NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/alt/hackers/1155

Internet Message Format | 1992-07-23 | 1.6 KB

Path: sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!agate!ocf.berkeley.edu!alanc
From: [e-mail] (Alan Coopersmith)
Newsgroups: alt.hackers
Subject: Re: not a test
Date: 24 Jul 1992 05:01:01 GMT
Organization: U.C. Bezerkeley Overloaded Computing Facility
Lines: 22
Approved: Apollos 4ever!
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: ocf.berkeley.edu

In article <[e-mail]> [e-mail] (Peter da Silva) writes:
)In article <[e-mail]> [e-mail] (Alan J Rosenthal) writes:
)>good god, please just use strtok().  Get to know your friendly neighbourhood C
)>library.  This is EXACTLY what strtok() is for.
)
)Good god, please don't use strtok. It's inherently non-reentrant and it's not
)even usable in cooperative threads. Define a new function that operates on a
)string handle or a string file, but let strtok die a well deserved death.

From the strings.h man page of Apollo Domain O/S 10.4: (which has a threads
interface based on Mach's)
     The strtok_r function is the reentrant version of strtok.  Upon
     successful completion, the first character of the next token is stored
     in **last_string, and a value of 0 (zero) is returned.

This from the dying O/S that is supplying many of its features to OSF/1...


--
------------------------------------------------------------------------
Alan Coopersmith            Internet: [e-mail]
University of California, Berkeley    Bitnet:   [e-mail]