Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!bionet!ames!dftsrv!nssdcs!setzer
From: setzer@nssdcs.gsfc.nasa.gov (William Setzer (IDM))
Newsgroups: comp.lang.c
Subject: Re: Declaration Question
Message-ID: <397@dftsrv.gsfc.nasa.gov>
Date: 25 Jul 89 22:34:43 GMT
References: <394@dftsrv.gsfc.nasa.gov> <1242@atanasoff.cs.iastate.edu>
Sender: news@dftsrv.gsfc.nasa.gov
Reply-To: setzer@nssdcs.UUCP (William Setzer (IDM))
Distribution: usa
Organization: NSSDC Greenbelt Md.
Lines: 15

In article <1242@atanasoff.cs.iastate.edu> hascall@atanasoff.cs.iastate.edu.UUCP (John Hascall) writes:
>In article <394> setzer@nssdcs (William Setzer (IDM)) writes:
>> [Stuff I said]
>   "char *foo[K][N]" is "array K of array N of pointer to char" which doesn't
>   match your picture, how about "char (*foo[K])[N]" which is "array K of
>   pointer to array N of char" which seems to match your picture
I thought that things adjacent to a name had priority over things farther
away, ie, '*foo[][]' = '(*(foo[]))[]'.  Oh well, I guess I'd better check
again.
>   and you will need to malloc the K sets of "N chars" yourself
Then that means that if you say 'char (*foo[K])[N]', then at best the
'[N]' part is lint fodder (no offense to lint intended).  What's the point?
--
William Setzer
setzer@epsl.umd.edu    ; My 'real' mail address