| To: | David Korn <yyy@xxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: AI 2000-05-010: proposed interface |
| From: | Ulrich Drepper <yyyyyyy@xxxxxxxxxx> |
| Date: | 14 Aug 2000 14:18:49 -0700 |
| Cc: | yyyyyyyyyyyy@xxxxxxxxxxxxx |
| References: | <200008142108.RAA37605@raptor.research.att.com> |
David Korn <yyy@xxxxxxxxxxxxxxxx> writes: > I believe that it would be a bit more complicated to use with strseq() > However, I believe that it would require two calls to this > function to test if a character is in the range, not 1 as with colseq(). I would never implement the range expression matching with any of these interfaces. When implementing it in the libc I have access to the internal data structures. This is obviously how everybody else did it since there is no interface like this so far. The additional interface is meant to provide the user a possibility to learn about the collation sequence. This can be done either way as well, I just think that one normally will find some kind of input text which then is sorted and examined. Anyway, if we cannot agree here there is nothing easier than writing down a second proposal. -- ---------------. ,-. 1325 Chesapeake Terrace Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA Red Hat `--' drepper at redhat.com `------------------------ |
| Previous by Date: | RE: AI 2000-05-010: proposed interface, Jason Zions |
|---|---|
| Next by Date: | RE: Re: AI 2000-05-010: proposed interface, Donn Terry |
| Previous by Thread: | Re: Re: AI 2000-05-010: proposed interface, David Korn |
| Next by Thread: | RE: AI 2000-05-010: proposed interface, Jason Zions |
| Indexes: | [Date] [Thread] [All Lists] |