Torsten Bronger wrote: > Hallöchen! > > I have a file that looks a little bit like a C header file with a > long list of variables (actually constants) definitions, e.g. > > VI_ATTR_TIMO = 0x54378 > ... > > Actually I need this in a couple of low-level modules that are > imported into the main module, and in the main module itself. They > may be also used in the programs that import the main module. > > Probably it doesn't mean a significant loss of performance anyway > since all of it is done only at startup, but I wonder whether it's > better to keep everything that depends on this "header" file > together so that it must be looked over by Python only once. Is > this correct, or is there some sort of implicit optimisation that > makes both variants almost equivalent? >
I'm not entirely clear what you are trying to do *but* - if you import the same module in several places (per interpreter instance of course) the import will only be done *once*. The other import statments just make that namespace available from the namespace that does the import. This means there is little extra overhead for importing a module that has already been imported elsewhere. HTH ? Regards, Fuzzy http://www.voidspace.org.uk/python/index.shtml > Tschö, > Torsten. > > -- > Torsten Bronger, aquisgrana, europa vetus -- http://mail.python.org/mailman/listinfo/python-list
