ATL / WTL / STL

14-May-04 6:22

I started with C++ recently Basically i am trying to overload the >> operator but couldn't succeed.The aim is to parse the string into different values.Infact this was provided with IBM open classes but not present with standard C++ string

#include <iostream>
#include <string>
using namespace std;

#define DELIM '~'

main()
{

string S="2~R~10/01/03~Personal Watercraft~Pers. Watercraft~9.0500~";
string s1,s2,s3,s4,s5,s6;

S >> S1 >> DELIM >> S2 >> DELIM >> S3 >> DELIM >> S;

S >> S4 >> DELIM >> S5 >> DELIM >> S6 >> DELIM;

}

Basically i am expecting the values between delimiters to be populated itno s1,s2...

I have declared this fucntion as in IOC
#ifdef __IOC_ANSI_STREAM
friend std::istream& operator>>(std::istream& aStream, string& aString)
#else
friend istream& operator>>(istream& aStream, string& aString)
#endif

Can you experienced guys let me know how to implement this or what has to be done to acheive the desired values.Thanks a lot in advance

Thanks
Sas

Andrew Walker14-May-04 16:56

Andrew Walker

14-May-04 16:56

I recommend that you don't try to overload the stream insertion and extraction operators for strings, developers expect their tools to work in consistant ways - and this isn't.

For really simple parsing jobs it's easier to just search the string for the delimiting character and copy that substring out. For more complex tasks use a regular expression library like boost::regex.

#include <vector>
#include <string>

std::vector< std::string > parseString(const std::string& strin)
{
    std::string str = strin;
    std::vector< std::string > words;
    size_t splt = str.find("~");

    while(splt != string::npos)
    {
        string out = str.substr(0,splt);
        words.push_back( out );
        str = str.substr(splt+1,str.length());
        splt = str.find("~");
    }
    words.push_back( str );
    return words;
}

If you can keep you head when all about you
Are losing theirs and blaming it on you;
If you can dream - and not make dreams your master;
If you can think - and not make thoughts you aim;
Yours is the Earth and everything that's in it.

Rudyard Kipling

ra_sasi17-May-04 4:10

17-May-04 4:10

Thanks a lot for the info.But then this will be great when we are writing a new program.But then when an application is to be ported from IBM Open classes to satndard C++/STL, i guess thsi is not enough.
Becasue while parsing IString,we can parse up to some portion and can retain the remaining in the original string itslef.Thats where i am becoming clueless as to how to impleemnt this in STL.

#include istring.hpp
#include istparse.hpp
main()
{
IString IS1= "12#34#56#78#90#ab";
IString I1,I2,I3,I4,I5,I6;
IS1>>I1>>DELIM>>I2>>DELIM>>I3>>DELIM>>I4>>DELIM>>IS1;
IS1>>I5>>DELIM>>I6;
}

Any ideas on how to acheive this type of functionality?? This is where i am not able to make out anything.

Thanks
Sas

Andrew Walker17-May-04 22:40

Andrew Walker

17-May-04 22:40

I'm assuming that you want to keep supporting some of the functionalities of IString - but want to start moving to STL? I had a quick look at the documentation of IString, it seems to have a conversion operator to char*, so you should be able to directly construct a std::string from an IString, provided your IString's will never have embedded nulls.

What follows is a (hack) way of extracting the data in a way similar to what you are doing, because the stream extraction operators will read until they reach whitespace (and AFAIK, you can't change the definition of whitespace). One function which gets around this limitation is getline, which (optionally) stops at a delimiting character. This won't work without modification under VC6 (requires an STL patch from dinkumware)

#include <iostream>
#include <sstream>

int main()
{
    std::istringstream iss("12#34#56#78#90#ab");
    std::string I1,I2,I3,I4,I5,I6;

    std::getline(iss,I1,'#');
    std::getline(iss,I2,'#');
    std::getline(iss,I3,'#');
    std::getline(iss,I4,'#');
    std::getline(iss,I5,'#');
    std::getline(iss,I6,'#');

    return 0;
}

'The Standard C++ Library' by Josuttis is a good reference if you get stuck, it covers all of this and more in depth.

ra_sasi20-May-04 12:10

Re: 8-bit graphics display

20-May-04 12:10

Thanks Andrew..i am enlightened by this..
I went by another approach by writing a new class to over load the extraction operator

Sas

ITableCreation

João Paulo Figueira12-May-04 7:28

João Paulo Figueira

12-May-04 7:28

What is the interface against which one typically queries
IID_ITableCreation? IID_IDBInitialize? IID_IDBCreateSession? I'm a bit lost
here.

Regards,
João Paulo Figueira
Embedded MVP

Re: ITableCreation

João Paulo Figueira12-May-04 23:52

João Paulo Figueira

12-May-04 23:52

I've sorted this one out: one has to query for ITableDefinitionWithConstraints...

Regards,
João Paulo Figueira
Embedded MVP

8-bit graphics display

ted_yu11-May-04 14:49

ted_yu

11-May-04 14:49

Among 256 colors in palette, 16 are allocated for
graphics and 240 are for grey-scale video.
I use same palette for selecting into hDC and for
making bitmap.
256 was returned by RealizePalette(hdc).
However, video quality under 8-bit display is worse
than 16-bit display.

Can you give me some clue ?

