|
|
Magic Mailing List |
|
From: Stefan Jones (stefan.jones AT multigig DOT com) Date: Fri Mar 28 2003 - 15:31:57 EST
On Fri, 2003-03-28 at 11:39, cfk wrote: > And yes, there are now two graphs, one of the input and one of the output of > a magic inverter cell. By, the way, I can just do a > > %;load /usr/lib/libspice > > and get everything running without the 'package require spice'. I think the > issues are very minor and perhaps could be just worked out so the > occaisional user can get his magic/irsim/tclspice working without having to > understand everything about paths, packages and such. I am hoping you might > consider working to make it easier on the user. > In this case it works, but in general if you do it from a wish command line it will cause some breakages with the plot readline completion and BLT (I think). I guess the "package require spice" step should be kept so the normal tcl programmer can make use of tclspice, but a button or alias should be made in magic to do this for you. I guess this is up to the magic guys. > I also have a couple of suggestions along that line. Both magic and irsim > have a make config step, tclspice has arguments to ./configure. Perhaps > tclspice could have a make config step and avoid the ./configure arguments > and have its make procedure more like irsim & magic (or conversly, magic > could have its make config changed to ./configure arguements). Similarly, > magic and irsim do a 'make tcl' followed by a 'make install-tcl' and > tclspice does a 'make tcl' followed by a 'make tcl_install'. Ok, I will change tcl_install to install-tcl to unify that step. I will not do the make config -> ./configure change as the tclspice Makefiles are autoconf / automake based thus require configure. ./configure --help is useful and works. Will do a new tclspice release soon when I have it mostly fixed. Thanks for the info, Stefan -- Stefan Jones <stefan.jones AT multigig DOT com> Multigig Ltd
|
|
|
|