C / C++ / MFC

Re: IRichEditOle::InsertObject shifts text one byte

22-Dec-19 22:20

I am not sure that I can see what is wrong with that. Anything inserted into the control will take space and thus move everything else along.

ForNow23-Dec-19 1:04

ForNow

23-Dec-19 1:04

Richard

I think the selection range has to be in line with the number of characters that are in

C++

LPCTSTR lpszNewText,

of the null terminated string parameter of ReplaceSel

When you InsertObject a bitmap that bitmap depending on its size takes up a certain amount of characters depending among other things the size of font of the

current selection and the size of the bitmap. I am not sure if there is a some way to calculate this i.e. if you have a string and you want to know how many pixels it takes you do GetTextExtent. the REOBJECT structure doesn't seem to have parameters for the size of the bitmap like BITBLT I just played around with the string and the selection size and got it to align

Re: IRichEditOle::InsertObject shifts text one byte

Richard MacCutchan23-Dec-19 2:03

CASE SOLVED AND CLOSED cout and cerr execute out of coded sequence

23-Dec-19 2:03

Most image files and types have the size encoded in them. I don't know how you are loading the bitmap into the reobject variable but presumably you can check the source file/object.

Vaclav_21-Dec-19 10:13

Re: SOLVED cout and cerr execute out of coded sequence

21-Dec-19 10:13

CASE SOLVED AND CLOSED
FYI
As I pointed out in my recent post, which apparently did no good, this has been brought up before.
I am guilty of not going back to the original post.
Being repeatedly told that I am not sharing my code is getting old.
So - no need to be bothered again , let's just chalk this for yet another loop-sided discussion blaming the problem on me.

Cheers

CASE CLOSED

Re: Unexpected sequence of cout / perror in console output - C / C++ / MFC Discussion Boards[^]

SOLVED

It is acceptable and normal to direct standard output and standard error to the same destination, such as the text terminal. Messages appear in the same order as the program writes them, unless buffering is involved. For example, in common situations the standard error stream is unbuffered but the standard output stream is line-buffered; in this case, text written to standard error later may appear on the terminal earlier, if the standard output stream buffer is not yet full.

This is NOT a dupe, just an attempt to analyse / concentrate on specific problem.
The attached code execution clearly shows that the code is NOT run in sequence coded.
See line numbers - they are OUT of sequence.
The lines start with 40 and end with line 49

This is superficial, but annoying problem I would like to resolve.
Or just identify the reason for this behaviour.
BTW - inserting sleep(1) at various places "solves the problem".

cerr is used to output (stderr) to console in red AND WORKS just fine.

Maybe cerr and cout have conflicting "priority" or "4 core processor " in use is an issue.

C++

cout << " START test area " << __LINE__ << endl;
cout << " START test area " << __LINE__ << endl;
cerr << "Function: " << __FUNCTION__ << " @line " << dec << __LINE__
        << endl;
//sleep(1);
//int i = 0;
cerr << "TRACE TEST Function: " << __FUNCTION__ << " @line " << dec
        << __LINE__ << endl;
cout << "\tEND test area " << __LINE__ << endl;
cout << "\tEND test area " << __LINE__ << endl;

 START test area 40
 START test area 41
Function: 	END test area 48
	END test area 49
main @line 42
TRACE TEST Function: main @line 47

modified 22-Dec-19 9:44am.

Richard MacCutchan21-Dec-19 21:32

Re: CASE SOLVED AND CLOSED cout and cerr execute out of coded sequence

21-Dec-19 21:32

I still cannot get the output in the "wrong" order. As mentioned below, I think there is a key piece of information that you have not shared with us.

Richard MacCutchan22-Dec-19 22:55

SOLVED CASE CLOSED cout in color and perror - revisited

22-Dec-19 22:55

Vaclav_ wrote:
Being repeatedly told that I am not sharing my code is getting old.

And there is an easy way to prevent that ...

Vaclav_20-Dec-19 4:46

20-Dec-19 4:46

SOLVED CASE CLOSED
Thanks for all the help from the group.
It is much appreciated.

Ask Mrs Google for "Eclipse ANSI" .
The question remains why it works without the 3rd part plug in when the app is crosscompiled.

