D - strings & windows programming

Robert M. M�nch (36/36) Jul 08 2002 Hi, I'm trying to do some Windows programming by hand to get a feeling h...

Sean L. Palmer (18/48) Jul 08 2002 the

Robert M. M�nch (15/35) Jul 09 2002 that

Sean L. Palmer (14/51) Jul 09 2002 this

OddesE (13/26) Jul 13 2002

"Robert M. M�nch" <robert.muench robertmuench.de> writes:

Hi, I'm trying to do some Windows programming by hand to get a feeling how
to use D with the MS SDK. Well, I have some problems/questions. This is the
code snippet. I try to read the user input from an edit-control:

    // get the number of characters
    int numInput = (WORD) SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE,
        EM_LINELENGTH, (WPARAM) 0, (LPARAM) 0);

D stores the length of a string first in an array. Am I right that this
informaiton is internal to D and can only be access thru the .legnth
attribute or can the size of the array be accessed with something like
length = myarry[0]?

So if the length is internal, than it's necessary to store the length in the
array for Windows:

    // Put the number of characters into first entry of buffer. For this we
have to expand the string by hand
    char people[]; people.length = numInput + 1;
    people[0] = (char)numInput;

Wouldn't it be convinient to be able to do?

    char people[];
    people[0] = (char)numInput;

Here the array should be expanded automatically. The index is known so that
it's clear how big people[] needs to be at least.

Ok, now I want Windows to copy the text from the edit control into my
string:
      SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
(LPARAM) &people);

Well this line ends up with rubish in people[]. I have to call it this way:
      SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
(LPARAM) &people[0]);

Why do I have to use the &people[0] syntax? Shouldn't this be the same
&people = &people[0]?

--
Robert M. M�nch
IT & Management Freelancer
Mobile: +49 (0)177 2452 802
Fax   : +49 (0)721 8408 9112
Web   : http://www.robertmuench.de

Jul 08 2002

"Sean L. Palmer" <seanpalmer earthlink.net> writes:

"Robert M. M�nch" <robert.muench robertmuench.de> wrote in message
news:agceq3$b2r$1 digitaldaemon.com...
 Hi, I'm trying to do some Windows programming by hand to get a feeling how
 to use D with the MS SDK. Well, I have some problems/questions. This is

the
 code snippet. I try to read the user input from an edit-control:

     // get the number of characters
     int numInput = (WORD) SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE,
         EM_LINELENGTH, (WPARAM) 0, (LPARAM) 0);

 D stores the length of a string first in an array. Am I right that this
 informaiton is internal to D and can only be access thru the .legnth
 attribute or can the size of the array be accessed with something like
 length = myarry[0]?

You're right.  It can only be accessed via .length attribute.  In fact, it's
not a char either; I believe it's an int.

 So if the length is internal, than it's necessary to store the length in

the
 array for Windows:

If you mean null-terminate it, well, that's easy:   mystring ~= cast(char)0;

     // Put the number of characters into first entry of buffer. For this

we
 have to expand the string by hand
     char people[]; people.length = numInput + 1;
     people[0] = (char)numInput;

 Wouldn't it be convinient to be able to do?

     char people[];
     people[0] = (char)numInput;

 Here the array should be expanded automatically. The index is known so

that
 it's clear how big people[] needs to be at least.

I don't know why you'd want to be able to do this.

 Ok, now I want Windows to copy the text from the edit control into my
 string:
       SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
 (LPARAM) &people);

 Well this line ends up with rubish in people[]. I have to call it this

way:
       SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
 (LPARAM) &people[0]);

 Why do I have to use the &people[0] syntax? Shouldn't this be the same
 &people = &people[0]?

No, and you shouldn't depend on undocumented compiler-dependent stuff like
that anyway.  If you want the address of the first character in a string,
you should say so explicitly.  The string (like any struct) could have other
things in it before the actual string data.  In this case, the length.

If this is too tedious, remember that you can write a wrapper class that
deals with edit controls, and you only have to write that class *once*.

Sean

Jul 08 2002

"Robert M. M�nch" <robert.muench robertmuench.de> writes:

"Sean L. Palmer" <seanpalmer earthlink.net> schrieb im Newsbeitrag
news:agcm7b$isb$1 digitaldaemon.com...

 Wouldn't it be convinient to be able to do?

     char people[];
     people[0] = (char)numInput;

 Here the array should be expanded automatically. The index is known so


that
 it's clear how big people[] needs to be at least.

