Re: Requesting for review of the Draft proposal for - procfs

[Carl Fredrik Hammar]

<olafBuddenhagen@gmx.net> writes:

> I don't quite see the need for a special library here -- what would it
> provide over plain libtrivfs?...

It seems to me that many (most?) of the translators will be /very/
simple.  For uptime, cpuinfo, cmdline etc. their task boils down to
gather some info and produce a string using with asprintf().

This is much like the hello translator, except with a parameterized
string.  For something so simple the hello translator is still ~200
lines of code and I suspect it will get tedious and error-prone to
write several such translators when the actual functionality can be
summed up in 30 lines or so.

This won't cover all the proc translators, but I suspect there will be
a need to factor out commonalities in other cases as well.  It
perhaps a mistake to view it as a /single/ library.

That said, I agree with you that it's best to start with writing some
translators and *then* refactor them.  It's very easy to fall into the
``must handle every single case'' trap, as I know I have done on
occasion.  :-)

Regards,
  Fredrik