Why does Microsoft's code examples have errors, and how do I fix these?

Question

Why does Microsoft's code examples have errors, and how do I fix these?

CDev-8220 385

The code examples on this page all seem to have errors, some of which I don't have a clue how to fix.

How do I fix the following:

code:

hResult = StringCchLength(pmyitem->psz, STRSAFE_MAX_CCH, pcch);

error:

argument of type "LPSTR"is incompatible with parameter of "STRSAFE_PCNZWCH"

code:

GetTextExtentPoint32(hdc, pmyitem->psz, *pcch, &size);
ExtTextOut(
	lpdis->hDC, 
	nTextX, 
	nTextY, 
	ETO_OPAQUE, 
	&lpdis->rcItem, 
	pmyitem->psz, 
	*pcch, NULL
);

error:

argument of type "LPSTR"is incompatible with parameter of "LPCWSTR"

RLWA32 49,641 Reputation points

2024-04-08T09:32:56.3833333+00:00

@Castorix31 and @Minxin Yu Sorry about the duplicative answer. Unfortunately, the continuing failure of the Q&A site to timely update for posts of comments/answers meant that your answers were not visible when I made my post.
Minxin Yu 13,506 Reputation points Microsoft External Staff

2024-04-11T07:56:31.2133333+00:00

Is your problem solved?

Accepted answer

2 additional answers

Your answer

RLWA32 49,641 Reputation points

2024-04-08T09:32:56.3833333+00:00

@Castorix31 and @Minxin Yu Sorry about the duplicative answer. Unfortunately, the continuing failure of the Q&A site to timely update for posts of comments/answers meant that your answers were not visible when I made my post.
Minxin Yu 13,506 Reputation points Microsoft External Staff

2024-04-11T07:56:31.2133333+00:00

Is your problem solved?

Answer 1

RLWA32 49,641

Windows API functions usually have two versions. There is a "W" version when building for UNICODE and an "A" version for non-UNICODE builds. For example, when building for UNICODE the compiler sees a function named ExtTextOutW which expects wide-character strings. However, the example code declares some variables as LPSTR which means it is a pointer to a narrow string. This causes the compiler errors.

The simplest solution is to change the character set that you are building for to either "Not Set" or MBCS. That should eliminate the kinds of errors you have described.

RLWA32 49,641 Reputation points

2024-04-11T16:54:59.4333333+00:00

@CDev-8220, You can set the character set for a C++ project using the project property shown below. You will probably want to change the character set for all platform/configuration combinations.
CDev-8220 385 Reputation points

2024-04-12T01:27:32.8933333+00:00

This cleaned up the errors, thanks.

However, I was trying to figure out which is more advantageous - 'Not Set' or MBCS, and I came across this page, which says:

Multibyte character sets (MBCSs) are an older approach to the need to support character sets, like Japanese and Chinese, that cannot be represented in a single byte. If you are doing new development, you should use Unicode for all text strings except perhaps system strings that are not seen by end users. MBCS is a legacy technology and is not recommended for new development.

String handling

In an application that uses MBCS, string handling poses special problems. Characters of both widths are mixed in a single string; therefore, you must remember to check for lead bytes.

Would it be better to use 'Not Set', and then use the tchar.h header file, to build single-byte, MBCS, and Unicode applications from the same sources - Enable both Unicode and MBCS in my program.

What's your experience, and what might you suggest?
RLWA32 49,641 Reputation points

2024-04-12T11:44:04.1533333+00:00

The character set property is principally used to determine how the compiler will interpret the generic mappings for Windows API functions and CRT library functions and also how the declaration of text variables will be resolved as either narrow-character strings or wide-character strings.

The generic mappings make it possible to write code that can be successfully built for UNICODE, MBCS or ANSI depending on the setting of the project's character set property. You cannot use UNICODE and MBCS together.

My personal preference is to build for UNICODE because that is Windows internal character, the V6 Common controls library is designed for UNICODE (read Does version 6 of the common controls support ANSI or not?), and many Shell functions and COM interfaces require the use of wide-character strings.

Answer 2

Castorix31 90,686

Use WCHAR (LPTSTR) everywhere :

typedef struct _MYITEM
{
	HFONT hfont;
	LPTSTR psz;
} MYITEM;

CDev-8220 385 Reputation points

2024-04-11T14:41:15.04+00:00

Use WCHAR (LPTSTR) everywhere

I don't understand what you mean.

@Minxin Yu I do not understand what you mean by:

Set MBCS or use wchar

I went here, but there is nothing there explaining how to set MBCS.

What do you mean by use wchar?

Are you saying wherever I see LPSTR change it to WCHAR?

@RLWA32 I have never had to change the character set to either "Not Set" or MBCS. Is there a documentation to help persons new to this?

Answer 3

Minxin Yu 13,506 Microsoft External Staff

Hi, @CDev-8220

The sample uses ANSI. StringCchLengthA and GetTextExtentPoint32A
Set MBCS or use wchar
User's image

Best regards,

Minxin Yu

If the answer is the right solution, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment".

Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.

Share via

Why does Microsoft's code examples have errors, and how do I fix these?

2 additional answers

Your answer