[BUG] large files?

Tim Rowledge tim at sumeru.stanford.edu
Thu Nov 8 04:48:34 UTC 2001


John M McIntosh <johnmci at smalltalkconsulting.com> is widely believed to have written:


> Ok, the file copy I know about, However the canonicalise file name 
> needs more input, er information. What do you want?
Most platforms can take a partial filename and canonicalise it such that
'bar.html' might be returned as
'/home/lusers/bush/webPages/foo/bar.html' or
'c:\thingy\fooble\foo\bar.html' etc. The nice about having the os do it
is that you can point the finger atthe vendor if it produces duff
filenames. Writing ones own algorithm can be unutterably painful; you
have to find out all the rules, both documented and hidden, for all
theplatforms... yuck. Been there, done that, still got the mental scars.

A prim that takes your potential filename string and passes it through
the OS and hands back the mangled, decapitated, shortened, rooted result
saves much anguish. Hopefully.

While we're worrying about filenames, how about adopting the technique
we settled upon at Interval; all filenames are URLs, and only the file
handling prim code ever sees anything else. People are getting used to
URLs, they have reasonably documented rules, they open the door to
a nice interace to networks, all sorts of advantages.

tim

-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
"How many Motie Warriors does it take to change a lightbulb?" "None. One of the dead ones will do it."





More information about the Squeak-dev mailing list