John M. Drescher11-May-04 16:50

John M. Drescher

11-May-04 16:50

I am not sure what you are asking. When you are using the windows api in 8 bit mode you can only display 240 different colors no matter what palette you try to use. This restriction does not exist if you use full screen directdraw. As in full screen directdraw you can use 254 colors + white + black. In 16 bit mode you have 65K colors so it can and will look better.

John

Re: 8-bit graphics display

ted_yu11-May-04 18:18

ted_yu

11-May-04 18:18

So why does RealizePalette() return 256 ?
You mean I should keep the size of my palette under 240 ? This way I will leave the top 16 empty.

R/

difference between MIDL Compiler Versions 6.00.0361 and 5.01.0164

Tarundeep Singh Kalra10-May-04 20:05

Tarundeep Singh Kalra

10-May-04 20:05

hi,
this is my observation , correct me if i am wrong .

I was defining a UDT which consist of a

[uuid(D8B3861A-74C6-11d2-0000-00C04FB17CDB)]
struct _BSSID_INFO
{
BSTR name;
short grade;
//classification type;
VARIANT_BOOL graduate;
};

in the IDL file.
Now when i was compiling with Default Visual Studio IDE Compiler it gave the
following errors :-

error MIDL2072 : inapplicable attribute : [uuid] [ Struct '_BSSID_INFO' ]
error MIDL2072 : inapplicable attribute : [uuid] [ Struct '_BSSID_INFO' (
Parameter 'bssid_info' ) ]
warning MIDL2039 : interface does not conform to [oleautomation] attribute :
[ Parameter 'bssid_info' of Procedure 'get_BSSID_INFO' ( Interface
'ICWiFiMgr' ) ]

but when i compiled with new vesrion of compiler and replaced it with new
MIDL.exe in the bin dir of VS6 it compiled correctly.

Any Inputs
--
Tarun..
Knowledge Seeker............
**********************************************************************
Tarundeep Singh Kalra
(tarundeep1976_nospam_at_yahoo_dot_nospam_dot_com)
tarun_at_tarunsadhana_dot_com
visit:www.tarunsadhana.com
*************************************************************************

Blue Pill or Red Pill........??

(WTL) XP style password edit box

Ernesto D.10-May-04 17:53

Ernesto D.

10-May-04 17:53

Hi!, my wtl app has a dialog that contains some edit boxes whit the password style, however, when i run the app and i type something in these boxes, instead of getting those nice "black dots" like you do on all XP password-edit boxes, i get some "weird" vertical lines, they kinda look like thick binary OR operators " | ".
any idea of what im doing wrong?

Thanks!

Jens Doose10-May-04 23:18

Jens Doose

10-May-04 23:18

You have to set the right font, I think it was Tahoma, but not 100% sure.

Jens

Jason De Arte11-May-04 9:38

Jason De Arte

11-May-04 9:38

I had the same problem a while back when I added a manifest to a MFC project - the password edit boxes had vertical lines instead of nice round circle characters.

The issue was that the edit control had the wrong font associated with it.

In that project I added the following in the dialogs OnInitDialog() function...
GetDlgItem(IDC_EDIT_PASSWORD)->SetFont( CFont::FromHandle((HFONT)GetStockObject(DEFAULT_GUI_FONT)) );

In WTL, it might* look something like this...

::SendMessage(GetDlgItem(IDC_EDIT_PASSWORD), WM_SETFONT, (WPARAM)AtlGetStockFont(DEFAULT_GUI_FONT), MAKELPARAM(FALSE, 0));

*your mileage may vary

[ Jason De Arte | Toy Maker | 1001010.com ]

Michael Dunn11-May-04 10:15

Michael Dunn

11-May-04 10:15

Make sure your dialog font is set to MS Shell Dlg and not MS Sans Serif. The dot thingy in XP is a Unicode character that MS Sans Serif does not contain.

--Mike--
Personal stuff:: Ericahist | Homepage
Shareware stuff:: 1ClickPicGrabber | RightClick-Encrypt
CP stuff:: CP SearchBar v2.0.2 | C++ Forum FAQ
----
Pinky, are you pondering what I'm pondering?
I think so Brain, but how will we fit the hamster inside the accordion?

Ernesto D.11-May-04 16:06

Ernesto D.

11-May-04 16:06

I set the font to Tahoma in resource editor and the problem dissapeared

Thanks guys!

streams problem

ra_sasi10-May-04 12:45

10-May-04 12:45

Hi,
i am facing some problem while working with streams in my C++ program using STL
main(){
..
..
char buf[256];
stringstream infLine(buf);
infline<

Re: streams problem

Joaquín M López Muñoz11-May-04 9:45

Joaquín M López Muñoz

11-May-04 9:45

Hi. First of all, the line

infline<<string1<<string2<<infline;

is a little suspicious. You sure you want to insert infline into itself? Won't you mean the following?

infline<<string1<<string2<<endl;

Now with the rest. I guess (only guess since you provide little info) you are trying to postprocess the results inspecting buf, but this is not how stringstream works. When constructing infline, the contents of buf are copied to its internal buffer, so infline won't use buf anymore. If you want to postprocess what's left inside of infline, use the str() member function:

stringstream infline; //no need to use buf
infline<<string1<<string2<<endl;
string result=infline.str(); // results copied to result

Does this help?

Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo

Re: streams problem

ra_sasi13-May-04 10:44