On Sun, 28 Jan 2007 04:54:21 -0800, David Harvey <[EMAIL PROTECTED]> wrote:
> > A minor annoyance: > > Yesterday I found I would like to be able to create new cells in the > middle of documentatino text. Currently you can only create new cells > next to existing ones. But sometimes the tutorial has so much text > that means you have to go scrolling around to find a cell so that you > can add a new one. I'm not sure what is the best way to deal with this. That is simply a NotImplementedError -- no more no less -- the solution is to implement adding new cells by say clicking above or below any existing cell. It's simply not implemented yet though. > I noticed some bugs on a windows machine running Mozilla 1.7.13. I > couldn't scroll the documentation window with the arrow keys, even if > I clicked on the background of the tutorial frame. Also, whenever I > evaluated a cell using shift-enter, the evaluation worked, but the > page instantly scrolled right back to the top, so I had to scroll > back down (using the scroll bar) every time I evaluated anything. That is *exactly* the sort of behavior I would expect if two people were using the public doc browser at the same time -- as I mentioned in my first email, they would be using the same underlying worksheet, and would get weird behavior if they both evaluate. This is really another NotImplementedError. I don't know a good solution for this; it's not a problem of it being a lot of work, but basically I have to mull the problem over and wait-until-I-have-a-flash-of-insight. Basically it is good that all the worksheets as you browse through the docs are all actually the exact same worksheet in SAGE -- this means only one PYthon process gets started if you eval cells -- which is good since it makes it feel snappier (no startup time, even when you go to the next page). Also, this is good, since sage processes still don't timeout... An alternative would be to generate a different worksheet (and hence different sage process for each page of docs you view. But then if you evaluate cells and move to a new page of the docs, there are two sage worksheets running, etc., Also, there will be tons of worksheets crufting up your notebook corresponding to tons of pages. So at a minimum they have to be cleared out when you exit. That would work, but it's more complicated, and it's really sad to keep starting a new python process when one evaluates cells. But maybe that makes a lot more sense -- it is exactly the sage model right now -- one process per worksheet. Right now, if a user opened the docs on her computer to two points, and tried to evaluate two examples at the same time, hell would break loose or they could conflict in weird ways. With a change as suggested above everything would work as expected, but there are efficiency issues. [...] OK, I just implemented this (with a rotating pool of 48 (say) doc_browser worksheets that get cleand up, so it'll be in SAGE-2.0. It's better that things work precisely as expected correctly, but potentially less efficiently. -- William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
