I use the following for debugging on my dev machine System.Diagnostics.Debugger.Assert(false); When it executes it pops a dialog that asks for the debugger to run via the 'retry' option on the first popup. Once the debugger starts there is a break point on the above line. There are limitations to this such as where the app is running and timeout issues with execution threads (with the debug line).

This should give you a starting point. http://www.google.com/search?hl=en&q=How+to+populate+a+TreeView+with+XML&meta=[^]

guess it depends on how the merge should work. if you only need to append one file to another, cmd.exe will do the job. xslt is fine, but you'll need something that starts the transformation process automatically.

I fully agree with the first reply, and want to add the following: Your idea of "automation" is a good idea, but your priorities are wrong. For docs, the most important thing is: writing it and maintaining it. Your process should be all around making it quick and easy to write, to keep it current, and to find. This should determine what tools you use - not the publishing process that connects these three. The "publishing process" of course should be automatic and fast, but if someone makes a chance and thinks "firing up the doc editor takes too long - I'll change that later", you have lost. 15 seconds can be "too long" already. Further, I yet have to find the tool chain that allows what you describe: a comfortable editor that outputs XML, for which you can download XSLT's to transform it into PDF. Maybe that tool is out there somewhere, for that's what the angle bracket fetishists XML guys say is the future. We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP blog: TDD - the Aha! | Linkify!| FoldWithUs! | sighist

Tomerland wrote: I "feel" that my topic has to do with workflow-programming. Is this right? Interesting question, however it bring up another question. Since you have far more information about your applications requirements than people reading your post in an internet forum, why wouldn't it make more sense for you to study what workflow is[^] to answer that question rather than asking people in an internet forum? :rolleyes: led mike

Thanks's for your answer, but I still are unable to catch the point probably :-))) As I understand you, Requirements are from the user and Features are from the implementation. Am I right? Let me try an example: We have 3 user requirements: R1: Read configdata from file after startup R2: Store configdata to file at closing I have the following implementation (-features) I1: Search file from directory-tree I2: Read config-file I3: Interpret config-file I4: Write config-file This means: R1 is implemented by I1 and I2 and I3 R2 is implemented by I1 and I4. For getting traceability now, I have to check which requirement is implemented by which feature. Is my thinking right in some way? With kind regards

Thank you !

Go to the "My Documents\Visual Studio 2008\Addins" directory, and remove anything related to the add-in from there.

The other day I manually removed one by simply deleting the .AddIn and .dll files from: My Documents\Visual Studio 2005\Addins. This was for the MatchOptionsCommand addin which was initially installed via a .vsi. But I suspect it depends on how the addin is installed in the first place. I wouldn't be able to remove, e.g., DevExpress's Refactor! addin in this way. Kevin

I perfromed a repair - it dit not help. :-(((

Unfortunately I'm only familiar with using the doxygen binary - not the source-code

Can you tell me how to clear the web-cach? Thanks for your answer

seems reasonable what you say Thanks ! !

Hallo, I found another part of solution. If I set for all 3 components the debugger-option to "mixed mode" than I can expand the "unmanged objects" in the "managed C++"-Wrapper :-)))) But I still cannot expand my "unmanaged C++"-objects in my "manged C#"-Code. I fear, this is not possible due to totally different languages Kind regards

A few ideas: 1. Since the line is executed millions of times, you may be able to speed it up with loop unrolling http://en.wikipedia.org/wiki/Loop_unrolling[^]. 2. Every time you do pY[j], it requires a multiplication and an addition. If you're processing the array sequentially, it would be faster to start with a pointer to pY[0], then increment it to get the next array element. If you show some more surrounding code, we may be able to see some other optimizations.

Really? But if there is no exception thrown and there is no pointer set to 0. What else should be testet?

Thanks for your replies !!! Another question: What is the code for handling this szenario in the right way? Kind regards

Tomerland wrote: But if I comment out "return pInst->add(a,b);" then everything works fine. The code you posted does not show the pInst member being initialized. You can't use it without initializing it to a valid value. Oh and this is the wrong forum for your question. This is a C++/CLI question so if you need more information you should ask in that forum. led mike