From: Sam S. <sd...@gn...> - 2002-10-17 15:29:37
|
> * In message <1034843459.5467.48.camel@semmel> > * On the subject of "stream-{read/write}-sequence (was Re: stream-element-type not exported from package gray)" > * Sent on 17 Oct 2002 10:30:54 +0200 > * Honorable Rudi Schlatte <ru...@co...> writes: > > would it be possible to add generic > functions stream-read-sequence and stream-write-sequence to clisp? > These are not in Gray's proposal, but are in cmucl, sbcl and acl. CLISP has STREAM-[READ|WRITE]-[BYTE|CHAR]-SEQUENCE, which are called by [READ|WRITE]-[BYTE|CHAR]-SEQUENCE, which are called by [READ|WRITE]-SEQUENCE. I can add STREAM-[READ|WRITE]-SEQUENCE too, if people want it. (the default method will have to figure out whether the user wants bytes or chars and do i/o one by one). > I'm playing with Portable AllegroServe on clisp, where Gray streams are > used to implement chunked transfer encoding support, and write-sequence > is called on these streams quite often. When write-sequence and > read-sequence are called with an argument of type > fundamental-output-stream or fundamental-input-stream, respectively, > they should dispatch the call to their stream-xxx-sequence counterpart. A good (performance-wise) implementation will always end up calling [READ|WRITE]-[BYTE|CHAR]-SEQUENCE on a built-in steam, so STREAM-[READ|WRITE]-[BYTE|CHAR]-SEQUENCE should be quite enough. > I think I can get the rest of (the single-threaded part of) paserve to > run, but it would be nice to have these. cool! -- Sam Steingold (http://www.podval.org/~sds) running RedHat8 GNU/Linux <http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/> <http://www.mideasttruth.com/> <http://www.palestine-central.com/links.html> UNIX is a way of thinking. Windows is a way of not thinking. |