the discussion so far
Matthew Flatt
(16 Jul 2005 12:41 UTC)
|
||
(missing)
|
||
(missing)
|
||
Re: the discussion so far
Alex Shinn
(20 Jul 2005 02:50 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(20 Jul 2005 02:56 UTC)
|
||
Re: the discussion so far
Alex Shinn
(20 Jul 2005 03:15 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(20 Jul 2005 03:24 UTC)
|
||
Re: the discussion so far
Alex Shinn
(20 Jul 2005 03:38 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(20 Jul 2005 03:49 UTC)
|
||
Re: the discussion so far
John.Cowan
(20 Jul 2005 04:24 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(20 Jul 2005 04:27 UTC)
|
||
Re: the discussion so far
John.Cowan
(20 Jul 2005 04:58 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(20 Jul 2005 05:04 UTC)
|
||
(missing)
|
||
(missing)
|
||
Re: the discussion so far
bear
(20 Jul 2005 02:45 UTC)
|
||
Re: the discussion so far
John.Cowan
(20 Jul 2005 03:56 UTC)
|
||
Re: the discussion so far
Jorgen Schaefer
(16 Jul 2005 13:05 UTC)
|
||
Re: the discussion so far
Matthew Flatt
(16 Jul 2005 13:21 UTC)
|
||
Re: the discussion so far
Jorgen Schaefer
(16 Jul 2005 13:58 UTC)
|
||
Re: the discussion so far Thomas Bushnell BSG (17 Jul 2005 02:42 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(17 Jul 2005 02:57 UTC)
|
||
Re: the discussion so far
Jorgen Schaefer
(17 Jul 2005 03:33 UTC)
|
||
Re: the discussion so far
bear
(16 Jul 2005 18:07 UTC)
|
||
Re: the discussion so far
John.Cowan
(17 Jul 2005 04:49 UTC)
|
||
Re: the discussion so far
Thomas Bushnell BSG
(17 Jul 2005 02:40 UTC)
|
Jorgen Schaefer <xxxxxx@forcix.cx> writes: > String collation is very complex, as the "preferred" order of > characters depends on the locale. But since STRING<? and friends > are often used for things like binary search trees where the exact > order is irrelevant and the only important thing is the existance > of any kind of total order, defining them the way this SRFI does - > by using the codepoint sequence - is good, because it is fast. If > the implementation wants to provide the locale-dependent string > collation, fine, but that's not useful for this SRFI to define. This would make sense *only* if users would know that string<? might give them wrong results on fancy systems if they use it for indexing. So how about specifying two functions, one that implements a total order for use where you don't care what the order is, and another which guarantees the human-sensible text sorting method. Simple systems can simply eq the procedures; fancy systems can make fancy differences. Programmers will be on alert, and can use the correct name for whichever they are using.