Roy Smith wrote: > In article <[EMAIL PROTECTED]>, > "John Machin" <[EMAIL PROTECTED]> wrote: > > > Roy Smith wrote: > > > > > It would be nice if struct.unpack() had a way to specify unpacking > > > repeated > > > items as a list, ie: > > > > > > data = struct.unpack ("! H 4(B) H 2B 12(B) 6(B) H I", strMessage) > > > (top, > > > ip, > > > messageCounter, > > > ackRequired, > > > dataType, > > > utc, > > > st, > > > numberOfLables, > > > dataWord) = data > > > > > > and you'd end up with ip, utc, and st being lists. > > > > :-) > > Extension: if you used [] in the format string, the results would be > > tuples. > > :-) > > I see your point. Actually, I think you want to force the sequences to be > lists (regardless of what syntax we end up with), on the theory that tuples > are for heterogeneous sequences and lists are for homogeneous ones.
And array.arrays are for sequences that are even more homogeneous :-) However, after reflection I'd go for lists on practical grounds, because they're mutable and tuples aren't -- often one wants to start fiddling with the input. > As for > the parens, I was thinking fortran-style format specifiers, which may or > may not be the best model for a modern language :-) Yes, please don't perpetuate 10HANTIQUATED concepts :-) > > > > Hmmm, maybe that's > > > worth a PEP? > > > > Great minds think more-or-less alike :-) > > I was in the process of writing one up when I saw your earlier post. I'll > mail you a draft when I've got it done. Great. Looking forward to it. Cheers, John -- http://mail.python.org/mailman/listinfo/python-list