 I don't know why you'd want to be able to do this.

Hi, IMO if we have dynamic arrays these are either dynamic or not but no
mish-mash. So why can't I write the following:

char test[];
test[123] = 'a';

If test.length was 0 than after this line it will be at least 123. IMO this
makes sense. Why should I have to start at the beginning?

 Well this line ends up with rubish in people[]. I have to call it this


way:
       SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
 (LPARAM) &people[0]);

 Why do I have to use the &people[0] syntax? Shouldn't this be the same
 &people = &people[0]?

 No, and you shouldn't depend on undocumented compiler-dependent stuff like
 that anyway.

This is undocumented? Hm... didn't knew about this. So what's the correct
way to do it in D?

 If you want the address of the first character in a string,
 you should say so explicitly.

That's what I tried first with &people but this didn't worked.

 If this is too tedious, remember that you can write a wrapper class that
 deals with edit controls, and you only have to write that class *once*.

I know but first I need to get the basis right and than I think about
re-using :-)). Robert

Jul 09 2002

"Sean L. Palmer" <seanpalmer earthlink.net> writes:

"Robert M. M�nch" <robert.muench robertmuench.de> wrote in message
news:age639$24ld$1 digitaldaemon.com...
 "Sean L. Palmer" <seanpalmer earthlink.net> schrieb im Newsbeitrag
 news:agcm7b$isb$1 digitaldaemon.com...

 Wouldn't it be convinient to be able to do?

     char people[];
     people[0] = (char)numInput;

 Here the array should be expanded automatically. The index is known so


 that
 it's clear how big people[] needs to be at least.

 I don't know why you'd want to be able to do this.

 Hi, IMO if we have dynamic arrays these are either dynamic or not but no
 mish-mash. So why can't I write the following:

 char test[];
 test[123] = 'a';

 If test.length was 0 than after this line it will be at least 123. IMO

this
 makes sense. Why should I have to start at the beginning?

Because you don't want the compiler to have to insert hidden checks to make
sure the array is big enough and potentially reallocate it before *every*
single element assignment.

 Well this line ends up with rubish in people[]. I have to call it this


 way:
       SendDlgItemMessageA(hWnd, IDC_NUMPEOPLE, EM_GETLINE, (WPARAM) 0,
 (LPARAM) &people[0]);

 Why do I have to use the &people[0] syntax? Shouldn't this be the same
 &people = &people[0]?

 No, and you shouldn't depend on undocumented compiler-dependent stuff


like
 that anyway.

 This is undocumented? Hm... didn't knew about this. So what's the correct
 way to do it in D?

The location in physical memory of the length property is undocumented and
compiler-specific.  Or it should be.

 If you want the address of the first character in a string,
 you should say so explicitly.

 That's what I tried first with &people but this didn't worked.

That's because you asked for the address of the dynamic array, not the
address of the first element.  Say what you mean.  And when programming D,
you have to say it in D, not some other language.  ;)

 If this is too tedious, remember that you can write a wrapper class that
 deals with edit controls, and you only have to write that class *once*.

 I know but first I need to get the basis right and than I think about
 re-using :-)). Robert

hehehe

Sean

Jul 09 2002

"OddesE" <OddesE_XYZ hotmail.com> writes:

"Sean L. Palmer" <seanpalmer earthlink.net> wrote in message
news:agcm7b$isb$1 digitaldaemon.com...
 "Robert M. M�nch" <robert.muench robertmuench.de> wrote in message
 news:agceq3$b2r$1 digitaldaemon.com...


<SNIP>
 Why do I have to use the &people[0] syntax? Shouldn't this be the same
 &people = &people[0]?

 No, and you shouldn't depend on undocumented compiler-dependent stuff like
 that anyway.  If you want the address of the first character in a string,
 you should say so explicitly.  The string (like any struct) could have

other
 things in it before the actual string data.  In this case, the length.

 If this is too tedious, remember that you can write a wrapper class that
 deals with edit controls, and you only have to write that class *once*.

 Sean

And I think Pavel is busy doing just that, writing classes
for interfacing with Windows stuff such as windows, edit
controls etc...


--
Stijn
OddesE_XYZ hotmail.com
http://OddesE.cjb.net
_________________________________________________
Remove _XYZ from my address when replying by mail

Jul 13 2002

D Programming

C/C++ Programming

Other

D - strings & windows programming