C / C++ / MFC

Re: AddRef and Release function of COM inteface

19-Oct-07 5:08

George_George wrote:
Intersted, what is Apartment model?

This is a nice summary[^]

George_George20-Oct-07 4:18

Problem in generating XML log files

20-Oct-07 4:18

Thanks led!

Good article.

have a good weekend,
George

Charsets , koi8-r

Daniel Kanev7-Mar-17 21:54

Daniel Kanev

7-Mar-17 21:54

Hi,
I want to convert different character encodings like koi8-r to ASCII.
Where can I find information about them?
I can see there are many charsets - utf-8, windows-1251 etc.

Re: Charsets , koi8-r

Nemanja Trifunovic18-Oct-07 2:50

Nemanja Trifunovic

18-Oct-07 2:50

Daniel Kanev wrote:
I want to convert different character encodings like koi8-r to ASCII.

You can't convert kio8-r to ASCII. The former covers Cyrillic alphabet, and the later Latin (the English subset of Latin, really). What exactly do you want to do?

Programming Blog
utf8-cpp

Message Closed

18-Oct-07 3:21

Daniel Kanev

18-Oct-07 3:21

Message Closed

modified 8-Mar-17 3:44am.

Re: Charsets , koi8-r

Nemanja Trifunovic18-Oct-07 3:34

Nemanja Trifunovic

18-Oct-07 3:34

If you want to preserve the cyrillic text, I suggest converting it to some form of Unicode. If you don't know much about it, maybe you can start by reading The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!)
[^]

Programming Blog
utf8-cpp

Rocky#18-Oct-07 2:14

18-Oct-07 2:14

Hello every one,

I'm trying to generate log files in XML using MSXML4. Right now I'm trying to get it working correctly in a test application, so that when I get it right I could just implement the same thing in the actual project i need to work on.

This operation is intended to log the calling of functions in the application and the XML it needs to generate should be something like this

<pre>
<CallSequence>
<Call>
<ObjectType>Unit</ObjectType>
<ObjectName>ChemSepUO_382</ObjectName>
<Interface>ICapeUnit</Interface>
<Method>Calculate</Method>
<CallSequence>
<Call>
<ObjectType>Material Object</ObjectType>
<ObjectName>Mix out</ObjectName>
<Interface>ICapeThermoMaterialObject</Interface>
<Method>SetProp</Method>
<Arguments>
<Inputs>
<Row>
<Property>Volume</Property>
<Phase>Overall</Phase>
<Basis>Mole</Basis>
<Calc>Mixture</Calc>
</Row>
<Row>
<Property>My Property</Property>
<Phase>phaseValue</Phase>
<Basis>Molecule</Basis>
<Calc>Mkkkl</Calc>
</Row>

</Inputs>
<Outputs>
<Row>
<Value>0.000018075</Value>
</Row>
</Outputs>
</Arguments>
<ReturnCode>No Error</ReturnCode>
<CallSequence>
<Call>
<ObjectType>Unit</ObjectType>
<ObjectName>ChemSepUO_382</ObjectName>
<Interface>ICapeUnit</Interface>
<Method>Third Level of Recursion</Method>
<ReturnCode>No Error</ReturnCode>
</Call>
</CallSequence>

</Call> ...</pre>

Its the nesting of function calls (callsequence nodes in this case) which is causing the problem. The problem is that in the actual application I dont have access to the code that is calling those functions I need to log. So this nesting looks almost impossible for now.

The XML that is generated right now looks like this where function4 is actually called from Function1 but its not getting nested inside.

<pre><LOG>
<CallSequence>
<Call><MethodName>Function1</MethodName>
<Interface>ICapeThermo</Interface>
</Call>
<Call><MethodName>Function4</MethodName>
<Interface>ICapeThermo</Interface>
</Call>
</CallSequence></LOG></pre>

I'm in real hot water now... looks like I'm completely out of ideas...
Can u give me any idea or help in anyway?

Thanks for your time

Rocky

You can't climb up a ladder with your hands in your pockets.

James R. Twine18-Oct-07 3:12

James R. Twine

18-Oct-07 3:12

   As an aside, who came up with the requirement for using XML for logging? Based on the XML shown above, you have much more XML markup than content. Logging something like a crash/dump is one thing, but this looks like runtime-logging...

   Logging should not be a drain on performance. But XML, and XML combined with MSXML does not exactly scream high-performance...

   Peace!

-=- James
Please rate this message - let me know if I helped or not!<HR>If you think it costs a lot to do it right, just wait until you find out how much it costs to do it wrong!
Avoid driving a vehicle taller than you and remember that Professional Driver on Closed Course does not mean your Dumb Ass on a Public Road!
See DeleteFXPFiles

Rocky#18-Oct-07 19:28

18-Oct-07 19:28

well first of all, thank you so much for answering!

This idea was given by my boss and other seniors and well I think you're right but this task has been given to me n now I gotta do something abt it right?

