Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!usc!cs.utexas.edu!uunet!willett!ForthNet
From: ForthNet@willett.UUCP (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: Non-Forth systems/languages.
Message-ID: <524.UUL1.3#5129@willett.UUCP>
Date: 22 Feb 90 01:57:04 GMT
Organization: Latest link in the ForthNet chain.  (Pgh, PA)
Lines: 69


 Date: 02-20-90 (21:10)              Number: 2936 (Echo)
   To: L.ZETTEL                      Refer#: 2926
 From: ZAFAR ESSAK                     Read: NO
 Subj: NON-FORTH SYSTEMS/LANGUAG     Status: PUBLIC MESSAGE


   To: Len Zettel (GEnie Category 3, topic 16, msg ??) 
   Re: Streams. 

 Thankyou for your response to my earlier message. 

 LZ> streams tend to impel you to..."Begin at the beginning, go 
 LZ> through to the end, then stop". 
 LZ> otherwise, depending on the capabilities of your operating 
 LZ> system, you may have troubles. 
 LZ> Certainly, backing up in a pure stream is usually less 
 LZ> convenient than putting the water back in the hose. 

 LZ> Long streams of stuff needing wholesale programmable 
 LZ> modification just don't happen to me that often. 

 In addition to the usefulness of streams when modifying long strings 
 of stuff streams can also be very useful in the implementation of a 
 database, where random access is the primary method of access. 

 The implementation used of Streams in Forth was developed by Dan 
 Phillips and fully documented with source in a proposal to the ANSI 
 Committee and uploaded to the Forth Net.  Maybe the benefits are the 
 result of the implementation used, but I don't think that's all there 
 is to it. 

 The ability for "backing up" or resetting the stream pointer is a 
 necessity and the syntax is as readable as BLOCK I/O. 

     e.g.  d1 ( old place ) <from> >dseek 
           d1 ( old place ) 1024 UM/MOD BLOCK + 

 There may be some performance penalty for the stream as compared to 
 the block approach depending on the implementation. 

 Another advantage Streams offer is not having to remember the 
 "boundaries" of blocks when data extends over from one to another. 
 Certainly not a trivial requirement when data is variable length. 

 Also the ability to switch streams at any time is required and can be 
 provided by having "Stream Stacks" or using some other method of 
 explicitly identifying the stream. 

 What I find most enticing about Streams is the versatility of a few 
 simple words such as IN ( --c) OUT ( c--) INPUT ( adr,n--) and 
 OUTPUT ( adr,n--). 

 LZ> As a long time admirer of the beauty in Kernighan and Plauger's 
 LZ> Software Tools, I have now and again contemplated implementing 
 LZ> them in Forth. 

 Having not read this I guess it is going to require a trip to the 
 library or bookstore.  In the meantime, can you provide a brief 
 summary of what you are refering to. 

 Thanks, Zafar. 
 ---
  * Via Qwikmail 2.01

 NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886   
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'