for the first time on that system with your 'dragon' username. Do any other setup you need to do, then logout. From an administrative username, search in the registry for references to 'dragon', and identify all of the fields associated with the username addition. Look to see where the directory/directories and file(s) were created when you logged in. Write an installer to duplicate this (don't forget about ownership and access).

Robert is your father's brother.

That still won't compile.

Also not a good idea to do homework (or really basic programming) for people. This encourages cheating and/or dumb questions from lazy students/programmers.

Zowee, another +5 for an extraordinary effort!

You could update your answer and change the totalage declaration to be

int totalage = 0;

That won't fix everything (per pasztorpisti) but at least the sum will be calculated correctly.

What goes around comes around. Keep watching. :-)

+5. Ears burning?

Great analysis; I owe you another +5 - watch for it (I gave you several others in the last day or two). I also share your affinity for OO and disdain for iostreams.

You didn't initialize the totalage variable (major point of solution #2). Or move averageage declaration to the bottom where it is used. In fact you could make a minor change to the code so it is not needed.

[I also agree with pasztorpisti about disdain for iostreams...]

Thanks for your +5.

Ha ha - http://lmgtfy.com/?q=what+does+tl%3Bdr+mean :-)
If I make a long post, I usually add a "tl;dr" line or two at the bottom.

Whew! (yet another) +5 from me. This one mostly for the effort.
You should add a "tl;dr" version too...

Yeah me too, I missed it as well. +5

Yet another +5. I think I've given you about 500 points this week. :-)

Nice links. +5

Perfect answer, +5

... and don't use the "Rebuild" option. Click "Clean" and then "Build". Sometimes "Rebuild" does the wrong thing.

+5, I didn't know that. Good stuff...