I hope it will not bother the group if I revisit this , now multiple issue.
1. The attached C++ code DOES NOT execute in expected sequence.
Both perror output in red, which is expected , but out of expected sequence.

2. The setw works as expected.

3. The cout "escape sequence " DOES NOT change color and also outputs the sequence as "control characters".
It works well in crosscomplied code.

The code fails using tools implementing "standard" "hello word" on local machine.

This SAME code works as expected running after the application is crosscompiled for
"remote" machine. And has been working well for few months of development.

As far as I can tell - implementing "control / escape " characters in cout does not require anything special - iostream should work .

C++

	cout << " START test area " << __LINE__ << endl;
	cout << __FUNCTION__ << endl;
	cout << " START test area " << __LINE__ << endl;
	cout << __FUNCTION__ << endl;
	perror("TEST perror ");
	cout << setw(30) << "\033[1;31mbold red text\033[0m\n";
	perror("TEST perror ");
	while(1);

<pre lang="text">
TEST perror : Success             printed in red 
TEST perror : Invalid argument   printed in red 
 START test area 334
main
 START test area 336    
main
     [1;31mbold red text[0m    shifted , but no color change

modified 24-Dec-19 10:59am.

leon de boer20-Dec-19 5:01

leon de boer

20-Dec-19 5:01

With cout you can do it directly without the escape code

cout << ansi::foreground_red << "bold red text" << ansi::foreground_white << endl;

In vino veritas

Vaclav_20-Dec-19 5:48

20-Dec-19 5:48

OK, stupid follow-up question.

How / where I activate "ansi"?

Richard MacCutchan20-Dec-19 6:07

20-Dec-19 6:07

I just tried your code after correcting it, and it worked fine printing "bold red text" in bold red. What is that while(1); statement supposed to be for?

Also, why are your results shown in a different order to the actual instructions in your code sample?

Vaclav_21-Dec-19 3:10

21-Dec-19 3:10

I suppose asking to share your solution is out of the question.
Since this is a TEST code , the while(1); purpose is to stop code flow.

Richard MacCutchan21-Dec-19 3:34

21-Dec-19 3:34

It's nothing special, just print the start test line, then two samples of perror, the first with no error, and the second with error 5.

C++

cout << "Function: " << __FUNCTION__ << endl;
cout << " START test area " << __LINE__ << endl;
errno = 0;
perror("\033[1;31mTEST perror "); // print the error details in red
cout << "\033[0m\n";              // restore to standard colours
errno = 5;
perror("\033[1;31mTEST perror ");
cout << "\033[0m\n";

Quote:
Since this is a TEST code , the while(1); purpose is to stop code flow.

Not sure what that means as there is nothing to stop.

Vaclav_21-Dec-19 9:05

21-Dec-19 9:05

Unfortunately that is not a solution.

Edited
perror always prints in red - using stderr / cerr .
Maybe I should just switch from cout to cerr to output in red,

Please observe that escape codes, which are still ignored, are printed instead of being reacted on.

That is THE ISSUE, the usage of while is NOT related to the issue at all.

I believe this is related either to complier or operating system.
I am searching for a solution in that direction for now.

Any other suggestions are welcome.

C++

cout << "Function: " << __FUNCTION__ << endl;
cout << " START test area " << __LINE__ << endl;
errno = 0;
perror("TEST perror "); // default print in red
perror("\033[1;31mTEST perror "); // print the error details in red
cout << "\033[0m\n";              // restore to standard colours
errno = 5;
perror("\033[1;31mTEST perror ");
cout << "\033[0m\n";

printf("%c[%dmHELLO!\n", 0x1B, 32);
const std::string red("\033[0;31m");
cout << red << "red text" << endl;

system("Color 1A");
system("Color 1B");
std::cout << "\t\t\t    Hello World" << std::endl;

cout << "Bluetooth client " << endl;

cout << " START test area " << dec << __LINE__ << endl;
cout << __FUNCTION__ << endl;

Function: main
 START test area 39
[0m
[0m
[32mHELLO!
[0;31mred text
TEST perror : Success
[1;31mTEST perror : Invalid argument
[1;31mTEST perror : Input/output error
sh: 1: Color: not found
sh: 1: 			    Hello World
Bluetooth client 
 START test area 58
main
set color @line   [31m 443
set color @line   [31m 449
PROGREESS TRACE START

modified 21-Dec-19 15:38pm.

