C / C++ / MFC

Richard MacCutchan1-Jan-15 4:51

1-Jan-15 4:51

Are you and Scholar247 the same person? If so please request removal of one profile.

As to your questions:

This site does not provide code to order. Google is the place for samples.
The consequences of a race condition may vary from incorrect results, to a program not responding.
There are many different situations when this may be appropriate, but it depends on the problem to be solved.
Impossible to answer as any problem's ease of resolution depends on a lot of factors.

Babu_Abdulsalam1-Jan-15 17:04

Babu_Abdulsalam

1-Jan-15 17:04

C++

int g_x = 0;
DWORD WINAPI Add(void* p) //p unused parameter
{
g_x++;
}
void main()
{
HANDLE m_hArr[10];
int i = 0;
for( ; i < 10; i++ )
{
CreateThread( NULL,0,Add,NULL,0,&m_hArr[i]);
}
//Wait for all the 10 threads to complete its execution
::WaitForMultipleObjects(10,m_hArr,TRUE,INDEFENITE);

//Now print the result.You expect 10 but it may not be... Smile | :)

cout<<"g_x = "<

Babu_Abdulsalam1-Jan-15 17:07

Babu_Abdulsalam

1-Jan-15 17:07

Because the order of the execution.
Suppose you have 4 floats a,b,c,d
and you want to sum up them.
In case of serial implementation, it will be like a + b + c + d
But in case of parallel, it may execute like
(a+b) + (c+d)
So the result wont be same.
This answers your 4th question.

Re: Multi-Core Programming

Babu_Abdulsalam1-Jan-15 17:09

Babu_Abdulsalam

1-Jan-15 17:09

2. Consequence of race condition:
Your resource will go to a corrupted state.
Not getting the expected result etc.,

Scholar2471-Jan-15 0:56

Scholar247

1-Jan-15 0:56

modified 2-Jan-15 9:24am.

Richard MacCutchan1-Jan-15 4:45

Re: Multi-Core Programming

1-Jan-15 4:45

Your question is not clear. What is the connection between mult-threading and rounding errors?

OriginalGriff1-Jan-15 5:39

OriginalGriff

1-Jan-15 5:39

Richard MacCutchan wrote:
What is the connection between mult-threading and rounding errors?

His homework question? Laugh | :laugh:

Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...

Re: Multi-Core Programming

Richard MacCutchan1-Jan-15 5:57

YAAQ Yet another academic question - overloaded or "strongly typed"?

1-Jan-15 5:57

Thanks, I'm not an expert in rounding, but I will stand a round (or around).

Vaclav_31-Dec-14 14:10

Vaclav_

31-Dec-14 14:10

Overheard many years ago that "C is strongly typed language".
So what's up with unspecified / unknown "standard definition " of int abs (int) overloaded by , again unknown source "standard", to <b>float abs ( float)</b>?
Is it just "progress" AKA from plain C to C "whatever is latest derivative of it" or just plain lack of real standards ?
Happy coding in 2015
Cheers Vaclav

Re: YAAQ Yet another academic question - overloaded or "strongly typed"?

Richard MacCutchan31-Dec-14 23:03

Re: YAAQ Yet another academic question - overloaded or "strongly typed"?

31-Dec-14 23:03

That's not C, it's C++; see the Remarks section at http://msdn.microsoft.com/en-us/library/kw1ee768.aspx[^] and http://msdn.microsoft.com/en-us/library/18z15bk0.aspx[^].

Vaclav_14-Jan-15 6:02

Vaclav_

14-Jan-15 6:02

Sorry I did not specifically say "C++".
I should have said "is C++ ( and derivatives ) strongly typed " to make the question clearer.
Maybe the question is just irrelevant with overloading, thus academic as I said in title.
Cheers Vaclav

Re: YAAQ Yet another academic question - overloaded or "strongly typed"?

Richard MacCutchan14-Jan-15 6:19

Re: YAAQ Yet another academic question - overloaded or "strongly typed"?

14-Jan-15 6:19

Exactly, you said "Overheard many years ago that "C is strongly typed language".", and went on to talk about overloading. I responded that it (overloading) was not C it was C++. So I apologise that my answer was not very clear.

