Incorrect use of mbstowcs?

[Ralph A. Moritz]

Hi all,

my knowledge of character encoding issues is pretty limited and I
simply
cannot find a problem with the following code. It compiles -- but
seems to
produce garbage at runtime. I'd appreciate any helpful suggestions.

wchar_t* itemNameW = new wchar_t[strlen(itemName + 1) *
sizeof(wchar_t)];
mbstowcs(itemNameW, itemName, strlen(itemName) +1);

TIA,
Ralph

[Alf P. Steinbach]

* Ralph A. Moritz:

Hi all,

my knowledge of character encoding issues is pretty limited and I
simply
cannot find a problem with the following code. It compiles -- but
seems to
produce garbage at runtime. I'd appreciate any helpful suggestions.

wchar_t* itemNameW = new wchar_t[strlen(itemName + 1) *
sizeof(wchar_t)];

itemName+1 is a pointer to the second character of itemName. Instead of
the length + 1 you therefore get the length - 1.

mbstowcs(itemNameW, itemName, strlen(itemName) +1);

You're not copying the terminating null-character.

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Alf P. Steinbach]

* Ralph A. Moritz:

Hi all,

my knowledge of character encoding issues is pretty limited and I
simply
cannot find a problem with the following code. It compiles -- but
seems to
produce garbage at runtime. I'd appreciate any helpful suggestions.

wchar_t* itemNameW = new wchar_t[strlen(itemName + 1) *
sizeof(wchar_t)];

Oh sorry, I didn't see that there was even more, due to the
line-breaking: don't multiply by sizeof(wchar_t).

mbstowcs(itemNameW, itemName, strlen(itemName) +1);

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Alf P. Steinbach]

* Ralph A. Moritz:

Hi all,

my knowledge of character encoding issues is pretty limited and I
simply
cannot find a problem with the following code. It compiles -- but
seems to
produce garbage at runtime. I'd appreciate any helpful suggestions.

wchar_t* itemNameW = new wchar_t[strlen(itemName + 1) *
sizeof(wchar_t)];
mbstowcs(itemNameW, itemName, strlen(itemName) +1);

Even more sorry, forgot to give the standard advice (which you should
heed): use std::wstring, forget about that manual memory management.
--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Ron Natalie]

Ralph A. Moritz wrote:

Hi all,

my knowledge of character encoding issues is pretty limited and I
simply
cannot find a problem with the following code. It compiles -- but
seems to
produce garbage at runtime. I'd appreciate any helpful suggestions.

wchar_t* itemNameW = new wchar_t[strlen(itemName + 1) *
sizeof(wchar_t)];

new already handles the size of the object. There's no need to
multiply by wchar_t. Strlen is already going to give you a pessimistic
view of the number of wchar's required.

mbstowcs(itemNameW, itemName, strlen(itemName) +1);

So what's in itemName?

What did you do to examine itemNameW to determine it's validity?