RE: RosAsm

From: wolfgang kern (nowhere_at_nevernet.at)
Date: 04/29/04 

Date: Thu, 29 Apr 2004 14:02:36 +0200

Sorry, I had to start a new thread,
the reply-chain is too long already,
and my provider can't handle it.
-----------------------------------
Betov wrote:

[..]
| 1) The only LOGFONT we have actually in the [Struct] Dialog is for:
|
| [LOGFONTA:
   ....
| [lfFaceName: B$ 0 #&LF_FACESIZE]

Now I see, I may have shortend the facename-buffer by using
  [lfFaceName: B$ 'Courier New' 0]
I'll extend with zeros to get 32 bytes right now.

May I rely on data align filler-bytes are Zeros?

| ... So, i suppose there is also a 'LOGFONTW' -as usual- (Our
| Structures File is actually far from complete. Guga is actually
| working hard at producing an real exensive and complete version...).

Not a real problem, I can paste from win32.hlp and modify it.

| If true, the Unicode version should, of course be used with the
| unicode version of all concerned api:
|
| call 'GDI32.CreateFontIndirectW' ; W and not A
|
| In your case this is the 'A' version. So the proper way is to
| set your Name with 'B$'. You cannot mix Unicode formulations
| with Ascii formulations.

Ok, I can live with it.

| 2) My brain was not pretty clear when i wrote my previous post:

| In RosAsm syntax, the Data sizes have two defaults that save
| you from a bit of typing:

[...] Got that.

| My sentence saying that i would be estonished if W$ 'String'
| and U$ 'String' produced different output was completely
| stupid. They of course do: In, W$ 'String', the 'W$' is
| ignored because of the default for Bytes Strings and this
| is exactely why i implement 'U$' to force them to Words'Strings...

| Sorry for the added confusion!... ;)

No problem, I'm already familiar with confusion :)
So I repeat: U$ 0fffc is NOT -4 "U"nsigned
             U$ 'string' means "U"nicode
             B$/W$ 'string' are Ascii

BTW: the right-click on empty lines shouldn't change the view
     and the 'tree/<<</>>>' functions behave not as I'd expect.
     Just my opinion ...
     How can I produce 83h ..code [ie:83 c1 80 ADD ecx,80h]
     (the immediate sign-extended byte group)
     I cannot use DB for memory, did I miss something in B_U_ASM ?

Thanks Rene,
__
wolfgang

Relevant Pages