Hello developers.

I'm trying to gain some understanding of fpc internals, in view of possible contributions, and I stumbled in something I don't understand.

The tmodeswitch type is, as far as I can grasp, an enumerated type used to enable or disable features according compilation switches, such as delphi objfpc, etc.

Some tokens are conditioned by compilation switches, as reflected by the appropriate field in arraytokeninfo while some others are not, and this I understand: some constructs are mac specific, others are delphi specific, etc.

But I fail to grasp the difference between *m_none* and *m_all*.

Could someone shed some light on this subject?

Thank in advance,

Giuliano

Giuliano Colla wrote

But I fail to grasp the difference between *m_none* and *m_all*.

These are used in tokens.pas. Tokens marked as m_all are treated as a keyword in all syntax modes (i.e., you can never use them as identifier without escaping them). Tokens marked as m_none are never treated as a keyword.

Jonas

Jonas Maebe ha scritto:

Thank you, now I understand.

Giuliano