Mike Nordell2-Feb-15 11:07

Mike Nordell

2-Feb-15 11:07

I should have said "is C++ ( and derivatives ) strongly typed "

No need. You just mixed concepts when including overloading - that's like comparing apples with the act of driving a car. Smile | :)

So restricting the question to C, it is indeed a strongly typed language, but in some instances not entirely strict about it. C++ is stricter.

For further reading, see f.ex. this wikipedia page.

Enabling/Disabling a CMFCRibbonButton

Richard Andrew x6430-Dec-14 18:24

30-Dec-14 18:24

How does one arbitrarily enable or disable a button on the Ribbon bar?

The OnUpdateCommand handler only allows you to specify a change in response to a button-click. But I want to enable or disable the button from elsewhere in the program.

The difficult we do right away...
...the impossible takes slightly longer.

Richard MacCutchan30-Dec-14 22:06

30-Dec-14 22:06

I think you want the ON_UPDATE_COMMAND_UI macro[^].

Richard Andrew x6431-Dec-14 7:38

31-Dec-14 7:38

Thanks for your response.

That macro allows me to add an OnUpdateCommandUI handler which gets called right after the button is clicked.

But how does it allow me to do something like ribbonButton.Enabled(FALSE); NOT in response to a button click?

The difficult we do right away...
...the impossible takes slightly longer.

Richard MacCutchan31-Dec-14 22:55

31-Dec-14 22:55

I guess it's just a matter of creating a function that can enable or disable the button, something like:

C++

void EnableButton(CMFCRibbonButton& ribbonButton, bool bEnable)
{
    ribbonButton.Enable(bEnable);
}

or better still, a function that enables/disables all the buttons based on a set of rules ...

C++

void EnableButtons(int nRule, bool bEnable)
{
    switch (rule)
    {
    case 1:
        ribbonButton1.Enable(bEnable);
        ribbonButton2.Enable(bEnable);
        break;
    case 2:
        ribbonButton1.Enable(bEnable);
        ribbonButton6.Enable(bEnable);
        ribbonButton7.Enable(bEnable);
        break;
    // ... etc
    }
}

You then call this function from elsewhere in response to some arbitrary decision of your own.

Richard Andrew x641-Jan-15 8:28

1-Jan-15 8:28

Now you've hit upon the exact problem: The CMFCRibbonButton does not have an "Enable" method, or any method like it, hence my dilemma.

Thanks again for your time.

The difficult we do right away...
...the impossible takes slightly longer.

Richard MacCutchan1-Jan-15 22:17

1-Jan-15 22:17

That seems strange, since the base class has an IsDisabled() function. I also noticed in Word2010 that some buttons are disabled so it must be possible. The SO question at http://stackoverflow.com/questions/502640/disable-enable-ribbon-buttons-for-mfc-feature-pack[^] suggests that it uses the normal MFC macros as I suggested earlier.

Richard Andrew x642-Jan-15 10:10

2-Jan-15 10:10

I hear what you're saying, but that IsDisabled() property is read-only. And the SO question you link to says basically what we already know about the MFC macros.

That's why I can't figure this out. Sigh | :sigh:

The difficult we do right away...
...the impossible takes slightly longer.

Richard MacCutchan2-Jan-15 10:41

Am a newbie to PARALLEL VERSION USING MPI , I need to convert the code below to parallel version using mpi.

2-Jan-15 10:41

Sorry nor can I. But given what I said about Word and the SO question, it must be possible. Unfortunately it is too many years since I used MFC in anger so I can't even try a few things. I wonder where all the CodeProject MFC experts are?

Member 1124308429-Dec-14 4:57

Member 11243084

29-Dec-14 4:57

/*
TEST WITH THE FOLLOWING

PROBLEM 1

inputs:
infinity:999
no. of cities: 4
no. of paths:6

PROBLEM 2

inputs:
infinity:999
no. of cities: 5
no. of paths:10

*/

#include <stdio.h>
#define ALL -1
#define MAXCITIES 10

enum BOOL{FALSE,TRUE};
long*visited;//visited nodes set here
long*min_circuit;//min inner circuit for given node as start node at position indexed 0
long*ham_circuit;//optimal circuit with length stored at position indexed 0
long min_circuit_length;//min circuit lenth for given start node

int n;//city count
long matrix[MAXCITIES][MAXCITIES];//nondirectional nXn symmetric matrix
//to store path distances as sourceXdestination
long INFI;// INFINITY value to be defined by user
// function resets minimum circuit for a given start node
//with setting its id at index 0 and setting furthr node ids to -1

