HTTP multipart reading is pretty easy to implement. You just want to be
careful about space&time performance -- not only because you might want
lots-and-lots of this processing going on on each server (and the costs
add up), but because the individual parts are quite often large (for the
purposes to which people often put multipart).
(Side note: This seems like one of the many essential library things
that might quickly shake out of a startup with strong Racketeers who
contribute back high-quality open source packages for their
non-trade-secret modules. Many of these packages *could* also be done
by a hobbyist volunteer, but a startup figuring out real-world
function/interface/performance requirements, and needing to do it and
make it work, might be more likely to do produce what's actually needed
by other practitioners. A few Racket startups doing this, and maybe you
get an effect like the story of the person who, to design footpaths
through a university quad, instead planted grass, and let the
grass-trampling by students getting between classes determine where to
place good footpaths.)
--
You received this message because you are subscribed to the Google Groups "Racket
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
