FLTK logo

Re: [fltk.coredev] On the building of fluid-cmd and fluid...

FLTK matrix user chat room
(using Element browser app)   FLTK gitter user chat room   GitHub FLTK Project   FLTK News RSS Feed  
  FLTK Apps      FLTK Library      Forums      Links     Login 
 All Forums  |  Back to fltk.coredev  ]
 
Previous Message ]New Message | Reply ]Next Message ]

Re: On the building of fluid-cmd and fluid... imacarthur Apr 25, 2023  
 
On Tuesday, 25 April 2023 at 15:00:30 UTC+1 Matt wrote:

I am confused. So under mingw, "fluid-cmd.exe" is a Windows subsystem app, but it still provides stdout and stderr output? I thought that the whole subsystem trouble is at run time and caused by the OS rather than the build system or shell?

Not quite: fluid-cmd.exe is built as a console app, under both MS and mingw at present.
My "objection" to that was only that we compile everything twice, even though the actual compiled .obj files are no different. 
The difference between "console" and "windows" apps is at the final link, when the PE header is constructed.
We could presumably compile everything just the once, then have two linker passes for the fluid.exe and fluid-cmd.exe.

The difference between a MS "console" app and a "windows" app, in effect, can be a little as a single bit in the PE header.
If the "console" bit is NOT set, the MS-DOS shell (and it's descendants) drops the stdio and detaches the app from the shell.
If the "console" bit is set, the stdio stays attached, and the shell waits for the process to finish.

If you run the app from the Msys shell, as many mingw users will, this does not happen; the Msys shell pretty much ignores the difference between "console" and "windows" apps, at least in this regard. It's more like bash in this regard.
Note that it does this whether the app was built with mingw or an other toolchain, e.g. MSVC.

The "fluid-cmd" console app was provided, as far as I know, so that MS-DOS shell users could still get stdio and so forth, but for Msys shell users it was kinda moot, since it "worked" anyway.

I'm not proposing to do away with fluid-cmd, just proposing we drop the redundant extra compilations to build it.
 

Also, if we stop providing fluid-cmd on mingw machines, existing build setups may still rely on its existence, right? 

Yeah, I think we'd still need to provide something called fluid-cmd - just in case. It is used in the fltk lib Makefiles now I think, under Windows.
It wouldn't affect my builds, I never use it, but I'm sure there are folks who do. 
I'd assume Greg uses it in his Windows builds, for example?

 

--
You received this message because you are subscribed to the Google Groups "fltk.coredev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fltkcoredev+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fltkcoredev/1776f20a-c625-4b91-898a-54fc1c546f5cn%40googlegroups.com.
Direct Link to Message ]
 
     
Previous Message ]New Message | Reply ]Next Message ]
 
 

Comments are owned by the poster. All other content is copyright 1998-2024 by Bill Spitzak and others. This project is hosted by The FLTK Team. Please report site problems to 'erco@seriss.com'.