ref: reimplement uuid init without string subscript extensions

[armcknight]

I noticed this at first by discovering the tests for the string subscripts.

First, I think subscripts are generally bad to use as a custom extension, because discoverability is nonexistent. You can't ⌘ click on them to jump to their def. You can't ⌘ O to quick open to them.

Second, we never even had a test for the one thing they were being used for: creating a UUID from a string. (I added such a test)

Third, they were used in one spot. I rewrote that spot to use a reduce. It turns out to be faster, almost twice as fast:

before:

relative standard deviation: 147.699%, values: [0.000473, 0.000064, 0.000047, 0.000044, 0.000042, 0.000041, 0.000041, 0.000040, 0.000040, 0.000040]

after:

relative standard deviation: 179.733%, values: [0.000387, 0.000036, 0.000027, 0.000024, 0.000023, 0.000023, 0.000022, 0.000022, 0.000021, 0.000021]

#skip-changelog

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 91.380%. Comparing base (0f4071f) to head (11e5d06). Report is 2 commits behind head on main.

Additional details and impacted files

@@              Coverage Diff              @@
##              main     #4133       +/-   ##
=============================================
- Coverage   91.387%   91.380%   -0.008%     
=============================================
  Files          606       605        -1     
  Lines        48301     48260       -41     
  Branches     17434     17396       -38     
=============================================
- Hits         44141     44100       -41     
  Misses        4067      4067               
  Partials        93        93               

Files	Coverage Δ
Sources/Swift/Extensions/StringExtensions.swift	100.000% <ø> (ø)
Sources/Swift/Protocol/SentryId.swift	100.000% <100.000%> (ø)
Tests/SentryTests/Protocol/SentryIdTests.swift	100.000% <100.000%> (ø)

... and 3 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 0f4071f...11e5d06. Read the comment docs.

[github-actions[bot]]

Performance metrics :rocket:

	Plain	With Sentry	Diff
Startup time	1223.04 ms	1235.63 ms	12.59 ms
Size	21.58 KiB	682.20 KiB	660.62 KiB

Baseline results on branch: main

Startup times

Revision	Plain	With Sentry	Diff
98a8c16494c5fb68f2218a5635a2443961ce7749	1234.69 ms	1265.02 ms	30.33 ms
8e76be4558a9e15fe63841bbb98faec7d921f24e	1272.67 ms	1286.38 ms	13.71 ms
591a01b3e320697e7cc86082284d786c359306e3	1197.94 ms	1222.53 ms	24.59 ms
5cabf7ed4789ca11913d7c24b6f8d9b66578cd26	1247.12 ms	1249.61 ms	2.49 ms
b9a9ffd9ab91b555e7b798a60311876a15b87b01	1221.18 ms	1235.37 ms	14.19 ms
c291fcc0ff35822d65951cf8c202a99bc129c7eb	1204.60 ms	1226.59 ms	21.99 ms
dacf894bd3ce125f76ea565dbc543db3da8bf94d	1223.96 ms	1250.41 ms	26.45 ms
7bc3c0d8e54a5921c8c1a9510088c683dba2c1b2	1261.16 ms	1278.38 ms	17.22 ms
dd0557fb0fec40f5527e4f40ead6f82e092fde0f	1246.31 ms	1258.46 ms	12.15 ms
ff5c1d83bf601bbcd0f5f1070c3abf05310881bd	1236.67 ms	1252.34 ms	15.67 ms

App size

Revision	Plain	With Sentry	Diff
98a8c16494c5fb68f2218a5635a2443961ce7749	20.76 KiB	431.00 KiB	410.24 KiB
8e76be4558a9e15fe63841bbb98faec7d921f24e	20.76 KiB	427.66 KiB	406.89 KiB
591a01b3e320697e7cc86082284d786c359306e3	22.84 KiB	401.67 KiB	378.83 KiB
5cabf7ed4789ca11913d7c24b6f8d9b66578cd26	20.76 KiB	419.67 KiB	398.91 KiB
b9a9ffd9ab91b555e7b798a60311876a15b87b01	21.58 KiB	418.43 KiB	396.85 KiB
c291fcc0ff35822d65951cf8c202a99bc129c7eb	21.58 KiB	669.82 KiB	648.24 KiB
dacf894bd3ce125f76ea565dbc543db3da8bf94d	20.76 KiB	426.34 KiB	405.58 KiB
7bc3c0d8e54a5921c8c1a9510088c683dba2c1b2	20.76 KiB	427.36 KiB	406.59 KiB
dd0557fb0fec40f5527e4f40ead6f82e092fde0f	22.85 KiB	411.75 KiB	388.90 KiB
ff5c1d83bf601bbcd0f5f1070c3abf05310881bd	20.76 KiB	430.98 KiB	410.22 KiB

Previous results on branch: armcknight/ref/remove-swift-string-subscript-extensions

Startup times

Revision	Plain	With Sentry	Diff
5b664d26b6ae4784c8ce1eaaf286418183781e10	1222.96 ms	1242.90 ms	19.94 ms

App size

Revision	Plain	With Sentry	Diff
5b664d26b6ae4784c8ce1eaaf286418183781e10	21.58 KiB	678.23 KiB	656.65 KiB

[brustolin]

First, I think subscripts are generally bad to use as a custom extension, because discoverability is nonexistent. You can't ⌘ click on them to jump to their def. You can't ⌘ O to quick open to them.

This is a matter of culture building. After we get used to it, there will be no doubt about how it works. It's also not possible to "⌘ O" the original string subscript, and that is not a problem.

This particular extension is very useful.

But, I don't mind changing this piece of code. If it works, it works.

[armcknight]

This is a matter of culture building

Absolutely. I'm trying to build a culture that results in a codebase that is simpler to understand and maintain. Swift is a very flexible language, and thus it's very easy to create an incomprehensible mess of code. I've watched people do it over and over again. This lesson should have been learned with C++ operator overloading, but here we are. "...so preoccupied with whether or not they could that they didn't stop to think if they should..." and I'm saying we shouldn't.

It's also not possible to "⌘ O" the original string subscript, and that is not a problem

This is a matter of opinion. In my opinion, it is a problem. In my opinion, Swift's substring APIs are horrible. Have fun writing a regex to grep through a codebase for a usage of a subscript, but only the one used for a substring access. It would be much easier to grep for .substring(from:to:).