[swift-evolution] [Review] SE-0159: Fix Private Access Levels

[Nevin Brackett-Rozinsky]

On Sun, Mar 26, 2017 at 12:57 PM, David Sweeris via swift-evolution <
swift-evolution at swift.org> wrote:

>
> On Mar 26, 2017, at 08:50, David James via swift-evolution <
> swift-evolution at swift.org> wrote:
>
> I propose instead that we revise to use Alternative #3, per Vladimir’s
> comment and revision.
>
> Revised version:
>
> *“3. Revert private to be file-based and introduce the scope-based access
> level under a new name (e.g.: scoped, local, etc), provided that the
> scope-based access modifier is not used at the top level of the file.” *
> (addendum via Vladimir’s revised comment)
>
>
> Yeah, within reason, I couldn't care less how "private"/"fileprivate" are
> *spelled*. What I'm against is removing the *functionality* of the
> current "private" without simultaneously providing a semantically
> equivalent replacement.
>


Conversely, I do not particularly care whether the functionality of
scope-based visibility stays or goes, but I find the spelling “fileprivate”
to be an ugly wart on an otherwise beautiful language.

Before the discussion on this review, I had a slight preference for
removing scope-based visibility in order to simplify the access control
model. After reading through the thread I have a slight preference for
keeping that functionality to enable compiler verification that invariants
are only touched via dedicated channels. I do not have a strong opinion
either way, and I would be fine to see scope-based visibility exist under
the name “scoped”, or be removed entirely.

However, the “fileprivate” keyword is one of the few things I actually
*dislike* about Swift. As in, when I am writing code I find it annoying and
offputting to use such a lengthy and inelegant term for one of the most
common access levels. I think it was a mistake to change the meaning of
“private” in SE-0025, and we should change it back posthaste.

Nevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.swift.org/pipermail/swift-evolution/attachments/20170326/7023554a/attachment.html>