On Dec 30, 2007, at 10:06 AM, Bob Hanson wrote: > possibly, but let's talk first about what you are really interested in > doing, then talk format. Arrays aren't necessarily the solution.
I agree here. > Pmesh > is not what you want for simple planes and objects -- that is for > complex mathematical descriptions of surfaces. Using specific > colorings > and shadings sounds like Jmol scripting to me. So I think you are > talking about a mix of objects, some of which are memory/filespace > intensive, such as mathematical surfaces, and some of which are simple > objects. Yes. Scripts make it very easy to build up scenes like this. As for format, having a single zip file (with a MANIFEST that could specify a script to load, which could refer to other files in that same zip file by name) would probably be the best solution. > Give us a few scenarios to work on. What would these sage-results > entail? The simplest example is a 3d plot. This is perfectly represented by a pmesh. We might have points (spheres and/or labels) attached to this, more than one plot overlayed (or side-by-side), and would like to have axes too. Another common object would be a (combinatorial) graph--spheres for nodes connected by lines. As an aside--what is the best way to create a line (both thick, e.g. a cylindrical "molecular bond" although not necessarily between to "atoms") and a plain line (for axes, e.g. the bounding box of http:// pages.pomona.edu/~gk014747/teaching/Fall2007/ilkplot.gif )? > Q: Do you ever map surface data with other data so as to color it > with, > say, a scalar field? Not yet, but could be very useful and valuable. For example, when trying to represent a complex-valued function in 3 dimensions, the magnitude can be used for the z-value and color can represent the argument. Or even for plain 3d plots color can be used (in addition to the z-value) to represent height (as in topo maps). And I can imagine it could be useful to map arbitrary color data to a plot too. This is what the voxel stuff is about, right? > > Bob > > Fernando Perez wrote: > >> Howdy, >> >> On Dec 29, 2007 10:59 PM, Robert Bradshaw >> <[EMAIL PROTECTED]> wrote: >> >> >> Just as an FYI: as of the last few days, numpy has developed a binary >> format for arbitrary arrays. The current plan is to have the base >> file format (default extension .npy, but there's a magic string >> header >> for extension-less identification) contain single arrays, and to use >> zip files for multi-array files with a dict-like interface. >> >> It's in a branch right now, here's the format spec: >> >> http://projects.scipy.org/scipy/numpy/browser/branches/lib_for_io/ >> format.py >> >> >> > Bob > > -- > Robert M. Hanson > Professor of Chemistry > St. Olaf College > Northfield, MN > http://www.stolaf.edu/people/hansonr > > > If nature does not answer first what we want, > it is better to take what answer we get. > > -- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900 > > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com 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/ -~----------~----~----~----~------~----~------~--~---
