; FILE data label will have size of bytes included from file
calminstruction(NAMED) file? name*
        local tmp,bytes
        arrange tmp, =label NAMED:NAMED.=bytes
        assemble tmp
        arrange tmp, =file name
        assemble tmp
        compute bytes, $ - NAMED
        arrange tmp, NAMED.=bytes
        publish tmp, bytes
end calminstruction    

model file 'boat.obj'
repeat 1, V:sizeof model
        display `V
end repeat    

For an unknown reason labels using the FILE data type resolve to a SIZEOF 1.

calminstruction(NAMED) file? name*
        local tmp
        compute tmp, $
        arrange name, =file name
        assemble name
        compute tmp, $ - tmp
        arrange name, =label NAMED:tmp =at =$ - tmp
        assemble name
end calminstruction    

The reason is the usual one - backwards compatibility with fasm 1.

calminstruction(NAMED) file? name*
        local tmp
        compute tmp, $
        arrange name, =file name
        assemble name
        arrange name, =label NAMED:1 =at tmp
        assemble name
end calminstruction     

Your reflection on fasmg—Flat Assembler G—touches on an intriguing aspect of software development, particularly when evolving a tool like an assembler. The backward leaning perspective you mention seems to refer to the compatibility and continuation of features from the original Flat Assembler (FASM) to fasmg. This aspect is crucial because it ensures that existing users and their projects remain supported and operational, offering a form of continuity and stability.

On the other hand, the forward leaning perspective highlights fasmg’s capacity to transcend its roots. fasmg isn’t just about maintaining what was already established; it’s about innovation and expansion. By being more flexible and featuring a macroinstruction language that allows it to assemble for virtually any architecture with enough description, fasmg breaks free from the constraints of its predecessors. This dual approach not only honors the foundation laid by FASM but also pushes the boundaries of what an assembler can do, opening up possibilities for new types of projects and advancements in assembly language programming.

This duality can be seen as a powerful approach in software evolution, balancing between honoring the past and embracing the future, thereby creating a robust platform for both current users and those who wish to push the envelope.