RE: [Boost-users] Hash Table
Muhammad Hussain <m <mailto:mhussain@i.kyushu-u.ac.jp> <mailto:mhussain@i.kyushu-u.ac.jp> <mailto:mhussain@i.kyushu-u.ac.jp> hussain@i.kyushu-u.ac.jp> wrote: Is there any library in boost that provides hash table and the methods to manipulate it? No, but there are common extensions to the standard containers library (aka STL) that are included with many compilers, called hash_map and hash_set. Unfortunately there are several implementations of these which do not have quite the same interface. Ben.
On Thu, 4 Mar 2004 15:49:02 -0000, Ben Hutchings wrote
Muhammad Hussain <m <mailto:mhussain@i.kyushu-u.ac.jp> <mailto:mhussain@i.kyushu-u.ac.jp> <mailto:mhussain@i.kyushu- u.ac.jp> hussain@i.kyushu-u.ac.jp> wrote:
Is there any library in boost that provides hash table and the methods to manipulate it?
No, but there are common extensions to the standard containers library (aka STL) that are included with many compilers, called hash_map and hash_set. Unfortunately there are several implementations of these which do not have quite the same interface.
There has been some discussion of a tr1 hash table implementation for boost on the developer list. Someone has offered an implementation, but as far as I know it hasn't gone in. You might search the mailing list archive and see if you can download. You could also look at the STLPort hast table. HTH, Jeff
On Thu, 4 Mar 2004, Jeff Garland wrote:
There has been some discussion of a tr1 hash table implementation for boost on the developer list. Someone has offered an implementation, but as far as I know it hasn't gone in. You might search the mailing list archive and see if you can download. You could also look at the STLPort hast table.
This brings up an interesting organizational question: What sort of review procedure should we have for this type of thing? A full review seems a little silly since the interface has already been fixed, so it's only implementation details that would be reviewed in full... perhaps it's another fast-track review candidate. Doug
On Thu, 4 Mar 2004 14:43:11 -0500 (EST), Douglas Paul Gregor wrote
This brings up an interesting organizational question: What sort of review procedure should we have for this type of thing? A full review seems a little silly since the interface has already been fixed, so it's only implementation details that would be reviewed in full... perhaps it's another fast-track review candidate.
Fast track makes sense to me. Although we probably need to move this discussion to the developer list :-) Jeff
participants (3)
-
Ben Hutchings -
Douglas Paul Gregor -
Jeff Garland