🚀 go-pugleaf

In article <DI8rpv.H5x@research.att.com>,
Andrew Koenig <ark@research.att.com> wrote:
>Distribution:
>In article <48h0vg$8i6@meg.tesser.com> ralphs@tesser.com (Ralph Shnelvar) writes:
>
>> I haven't even peeked at the STL but am wondering if there is support
>> for a disk based sorting and/or indexing algorithms.
>
>Sure.  All you have to do is write a class that meets the
>relevant iterator constraints and grants access to your
>disk file.  Then all the algorithms will work.
>-- 
>				--Andrew Koenig
>				  ark@research.att.com

A cleaner way to do it is to make C++ persistent, and then use normal
"in memory" data structures to do your sorting.  Of course, you need
to pick data structures and algorithms that have good locality.

Mumit Khan has used our Texas persistent store (an efficient, high-tech,
but fairly portable p-store) for Persistent GNU STL.  Unfortunately,
GNU STL's current allocator scheme breaks orthogonal persistence, but if
you want all of your data structures to be persistent anyway, that's OK.

The next release of Texas will contain better support for STL.  (The
current release has trouble with nested classes, and requires a little
hand-fiddling to make it work.  The next release will handle nested
classes semiautomatically---you'll have to provide a list of header
files relevant to nested classes so that we can correctly generate
type descriptor information for them.)

-- 
| Paul R. Wilson, Comp. Sci. Dept., U of Texas @ Austin (wilson@cs.utexas.edu)
| Papers on memory allocators, garbage collection, memory hierarchies,
| persistence and  Scheme interpreters and compilers available via ftp from 
| ftp.cs.utexas.edu, in pub/garbage (or http://www.cs.utexas.edu/users/wilson/)      


---
[ comp.std.c++ is moderated.  Submission address: std-c++@ncar.ucar.edu.
  Contact address: std-c++-request@ncar.ucar.edu.  The moderation policy
  is summarized in http://dogbert.lbl.gov/~matt/std-c++/policy.html. ]