At the moment this whole call Tree scenario looks pretty vague if u ask me!

Rocky

You can't climb up a ladder with your hands in your pockets.

led mike19-Oct-07 5:13

19-Oct-07 5:13

James R. Twine wrote:
But XML, and XML combined with MSXML does not exactly scream high-performance...

There are all kinds of production loggers that use XML. Performance is not a problem. "Logging" is not actually specific enough to determine what logging mechanism is appropriate.

James R. Twine19-Oct-07 7:09

James R. Twine

19-Oct-07 7:09

   A good and valid point.

   I was merely going on my experience with things that I have and do with on/with. And with these systems, performance does matter and can be a problem, so I cannot take for granted things that other developers get away with.

   Peace!

led mike18-Oct-07 4:25

18-Oct-07 4:25

I am not aware of any logging scheme that produces a call tree. It might not even make sense logically. With normal logging if you log every method entry there will be a natural sequence that is only upset in multi-threaded environments. In that case you can include the ThreadID in the log so they can be queried by thread to produce each threads actual sequence.

Rocky#18-Oct-07 19:33

18-Oct-07 19:33

well actually I dont think there's any multithreading involved here. but still the problem is that I have the functions which are called but I dont have the code that actually calls those functions. and I think to get the call tree right I need that part as well bcz you can't possibly judge inside a called function that at what level in the calling heirarchy it was called (I hope u get my point bcz it was a really hard one to explain Smile | :)

)

anyway thanks for ur time I appreciate this!

Rocky

You can't climb up a ladder with your hands in your pockets.

led mike19-Oct-07 5:13

19-Oct-07 5:13

Rocky# wrote:
but still the problem is that I have the functions which are called but I dont have the code that actually calls those functions.

I have no idea what that means.

Rocky#19-Oct-07 20:55

19-Oct-07 20:55

OK let's just forget it for the time being, I've told them I can't make up a Call Tree this whole idea is a mess. WTF | :WTF:

Rocky

You can't climb up a ladder with your hands in your pockets.

virtual destructor

George_George18-Oct-07 2:10

18-Oct-07 2:10

Hello everyone,

What is the purpose of virtual destructor? If currently, no derived class?

thanks in advance,
George

James R. Twine18-Oct-07 2:17

James R. Twine

18-Oct-07 2:17

Planning for the future.

Peace!

toxcct18-Oct-07 3:53

toxcct

18-Oct-07 3:53

James R. Twine wrote:
Planning for the future.

\o/

[VisualCalc][Binary Guide][CommDialogs] | [Forums Guidelines]

George_George18-Oct-07 4:08

18-Oct-07 4:08

Thanks toxcct,

Could you describe in more details please what advantages we could get if we have a derived class please? Compared with non-virtual destructor?

regards,
George

toxcct18-Oct-07 4:12

toxcct

18-Oct-07 4:12

WTF is the problem with you ?
i'm starting to be tired of you and your F*CKING voting mania.

i was not answering anything related to your question, i was just joking with James, so
1) why on hell do you vote me 2 ?
2) why did you down voted his post ?
3) why do you ask me about virtual desctructors if i don't talk to you ?
4) WTF ?! Mad | :mad:

[VisualCalc][Binary Guide][CommDialogs] | [Forums Guidelines]

Stephen Hewitt18-Oct-07 14:42

Stephen Hewitt

18-Oct-07 14:42

Chill out! It's a public forum.

Steve

George_George18-Oct-07 4:06

18-Oct-07 4:06

Thanks James,

Could you describe in more details please what benefits we could get if we have a derived class please? Compared with non-virtual destructor?

regards,
George

David Crow18-Oct-07 2:51

David Crow

18-Oct-07 2:51

George_George wrote:
What is the purpose of virtual destructor?

To ensure that the proper derived class destructor is called if an object of a derived class is deleted through a pointer to a base class. If a class has any virtual functions, it should have a virtual destructor.

George_George wrote:
If currently, no derived class?

Can you guarantee that your class will never be derived from? If so, then a virtual destructor is not necessary.

"Normal is getting dressed in clothes that you buy for work and driving through traffic in a car that you are still paying for, in order to get to the job you need to pay for the clothes and the car and the house you leave vacant all day so you can afford to live in it." - Ellen Goodman

"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne

George_George18-Oct-07 4:14

18-Oct-07 4:14

Thanks DavidCrow,

DavidCrow wrote:
Can you guarantee that your class will never be derived from? If so, then a virtual destructor is not necessary.

Your reply makes senses.

DavidCrow wrote:
To ensure that the proper derived class destructor is called if an object of a derived class is deleted through a pointer to a base class. If a class has any virtual functions, it should have a virtual destructor.

Could you show a simple sample (or some pseudo code) why *ensure that the proper derived class destructor is called* please? Compared with non-virtual destructor?

regards,
George