Re: CGA standard
Alastair Reid <[email protected]>
| Newsgroups | gmane.comp.lang.haskell.gui |
|---|---|
| Message-ID | <[email protected]> |
Krasimir Angelov <[email protected]> writes: > I am under the impression that there is a consensus of all points > mentioned [below] but they have to be specified in the standard. My impression differs on 3 of the 4 points. > Summary: > - CGA should use callback functions for event handling The concensus does seem to be that this is readily implemented and can be used to implement the other styles of interface being proposed. > - All primitives should be thread safe There seems to be some disagreement over this and strong arguments for considering requiring single-threaded use. (There are also strong arguments for requiring thread-safety.) > - Calling whoever of the primitives should not block the concurrent > threads. An exception can be done only for the primitives who are > executed in a very short constant time This decision is contingent on a multithreaded design. > - For each event there should be two functions: register and unregister My understanding is that there should be one function 'register' which returns an 'unregister' action as part of its result. This neatly avoids the problem of needing equality tests on functions (semantically difficult) or hacks to get around this lack. -- Alastair Reid [email protected] Reid Consulting (UK) Limited http://www.reid-consulting-uk.ltd.uk/alastair/