Re: [Gcl-devel] tan compilation problem

[Camm Maguire]

Hi Vadim!  this should be fixed in cvs now.  Please check it out?
(And the logic of the patch too, if you don't mind.)

Take care,


"Vadim V. Zhytnikov" <address@hidden> writes:

> Camm Maguire ÐÉÛÅÔ:
> > BTW, Vadim -- what is the right thing to do about the path issue?
> > I.e., can you figure out what will break if we just implement the ansi
> > behavior?  And could you describe for me again what that behavior is?
> > 
> > Take care,
> 
> In fact my old report was a bit misleading.  At present I can't
> count _all_ differences between ansi standard and current gcl
> behavior. But some example to get some idea
> 
> GCL:
>    (pathname-directory "/aaa/bbb") -> (:root "aaa")
>    (pathname-directory "aaa/bbb")  -> ("aaa")
> 
> ANSI:
>    (pathname-directory "/aaa/bbb") -> (:absolute "aaa")
>    (pathname-directory "aaa/bbb")  -> (:relative "aaa")
> 
> Different keywords, different results etc.
> 
> But maybe the most important difference which directly affects
> Maxima load() command could be quickly fixed without altering
> the rest of pathname related GCL functions?
> 
> Compare these examples:
> 
> ANSI:
>     (merge-pathnames "aaa/bbb" "/sss/###.mac")  ->  "/sss/aaa/bbb.mac"
>     (merge-pathnames "/aaa/bbb" "/sss/###.mac") ->  "/aaa/bbb.mac"
> 
> GCL:
>     (merge-pathnames "aaa/bbb" "/sss/###.mac")  ->  "aaa/bbb.mac"
>     (merge-pathnames "/aaa/bbb" "/sss/###.mac") ->  "/aaa/bbb.mac"
> 
> Best wishes,
> 
> -- 
>       Vadim V. Zhytnikov
> 
>        <address@hidden>
>       <address@hidden>
>       <address@hidden>
>      <address@hidden>
> 
> 
> 
> 
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah