> also, it's not clear what happens if transaction is rolled back -- is
> prepared statement rolled back too, or it is memorized?
> it seems this way our idea of memorized statements and db session can be
> desynchronized, but actually in _oposite direction_: our code will think
> that it's there,
IE> driver for it. Do you have a reason, other than good code hygiene,
IE> for having such an abstraction?
i had some problems with this memutil thing recently -- I was trying 32-bit
SBCL on 64-bit machine.
everything was going fine but memutil -- it was compiled with 64-bit gcc and
SBCL coul
On Apr 29, 2008, at 4:06 PM, Leslie P. Polzer wrote:
This is a legacy component that plays very nicely with BDB. It
should
be possible to abstract over buffer-streams so you could put a
different serializer in place for, say, a lisp-only backend. I
haven't been highly motivated to do this
> This is a legacy component that plays very nicely with BDB. It should
> be possible to abstract over buffer-streams so you could put a
> different serializer in place for, say, a lisp-only backend. I
> haven't been highly motivated to do this yet given there hasn't been a
> driver for it. Do
This is a legacy component that plays very nicely with BDB. It should
be possible to abstract over buffer-streams so you could put a
different serializer in place for, say, a lisp-only backend. I
haven't been highly motivated to do this yet given there hasn't been a
driver for it. Do you
> how do you thik it is supposed to work?
Yeah, I wonder. :)
> "The remaining problem outlined in the section on Serialization is that
> operations which mutate collection types do not have persistent side
> effects."
What I have mainly carried with me from this section of the
documentation wa
Alex is exactly right, as pointed out by the docs. set-valued slots
partially solve this annoyance with persistent collections stored in
slots, but they are unordered.
Unfortunately aref is not a generic function so you'd have to play
package naming games to hijack it which I abhor. I'm o
LPP> (32): (setf (aref (slot-value sch 'a) 2) 1)
LPP> 1
LPP> (33): (slot-value sch 'a)
LPP> #(0 0 0 0)
how do you thik it is supposed to work?
to my knowledge, elephant does not hijack (setf aref), so (setf aref) just
modifies a temporary array returned by a slot read.
db is written _only_ o
The buffer stream type provided by the memutil module is an essential
part of the serializer. Is this really necessary, or could we use
a more portable way?
Leslie
___
elephant-devel site list
elephant-devel@common-lisp.net
http://common-lisp.net/mai
There are array tests and slot tests, but there don't seem to be any
covering the combination of those. Seems to have turned out to be
a problem:
(29): (defpclass schmuck () ((a :type array :initform (make-array 4)))(:index
t))
#
(30): (setf sch (make-instance 'schmuck))
#
(31): (slot-value s
10 matches
Mail list logo
