S29 update ready

[Aaron Sherman]

I've gathered my ducks in a row, used the feedback that I've gotten so
far, and I think I'm ready to officially update S29. For that I need two
things:

1) I'd really like Larry to glance over the changes and $s29.bless but
all comments are welcome
2) I'll need commit rights to whatever repository is authoritative for
Functions.pod these days

Here is the current version (I've added chr and ord docs now):

http://www.ajs.com/svn/ajs/perl6/Functions.pod

If anyone has any comments at all, please send me mail or respond to the
list. I'll try to address everything I can.

This is the first big update. There will be more if this all goes well.

[Larry Wall]

On Sat, Jul 08, 2006 at 06:04:40PM -0400, Aaron Sherman wrote:
: I've gathered my ducks in a row, used the feedback that I've gotten so

: far, and I think I'm ready to officially update S29. For that I need two
: things:
:
: 1) I'd really like Larry to glance over the changes and $s29.bless but
: all comments are welcome

I'll be glad to do that. But first things first...

: 2) I'll need commit rights to whatever repository is authoritative for
: Functions.pod these days

A pugs commit bit is on its way to you. Just keep checking it into the
pugs repository for now. It's a good place for collaboration. Don't
be afraid to check in intermediate results--we can always revert it. :)

Larry

[Joe Gottman]

I have one minor comment about join. You should specify its behavior when
it is passed an empty list. Does it return undef or the empty string?

Joe Gottman

[Darren Duncan]

At 8:32 PM -0400 7/8/06, Joe Gottman wrote:
>I have one minor comment about join. You should specify its behavior when
>it is passed an empty list. Does it return undef or the empty string?

I think it makes the most sense for it to return an empty string,
which is a reasonable degenerate case of a string-join. -- Darren
Duncan

[Darren Duncan]

Or to put it another way, I see string-join being like a prefix
version of the ~ operator (though with extra features), and so the
reduce rules of ~ should perhaps determine it. According to Synopsis
3, "[~]()" returns the empty string. -- Darren Duncan

[Aaron Sherman]

join returns a string, and requires a list. undef would indicate an
exceptional condition (for example the list contained an object which
could not be stringified), so I don't see why a valid (but empty) list
would cause an undef to be returned. The later point about C<[~]()> is
also quite helpful. I've made a note in my copy of S29, which will be
committed to the pugs repo as soon as I have commit rights.