Hi Dave, thanks for your comments. On Tue, Sep 4, 2012 at 5:28 PM, David MacMahon <dav...@astro.berkeley.edu>wrote:
> Hi, Wes, > > Thanks for taking on this effort! Here are a few comments: > > On Sep 3, 2012, at 4:43 AM, Wesley New wrote: > > > Removing the following yellow blocks: > > XAUI - Only supported by BEE2 and iBOB > > I didn't realize that XAUI is not supported on the ROACH. I guess if > someone needs it on ROACH they can update the existing yellow block from > the BEE2/iBOB branch to work on ROACH and reintroduce it into the library. > Jason has corrected me on this, the XAUI block is supported by ROACH, so I wont be removing it. > > > The environment variable BEE_XPS_LIB_PATH is changing to XPS_BASE_PATH, > this will mean a change to your matlab startsg scripts > > You're probably planning on something like this anyway, but as a migration > aid, please update the scripts to check for the new variable first. If the > new variable is not set, then check for the old variable (and print a > deprecation warning). If you want, you could print a big error message > instead of a warning to force the user to update the variable right away, > but I suspect that some users may need the intervention of others to get > things like that changed. The main point is please don't have the scripts > fail silently/mysteriously if the user has not updated the name of this > environment variable. I guess the same thing goes for the value of the > variable (i.e. check the last component of the path to see if it is still > "xps_lib" and print a warning/error message). It's always nicer to get > friendly informative error messages than cryptic ones from tools failing in > unforeseen ways. > Good point, Ill definitely implement it in this way. > > I suppose there could be a corresponding cleanup of the BEE2/iBOB branch > (e.g. remove the QDR yellow block, etc.), but I imagine that is not high on > your priority list. :-) > As you noted, this is not on my list of To-Dos. > > Thanks again, > Dave > >