Ok, I think I've found the problem. Perhaps this should point to the need for making clear and complete instructions about creating your own copy.
I found that in my local copy there was a sage script that pointed to the systemwide sage. When I changed that I now get my local copy. So perhaps the instructions should say: Copy the whole directory tree. Inside the top level directory there is a script called sage. Edit that so that SAGE_ROOT points to your local copy, and use that version of sage to run sage (e.g. by making a symbolic link to it in one of your local directories in search order ahead of the system wide copy). So, another question -- it seems that sage -ba didn't regerenate the individual doc files, since if I do something like EllipticCurve?? the path that it displays is in the systemwide directories. Should sage -ba do that? How do I regenerate these files? Victor On Aug 5, 12:50 pm, VictorMiller <victorsmil...@gmail.com> wrote: > First, when I run > > mysage -sh > > and look at the environment variables involving the string SAGE, I > find they the all point to my local copy. There is a SAGE_SERVER > variable > which has the valuehttp://www.sagemath.org, but that shouldn't do > anything since the computer that I'm on is not connected to the > internet. > > as I said before, if I just start mysage, and on the command line I > type > > sys.path > > all of the values there point to my local copy. But when I run > notebook, start a new worksheet, and type sys.path > > all of the values point to the systemwide version. This also happens > if I run > > mysage -notebook foo > > with my current directory in /tmp > > At that point I'm asked to make a new password for admin, which I do. > I then see a notebook server with no worksheets defined. I start up a > new worksheet, and type sys.path -- I get the systemwide paths. I've > checked then when I do > > which sage > > it finds nothing (remember that I've called my script mysage). This > is extremely puzzling. > > Victor > > On Aug 4, 11:45 pm, Gonzalo Tornaria <torna...@math.utexas.edu> wrote: > > > Victor, > > > Do you have a script "sage" in your ~/bin which runs the system wide > > sage? If you run your local sage with "mysage", what happens when you > > run "sage"? What happens if you do > > > $ mysage -sh > > $ sage > > > will this run your local or the systemwide sage? > > > AFAIK, the notebook process which you run by issuing "notebook()" on > > the sage prompt doesn't execute the math itself, it only handles the > > notebook stuff. Rather, it spawns a new "sage" process, and > > communicates with it to do the math. So, it may be that for some > > reason it is spawning the system-wide install of sage rather than your > > own copy. > > > Don't recall the exact details, though, so I may be way off target. > > > Best, Gonzalo > > > On Tue, Aug 4, 2009 at 6:50 PM, VictorMiller<victorsmil...@gmail.com> wrote: > > > > It isn't. Systemwide SAGE is only in my path if I run a particular > > > script. I tried this in a fresh shell and checked that the systemwide > > > SAGE wasn't there. The problem was still there! > > > > Victor > > > > On Aug 4, 5:12 pm, William Stein <wst...@gmail.com> wrote: > > >> On Tue, Aug 4, 2009 at 12:12 PM, VictorMiller > > >> <victorsmil...@gmail.com>wrote: > > > >> > More info. When I type > > > >> > notebook() > > > >> > after typing the banner telling me to open my web browser > > >> > it prints a path to a system files copy of sob.py (not my local > > >> > copy!) > > >> > and a deprecation warning about the md5 module. > > > >> Why don't you try editing your PATH so that the systemwide sage isn't > > >> even > > >> in your PATH? > > > >> -- William > > --~--~---------~--~----~------------~-------~--~----~ 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 URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
