Re: [boost] Re: persistence : current state, and plans for future ?

From: Beman Dawes (bdawes_at_[hidden])
Date: 2001-04-17 18:22:02

At 06:15 PM 4/17/2001, krempp_at_[hidden] wrote:

 >Hum..

 >I checked out branch "persistence-initial" but I am not sure this is
 >actually what you are speaking about,
 >since the persistence_demo.cpp still throws (probleme of EOF and
 >file.exceptions set in definiton of save_file )
 >
 >PLUS there is an obvious bug here :
 >
 > ...

Yes.  But it is Jens' code, so I'll let him make the fixes.  Because of day 
job pressures he isn't always able to response instantly, but I expect he 
will respond in a reasonable amount of time.

 >about the XDR : with my g++ (3.0-pre) I had to define
 >void write(std::size_t n)
 >void write(int n)
 >[ and same with laod)
 >because it was called with those type, but they did not match any of
 >the boost *least32 types.
 >Is it normal ?

Yes, that's a known issue.  My reaction is that it would be safer to 
overload for each built in type.

A related question:  For XDR loads, the external format is always 
32-bits.  So what should happen when the internal target won't hold the 
actual value read?  Or should only *least32 types be allows, so the issue 
never arises.  I have no idea if Jens has thought about that issue, but he 
may have thought of it and that is why he didn't provide the overloads.  If 
so the fix will be different.

 >( I dont really care about XDR, but this prevented me from compiling
 >test1, so I felt the need to make it run..)

Jens would appreciate it I'm sure if you report problems you run into, and 
any fixes you come up with.  Known problems are 1) wchar_t fails on many 
compilers because of poor wchar_t support, 2) floating point types have 
problems with text readers and writers (but not binary readers and 
writers), and 3) the Microsoft compiler can't handle the save/load 
overloads on containers.

Cheers,

--Beman