On Oct 22, 2009, at 4:45 PM, Dr. David Kirkby wrote:

>
> Peter Jeremy wrote:
>> On 2009-Oct-16 13:05:02 +0100, "Dr. David Kirkby" <david.kir...@onetel.net 
>> > wrote:
>>> I'm updating a configure script which will warn users if their  
>>> operating system
>>> is too old. In the case of a Solaris 9 machine it will say:
>>
>> Whilst I agree in principal, I have some concerns about the wording.
>>
>>> configure: WARNING: You are using Solaris 9, which is a very old  
>>> version
>>> configure: WARNING: last released in 2004. Solaris 10 was first  
>>> released in
>>> configure: WARNING: 2005, and there have been many updates since  
>>> then. You
>>> configure: WARNING: are advised to upgrade to Solaris 10 or later,
>>
>> This is all very nice in principal but may not be possible in  
>> practice
>> due to requirements for other applications.  Eg, we still have a  
>> Solaris
>> 2.6 server to support an old application.  Whilst the application  
>> might
>> work with Solaris 10, that would require effort to upgrade the OS and
>> then regression test the application - for which there is no funding.
>
> I had changed the wording somewhat anyway, but I would admit I had not
> considered other applications.
>
> I also added a check for the old sun4m and sun4c systems, as they  
> can not be
> upgraded to Solaris 10.
>
> I'm well there may be reasons for sticking to an older OS, though  
> Solaris has
> excellent binary compatibility. Obviously there are exceptions, but  
> generally it
> it runs on version N, it will run on N+1, N+2 etc.
>
>> I suspect you will find that Solaris 9 is still in fairly common use.
>> Whether it's used widely enough to justify a Sage release, I'm not
>> certain.  (And I'm not certain how much effort would be needed to
>> build Sage on Solaris 9).
>
> I would agree Solaris 9 is in quite common use. If Wikipedia is to  
> be trusted,
> Solaris 9 will be supported by Sun until 2014, and even Solaris 8 to  
> 2012. So
> neither should be written off.
>
> The current wording I have is:
>
> if test x$build_os = 'xsolaris5.0    || test x$build_os =  
> 'xsolaris5.1' \
> || test x$build_os = 'xsolaris5.2'   || test x$build_os =  
> 'xsolaris5.3' \
> || test x$build_os = 'xsolaris5.4'   || test x$build_os =  
> 'xsolaris5.5' \
> || test x$build_os = 'xsolaris5.5.1' || test x$build_os =  
> 'xsolaris5.6' \
> || test x$build_os = 'xsolaris5.7'   || test x$build_os =  
> 'xsolaris5.8' \
> || test x$build_os = 'xsolaris5.9'

Is there not a way to say \le solaris6.x? Listing all blacklisted  
systems explicitly seems both brittle and cumbersome.

>>> I think I'll have to leave any tests for linux (or let someone else
>>> write them), as I believe there are too many different distributions
>>> to worry about.
>>
>> It would be nice to have a list of base OS shared libraries that Sage
>> expects and do a check against the available libraries.  I'm not sure
>> how much effort this would be.
>
> There are many things that are checked in configure scripts, but I  
> do not know
> what are actually needed. Does Sage need 'yacc' or 'bison' for  
> example? If I
> knew it did, then I could add a test for that early on, to save  
> someone the
> trouble of spending hours building Sage, only to find they need  
> something which
> we could have checked for earlier.
>
> I would imagine producing a list of what libraries and programs are  
> needed for
> every Sage package would be a very daunting task. If each developer  
> agreed to
> document exactly what one package needed, then it would be possible.  
> For for any
> single person to do so for 100+ packages would be too much to ask. I  
> certainly
> would not wish to do that.

You mean like the list of dependancies at 
http://www.sagemath.org/doc/installation/source.html 
  or http://www.sagemath.org/mirror/src/README.txt ? One of the goals  
of Sage is to have very few dependancies. Or are those out of date now?

- Robert


--~--~---------~--~----~------------~-------~--~----~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to