Richard MacCutchan21-Dec-19 21:25

How to close file in C++?

21-Dec-19 21:25

Taking your code above and running it produces the following output:

C:\Users\rjmac\Documents\Code\C++>Test
Function: DoTest
 START test area 17
TEST perror : No error               --> in normal black
TEST perror : No error               --> in bold red

TEST perror : Input/output error     --> in bold red

HELLO!                               --> in green
red text                             --> everything from here in normal red
                            Hello World
Bluetooth client
 START test area 36
DoTest

Which does not match with what you have produced. I can only conclude that there is something else going on in your system, or the actual code you run is different, that you are not telling us about.

Vaclav_19-Dec-19 17:13

19-Dec-19 17:13

Yet another "problem" got me stumped.
To verify "networking" I use system call such as

system("vim /etc/hosts");

It does the job, however, I cannot figure out how to close the file.
The standard "vim" way is "Esc :q!" , and it is unclear which character should be "send" via standard console and which ones after responses from vim.
Either manually (cin) or as a string starting with "Esc".

Cheers

CPallini19-Dec-19 21:35

CPallini

19-Dec-19 21:35

What kind of verification does vim?
I am asking such a question because, I believe, while it is not straightforward closing vim, there could be better alternatives to 'verify networking'.

Stefan_Lang19-Dec-19 21:59

Stefan_Lang

19-Dec-19 21:59

I wondered about that too: if the point is just to verify that you can access a certain file across a network file path, then trying to open it with fstream::open() should do the job.

Whereas, emulating console input that is supposed to be passed into an external process would be kind of tricky, to say the least...

GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)

Vaclav_20-Dec-19 4:51

20-Dec-19 4:51

I am just looking to verify "hosts" names.
And I realize that using vim is asking for complications, especially when I actually need to use / select "name" for further processing.
vim is bad idea.

Richard MacCutchan19-Dec-19 21:56

Do "setw(offset) " BEFORE cout ?

19-Dec-19 21:56

If vim is not in insert or append mode then you can close it by sending ZZ.

Vaclav_18-Dec-19 4:22

18-Dec-19 4:22

OK, I can do search and replace "cout<<" with "cout<< setw(offset)<< "
to shift the output start column to console at offset.

BUT

anybody has an idea / hack how to PRESET "cout" ?

C++

<pre>	int offset = 10;

	setw(offset);            // preset to offset , but this does not work , obviously 

	cout << setw(offset ) << "Test " <<  endl;
	cout << setw(offset ) << "Test " <<  endl;
	cout << setw(offset ) << "Test " <<  endl;

	offset += 10;
	cout << setw(offset ) << "Test " <<"  space " << "Test " <<  endl;
	cout << setw(offset ) << "Test " <<"  space " << "Test " <<  endl;
	cout << setw(offset ) << "Test " <<"  space " << "Test " <<  endl;
	cout << setw(offset ) << "Test " <<"  space " << "Test " <<  endl;

Richard MacCutchan18-Dec-19 4:56

18-Dec-19 4:56

No, The width feature affects the following item only. In the same way you cannot have a preset in printf.

leon de boer18-Dec-19 5:19

leon de boer

18-Dec-19 5:19

Usually when doing that sort of thing its easier to swing to ANSI which is semi portable on Windows and Linux consoles and you can have colours. Still cant set it global pre calls.

An example

printf("\x1B[%d;%df\x1B[31;40mRed line at 10,20\n", 10, 20);
printf("\x1B[37;40m");

In vino veritas

Vaclav_19-Dec-19 17:26

19-Dec-19 17:26

Leon,
I am getting used to using cout, despite of its irrational behaviour.
In my main project I went overboard with debugging colouring scheme and it sort of works.
I just started the "client" part of the project and reusing a class from my main project where cout colouring and cout lines work just as expected.

In derived project - no coloring, cout lines have extra line feed and I am getting non printable characters to boot.
Just another bug to find.

BTW
I am finding out this setw does not always work....

Richard MacCutchan19-Dec-19 21:50