Major architectural flaw in FASM?

Help needed,

FASM cannot handle more than 1 error at a time. So, if your source program has say 45 errors, you'll have to correct and recompile your code 45 more times in order to finally execute it.

This may be because it doesn't support lookaheads (but surprisingly forward references are easily resolved), and also may be using a multiple top-down parser approach. If anybody figures out a way to solve this problem, please get in touch with Privalov immediately.

Posted on 2003-11-29 15:24:19 by art_sands
I would imagine multiple passes would be the answer to this issue - creating a database of symbolic references to the elements of the program would mean you are effectively looking ahead all the way to the end of the program. The idea would be to first resolve all references, creating temporary symbols for unresolved refs, in much the same way as a multipass disassembler operates, before attempting to issue opcodes in line with the source.
Posted on 2003-11-29 15:46:32 by Homer