* Stephen P. Potter ([EMAIL PROTECTED]) [18 Aug 2000 03:06]:
> I don't see these ideas in RFCs:

[Reordered a bit and some items removed]

> * Socket (such as C<accept>, C<bind>, etc)
> * Math (such as C<abs>, C<atan2>, C<cos>, etc)
> * IPC (such as C<msgctl>, C<semctl>, C<shmctl>, etc)
> * "User" (C<getpw*>, <endpw*>, C<setpw*>, C<getlogin>)
> * Group (C<getgr*>, C<endgr*>, C<setgr*>)
> * Network (C<getnet*>, C<gethost*>, C<getproto*>, etc)
> * Process (C<getpgrp>, C<getppid>, C<getpriority>, C<setpgrp>, C<setpriority>)

For socket stuff, a lot of people use IO::Socket, so that would not be a
loss (but IO::Socket would need to be rewritten to not use them if it
does). The rest, probably a good idea. Would it make things much
smaller (most of the above are native Unix calls so would have minimal
overhead [he says, not having seen the source code])?


[...]
> * The match operator, C<m>, is always required (bare C<//> becomes a
> fatal error).

Why? Seems a waste. It would just add more clutter to many lines.

> * Replace C<??> with flag to C<m//>, and remove special meaning of
> C<m??>.

Fine by me, but then, I don't use ?? often. (Maybe I should.)

> * Replace C<m//>, C<tr///>, and C<s///> with equivalent regularized
> functions that take mulitple arguments instead of using specialized
> syntax.  It would be best if the names could be more "complete", like
> match(), translate(), and substitute() (although translate and
> substitute are rather long).

Far too long. Keep them short --- they're useful and frequently used.
Hence, I suggest: m, tr and s. =)

> * Disallow use of C</> as delimiters for quote functions and require
> use of matching pair characters.

Why? What if my text happens to use the various matching pair
characters? I *like* using / because my text very rarely has it in it
but is littered with [{()}]"' etc.

> * Remove special meaning from C</>.  Integrate that functionality with
> C<'> and C<">.

Why and how?

> * Remove C<y///> as a synonym for C<tr///>

Why?

> Some of these are mine, some of them are ideas I remember seeing
> someone else mention, but I don't see RFCs for.  Am I missing them, or
> do they need to be written up? 

They do --- preferably with complete justification behind them. A lot of
them seem to be pointless reshuffling of things to make using them
harder.

-- 
iain truskett, aka Koschei.                    <http://eh.org/~koschei/>
     Q: How do I find the largest element in an array?
     A: Write a foreach elements one at a time, and stop when you get to
        the largest one.                                     -- perliaq.

Reply via email to