On Tue, Oct 13, 2009 at 7:27 PM, Robert Bradshaw <rober...@math.washington.edu> wrote: > > On Oct 13, 2009, at 6:45 PM, William Stein wrote: > >> On Tue, Oct 13, 2009 at 6:38 PM, Robert Bradshaw >> <rober...@math.washington.edu> wrote: >>>> This thread is mainly about Gonzalo's proposal that we target >>>> something like busybox (or my suggestion "python") instead of POSIX >>>> standard shell usage. Somehow it is amazingly difficult to keep >>>> this >>>> discussion on track! >>>> >>>> Definitely your 1-3 are definitely a good idea though... it's hard >>>> to >>>> argue with them. >>> >>> Ah, I got my threads crossed. I't still a bit unclear, are we talking >>> about the scripts in sage/local/bin, or install scripts for the >>> various packages, or both? >> >> I think we are talking about both. >> >>> I like the idea of moving towards using Python as the scripting/build >>> coordinating language, but that might make using non-standard >>> compliers and/or cross compiling even more difficult. >> >> Why? > > I don't know how much we can count on distutils/distribute/scons/? to > support non-gnu toolchains. On the other hand, if we're just using > pure Python, then I agree that it would lend itself to a much cleaner > build system (though then we risk re-inventing the wheel...) > > I don't think requiring (some) Python, and using that to bootstrap the > build process before building our own Python, is to onerous, though it > would be funny if the "source Python distribution" had Python as a > prerequisite.
Even funnier is that Sage *does* have perl as a prerequisite, since the PARI build system is written in Perl. Nobody mentions or worries about this, since perl is ubiquitous -- certainly much more common than "make" or gcc. Before Sphinx, Python's documentation build system had perl as a prerequisite too. I would be OK to require "some random python be installed systemwide" as a prerequisite to build Sage. For windows we require Python... but if you don't want to install Python we do have a little "bootstrap" dos script that build's Sage's python. -- William > >> I can only imagine it making things easier since we can structure >> are code more cleanly and factor out common things. Anyways, one can >> always do "os.system(...)" so shell capabilities are a subset of >> Python. >> >> This isn't a purely theoretical discussion, since the native Windows >> porting work of Sage has a Python-based build system already. >> >> I'm not arguing for changing anything in the Sage build system right >> now. I'm just suggesting we should keep an open mind, so when some >> incredible person shows up at a SAge days (say) who is willing to dive >> in and do some amazing 1-month coding sprint to improve the build >> system substantially -- say for a major Sage release like 5.0 -- we >> are ready. > > +1 > > - Robert > > > > > > -- William Stein Associate Professor of Mathematics University of Washington http://wstein.org --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
