Swiftweekly.github.io: [142] Issue #142 - September 5, 2019
To contribute to this issue, simply leave a comment here. Please also review our contributing guidelines.
The current draft for this issue in _drafts/. If you want to contribute directly, feel free to open a pull request.
BasThomas
All 28 comments
Rearchitecting JSONEncoder to be much faster
Currently, JSONEncoder can spend a lot of time doing the following things:
- Creating intermediate Swift/ObjC data structures to pass to JSONSerialization
- Creating a new Swift String for each key conversion
Inside of JSONSerialization, using libc functions to convert numbers to stringsAll of these things can be resolved with a different architecture. There may be other good ones, and if so I'd love to hear about them, but here's what I'll pitch:
Whenever an encode method is called, write the JSON out directly, rather than storing anything in dictionaries or delegating any work to JSONSerialization. The one tricky bit is that it may be told to encode something that needs to come later after other, unseen, pieces of data are encoded. For example:
https://forums.swift.org/t/rearchitecting-jsonencoder-to-be-much-faster/28139
fassko
on 22 Aug 2019
Three outstanding proposals
Accept Values for Custom Compilation Flags
Single Quoted Character Literals
Collection conditionals SR-8743
https://forums.swift.org/t/three-outstanding-proposals/28138
fassko
on 22 Aug 2019
After a lot of fiddly work, the Swift runtime and standard library no longer depend on Foundation! (they’ll still use it for bridging if it’s present)
https://github.com/apple/swift/pull/26630
Next up: finding out where the last thing linking CoreFoundation is hiding and decoupling from that too
fassko
on 24 Aug 2019
262 returned for review
https://forums.swift.org/t/returned-for-revision-se-0262-demangle-function/28186
fassko
on 24 Aug 2019
Formally revoking SE-0042
SE-0042 was formally accepted more than three years ago. However, since the proposal was accepted before having a working implementation was a requirement, and the change would be source breaking, and also require massive rewrites of the compiler; it is now clear that this proposal will never actually be implemented. It has been obvious for a long time.
fassko
on 24 Aug 2019
A new binary-search attempt
Continuing the discussion from An always-sorted set:
I just updated the gist to Revision 2, which updates the "BinarySearch.swift" file.
https://forums.swift.org/t/a-new-binary-search-attempt/28195
fassko
on 24 Aug 2019
Confirming a sort or partition
Continuing the discussion from A new binary-search attempt:
I was thinking of some extension from the binary search on collections that can work for sequences. Then I came up with checking if a sequence matches a given partition (with a partitioning predicate) or sort (with a comparison predicate and monotonic flag). These match the std::is_partitioned, std::is_sorted, and std::is_sorted_until functions from C++.
https://forums.swift.org/t/confirming-a-sort-or-partition/28206
fassko
on 24 Aug 2019
What is libswiftabi /
Following the returned for revision announcement for SE-0262, one of the bullet points to discuss was a new "Runtime" module in Swift, separate to the stdlib. I want to open discussion on what that might look like and what other features it might entail. The core team believes that a demangle function would make for a "natural candidate" for such module, but Joe also mentioned that other functionality from the Swift runtime could be used for things like reflection. Could this module be where full reflection support would be implemented, or just some of the building blocks? What other "low-level" features could this potentially offer?
Another important question to consider: what would be a good name for this module? Kidding aside, it seemed implied that we could name this Runtime, but considering potential other features, are there any other names that could fit the bill as well?
https://forums.swift.org/t/what-is-libswiftabi-swift-runtime-h/28201
fassko
on 24 Aug 2019
Future of Diagnostics UX
Lately, I've been thinking about how the diagnostics user experience for Swift could be improved in the long term. I thought it would be valuable to start a conversation around what kind of improvements people would like to see and what we can learn from other compilers, maybe resulting in some kind of rough roadmap if there's enough interest.
fassko
on 24 Aug 2019
https://twitter.com/jckarter/status/1166029901335412738
Another cool optimization from Erik to eliminate temporary copies during in-place updates of enum payloads like:
switch e {
case .A(var arr):
arr.append(x)
e = .A(arr)
https://twitter.com/Catfish_Man/status/1166786023667318784
Extremely cool proposal from @Ilseman for slicing and dicing collections ergonomically in Swift
https://forums.swift.org/t/pitch-offset-indexing-and-slicing/28333
BasThomas
on 28 Aug 2019
Allowing explicit @escaping for optional closures in function parameters
I drafted a formal proposal to introduce explicit @escaping for optional closures in function parameters. Unfortunately, I don't have the necessary knowledge to build a working implementation for this proposal. I was hoping that someone could point me on the right direction or pair me in the process.
fassko
on 29 Aug 2019
Adding a “does X conform to Y” request
https://forums.swift.org/t/adding-a-does-x-conform-to-y-request/28349
fassko
on 29 Aug 2019
fassko
on 29 Aug 2019
fassko
on 30 Aug 2019
https://twitter.com/johannesweiss/status/1167538133401067520
Coincidentally @gottesmang just demonstrated how much a bug+benchmark can help:
The NIO ChannelPipeline benchmark runs in HALF the time. This improvement should make a real difference in NIO!
Michael, thanks so much, this is awesome.
https://github.com/apple/swift/pull/26956 https://twitter.com/johannesweiss/status/1167526400330539009
BasThomas
on 31 Aug 2019
Starter bug https://bugs.swift.org/browse/SR-11397
fassko
on 31 Aug 2019
https://twitter.com/swift_unwrapped/status/1168525041434333184
Episode 78: Binary Dependencies in Swift Package Manager https://spec.fm/podcasts/swift-unwrapped/307679
BasThomas
on 2 Sep 2019
fassko
on 4 Sep 2019
Pitch: Fully qualified name syntax
Swift's name lookup is…messy. Very messy. This causes a number of problems, and today I'd like to talk about a potential solution for one of them.
...
https://forums.swift.org/t/pitch-fully-qualified-name-syntax/28482
fassko
on 4 Sep 2019
The August release is out: Swift 5.0.3 for Linux, here we go: https://forums.swift.org/t/swift-5-0-3-for-linux/28523 .
Thanks very much for the contributions! Docker images should be available soon too.
https://twitter.com/johannesweiss/status/1169284396718968833
fassko
on 4 Sep 2019
Starter bugs:
https://bugs.swift.org/browse/SR-11421
https://bugs.swift.org/browse/SR-11420
https://bugs.swift.org/browse/SR-11419
https://bugs.swift.org/browse/SR-11418
fassko
on 5 Sep 2019
Great article about sort in Swift https://swiftrocks.com/introsort-timsort-swifts-sorting-algorithm.html
fassko
on 5 Sep 2019
263 in review
BasThomas
on 5 Sep 2019
Thanks again for all the great help, @fassko — I wouldn't be able to compile the newsletter currently without you. 😊
BasThomas
on 5 Sep 2019
@BasThomas no worries, if it's too much in your place I could take over some of the next issues.
fassko
on 6 Sep 2019
It's OK — especially gathering the content helps. 😊
BasThomas
on 6 Sep 2019
Related issues
jessesquires
·
10Comments
jessesquires
·
6Comments
gregheo
·
3Comments
jessesquires
·
4Comments
BasThomas
·
4Comments
Most helpful comment
https://twitter.com/swift_unwrapped/status/1168525041434333184