void reset_min_circuit(int s_v_id)
{
min_circuit[0]=s_v_id;
for(int i=1;i<n;i++) min_circuit[i]="-1;
}
//" marks="" given="" node="" id="" with="" flag
="" if="" it="" all="" nodes="" flag

void="" set_visited(int="" v_id,bool="" flag)
{
="" if(v_id="=ALL)" for(int="" i="0;i<n;i++)" visited[i]="flag;
" else="" visited[v_id]="flag;
}
//" function="" sets="" hamiltonion="" circuit="" for="" a="" path="" length
="" setting="" at="" index="" 0="" and="" furthr="" from="" current="" min_circuit

void="" set_ham_ckt(long="" pl)
{
="" ham_circuit[0]="pl;
" ham_circuit[i+1]="min_circuit[i];
" ham_circuit[n+1]="min_circuit[0];
}

//function" valid="" by="" finiding="" min="" inner="" given
="" combination="" start="" vertex="" next="" to="" such="" that
="" the="" 2nd="" of="" circuits="" is="" always="" s_n_v="" dest="" is
="" s_v="" possible="" values="" s_n_v,="" then="" returns="" the
="" length="" this="" combination

long="" get_valid_circuit(int="" s_v,int="" s_n_v)
{
="" int="" next_v,min,v_count="1;
" long="" path_length="0;
" min_circuit[0]="s_v;
" min_circuit[1]="s_n_v;
" set_visited(s_n_v,true);
="" path_length+="matrix[s_v][s_n_v];
" v="s_n_v;v_count<n-1;v_count++)
" {="" if(="" matrix[v][i]<infi="" &&="" !visited[i]
="" matrix[v][i]<="min
" )
="" set_visited(next_v,true);
="" }
="" return(path_length);
}

int="" main()
{
=""
="" printf("make="" sure="" that="" infinity="" value="" <="" sum="" distances\nset="" (signed="" long):");
="" scanf("%ld",&infi);
="" pathcount,i,j,source,dest;
="" dist="0;
" new_circuit_length="INFI;
" printf("enter="" no.="" cities(max:%d):",maxcities);
="" scanf("%d",&n);
="" count:");
="" scanf("%d",&pathcount);

="" paths:<="" source_id="" destination_id="" distance="">\n ids varying from 0 to %d\n",n-1);
//init all matrix distances to infinity
for(i=0;i<n;i++)
for(j="0;j<n;j++)
" matrix[i][j]="INFI;
//populate" the="" matrix
=""
="" for(i="0;i<pathcount;i++)
" return="" 0;
="" {
="" printf("[path="" %d]:",i);
="" scanf("%d="" %d="" %ld",&source,&dest,&dist);
="" if(source!="dest)
" matrix[source][dest]="matrix[dest][source]=dist;
" }

="" visited="new" long[n];
="" min_circuit="new" ham_circuit="new" long[n+2];
="" min_circuit_length="INFI;
" algorithm
="" for="" each="" vertex,="" s_v="" as="" a="" staring="" node
="" for(int="" s_v_id="0;S_V_id<n;S_V_id++)
" {="" and="" non="" start="" vertex="" i
="" set="" all="" to="" unvisited
="" set_visited(all,false);
="" visited
="" set_visited(s_v_id,true);
="" reset="" init="" minimum="" circuit
="" reset_min_circuit(s_v_id);
="" obtain="" circuit="" combination="" of="" new_circuit_length="get_valid_circuit(S_V_id,i);
" if="" newer="" length="" is="" less="" than="" previously
="" calculated="" min="" then="" it="" the
="" current="" in="" hamiltonion="" if(new_circuit_length<="min_circuit_length)
" set_ham_ckt(min_circuit_length="new_circuit_length);
" }
="" any="" found
if(min_circuit_length<infi)
{
="" printf("\n\nminimum="" is:="" %ld\ncircuit="" is:\n",min_circuit_length);
="" printf("<%ld=""> ",ham_circuit[i]);
}
else printf("\n\nNo hamiltonian circuit !");

delete []visited;
delete []min_circuit;
delete []ham_circuit;
}

Re: Am a newbie to PARALLEL VERSION USING MPI , I need to convert the code below to parallel version using mpi.

Richard MacCutchan29-Dec-14 23:03