Re: HLA v2.x and / or LASM suggestion: Win32 Resources
From: hutch-- (hutch_at_movsd.com)
Date: 01/06/04
- Next message: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Previous message: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- In reply to: Randall Hyde: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Next in thread: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Reply: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 6 Jan 2004 08:18:50 -0800
I wonder about the big deal with resources in 32 bit Windows. The old
16 bit Windows tacked on resources but when Microsoft got the VAX guys
to develop the Windows version of portable executable where resources
are a section in the PE file.
Microsoft languages convert the RES file to an OBJ file and then link
it into the final executable, the toys have to fudge it to get their
resources in. Working with code in modular components is part of the
OS file format design and the most efficient way to do this is with
libraries and a linker that puts all of the components together to
make the executable code.
Any of the big professional tools can do this, C compilers, large
assemblers and the like, the only reason why the toys cannot do it is
that it is too difficult technically to cobble together. With tools
designed at a professional level where compatibility is determined by
OS file specifications, you can swap around things like compilers and
assembler and linkers and resource compilers, with the toys you are
stuck with the way they are cobbled together and the scale of the
output shows it.
Monolithic files with no external file or library support are among
the worst possible code designs. Finding things are a problem and
laying out code in an efficient manner is almost impossible. Without a
linker to resolve interdependency between procedures, you can only add
code on a trial and error basis until it works, with a library and a
linker, you just call the procedure you need and the linker resolves
the dependencies.
Often the model people use for assembler code is tiny little test
programs and while it may feel profound to hack these types of
programs together, the style of architecture is not extendable to
larger applications because of its lack of modularity. Profundities
about tiny programs with lousy monolithic architecture fall flat when
anthing useful has to be written of any size.
Regards,
hutch at movsd dot com
- Next message: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Previous message: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- In reply to: Randall Hyde: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Next in thread: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Reply: Betov: "Re: HLA v2.x and / or LASM suggestion: Win32 Resources"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|