View previous topic :: View next topic |
Author |
Message |
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 7924 Location: Salford, UK
|
Posted: Thu Mar 02, 2017 9:09 am Post subject: |
|
|
The standard Silverfrost INCLUDE files contain interfaces to functions and integer parameters that are in Silverfrost and common third party DLLs. There is no related object code for these files. In that sense that do only provide "symbols". The same is true for the corresponding modules. |
|
Back to top |
|
|
mecej4
Joined: 31 Oct 2006 Posts: 1886
|
Posted: Thu Mar 02, 2017 2:06 pm Post subject: Re: |
|
|
JohnCampbell wrote: | I have been puzzled by this thread.
I would have expected that INCLUDE and USE would create different variable definitions.
John |
Not necessarily.
The INCLUDEd source may contain only comments. It may contain only executable statements and no variable definitions. Or the converse. Included lines may be rejected by the compiler if they lead to syntax errors.
The compiler will judge the included lines only after they have been merged in place in the including source code.
The USE statement, on the other hand, can appear only after a subprogram declaration and before any variable declarations. The module that is being referred to must already exist. That module may provide variables or module procedures or both. |
|
Back to top |
|
|
JohnCampbell
Joined: 16 Feb 2006 Posts: 2554 Location: Sydney
|
Posted: Fri Mar 03, 2017 1:08 am Post subject: |
|
|
While it may "Not necessarily" be the case, I was trying to highlight that it can be that local variables can be defined, which makes INCLUDE and MODULE/USE different in their use.
As clearwin.ins appears to only define Names or Symbols, it may be safe.
But in general, when using a library of routines, a module or include file can not be safely mixed, which is the case of all libraries I have developed.
I am still uncertain about the limitations on using a MODULE in a .dll
John |
|
Back to top |
|
|
|