[Ecls-list] open :supersede and rename-file advise

[Juan Jose Garcia-Ripoll]

On Nov 29, 2007 9:47 PM, Geo Carncross <geocar at gmail.com> wrote:
 > See my response to (2), but more on that: I'm not sure many users
> fork() with lisp, with files open for :supersede, and have written,
> and use the stream on each side differently. I don't know if it's
> necessary to optimize for this.

This is how I see it:

i) The truncation semantics is not going to disappear. That means
users who expect to be able to read and write a file from two
processes which open with :supersede will continue to do so for the
time being. Furthermore, we will add :truncate as a synonym for
:supersede's current behavior.

ii) We add a new flag which is not standard. Users willing to use it
should be aware of the consequences: the behavior of :lazy-supersede
or :atomic-whatever, as of now will be specified for single processes.
Anything related to safety and fool-proof implementation can be
extended later on as we see how people use this feature and we learn
more about the implementation.

iii) Related to ii), we will watch the reaction of other
implementations and users towards this idea of resurrecting
:supersede's original behavior. Further improvements should be decided
on a community basis and, if possible, trying to reach consensus with
other implementors.

Regards,

Juanjo
-- 
Facultad de Fisicas, Universidad Complutense,
Ciudad Universitaria s/n Madrid 28040 (Spain)
http://juanjose.garciaripoll.googlepages.com