Fix thousands of typos across hundreds of files. #74983

Sajjon · 2024-07-05T07:40:48Z

Fix thousands of typos across hundreds of files.

Might contain some breaking ones, I might need help identifying which ones to revert.

I've used the awesome CLI tool typos to create this PR in combination with a couple of hours of manual work to sort between false positives, and some "weak true positives" (using command typos -w the cli tool will automatically change files for "strong" positives, but for some typos it is "less sure" about, it will not make any changes on disc, but output those in the terminal as probable typos, "weak positives" I call those, some of which are true, some of which are false).

I will remove the _typos.toml config file before merging this PR (and possibly other PRs "split out" from this, if we go with the "batch" approach).

N.B. I've done a lot of manual work apart from using the typos cli tool, which have taken many hours, e.g. git added some true positives of some files, but ignored some false positives in the same files, and then "typos ignored" those files. The implications of this is that if you were to run the same version of typos with the _typos.toml config file, you would NOT get the same result, some typos would be missed.

I did it like this since many files contained lots of specific false positives, where it made no sense to add those strings to extended identifiers list.

AnthonyLatsis · 2024-07-05T11:00:46Z

Do you mind breaking this up into several PRs so that it is hundreds of typos over tens of files? It will be easier to review for us, and you will be less likely to catch a conflict or change request on a given PR.

Sajjon · 2024-07-05T11:08:30Z

Do you mind breaking this up into several PRs so that it is hundreds of typos over tens of files? It will be easier to review for us, and you will be less likely to catch a conflict or change request on a given PR.

@AnthonyLatsis sure I could do that, will take me some time to change into that though (I'm on vacation now and have limited time available).

EDIT:
Perhaps 10 PRs with 53 changed filed each would be reasonable ?

Sajjon · 2024-07-05T11:17:31Z

@rjmccall you reviewed my PR swiftlang/swift-evolution#2479 (comment)

Any input on this one, and if you want me to chunk it up into several PRs, if so, how many?

AnthonyLatsis · 2024-07-05T11:23:05Z

Perhaps 10 PRs with 53 changed filed each would be reasonable ?

You can surely do that, but even 4 PRs with up to 400 additions should be fine.

Lancelotbronner · 2024-07-05T13:44:25Z

benchmark/single-source/SortIntPyramids.swift

@@ -37,7 +37,7 @@ let pyramidTemplate: [Int] = (1...pH) + (1...pH).reversed()
 // A^R - reversed array A, + - array concatenation operator,
 // A indices are in range 1...A.length.
 // define adjacent pyramid as A + A^R + A + A^R,
-// defne adjacent pyramid hight as A[A.length].
+// defne adjacent pyramid height as A[A.length].


defne -> define?

Sajjon · 2024-07-05T13:51:39Z

_typos.toml

@@ -0,0 +1,263 @@
+[files]


This is the config file for the awesome CLI tool typos I've used to create this PR.

I will remove it before merging this PR (and possibly other PRs "split out" from this, if we go with the "batch" approach).

N.B. I've done a lot of manual work apart from using the typos cli tool, which have taken many hours, e.g. git added some true positives of some files, but ignored some false positives in the same files, and then "typos ignored" those files. The implications of this is that if you were to run the same version of typos with this config file, you would NOT get the same result, some typos would be missed. I did it like this since many files contained lots of specific false positives, where it made no sense to add those strings to extended identifiers list.

EDIT: this was relevant info, copied to PR description.

rjmccall · 2024-07-05T21:17:07Z

docs/SIL.rst

@@ -1181,7 +1181,7 @@ information when inlined.
 The function is a compiler generated thunk.
 ::

-  sil-function-attribute ::= '[dynamically_replacable]'
+  sil-function-attribute ::= '[dynamically_replaceable]'


I regret to inform you that this is correctly documenting the actual attribute.

rjmccall

I would suggest breaking it up into groups that a reviewer could reasonably scan through in about 5 minutes. I got down to RegionAnalysis.h before I started losing focus and needed a break.

include/swift/ABI/Task.h

include/swift/AST/Decl.h

beccadax · 2024-07-05T21:29:15Z

include/swift/AST/DiagnosticsSema.def

@@ -7497,12 +7497,12 @@ ERROR(wrap_invalid_attr_added_by_access_note, none,

 ERROR(cannot_default_generic_parameter_inferrable_from_another_parameter, none,
      "cannot use default expression for inference of %0 because it "
-      "is inferrable from parameters %1",
+      "is inferable from parameters %1",


While "inferable" is a more common spelling, "inferrable" is used in some published works:

Swift is consistent enough in its use of "inferrable" that I'd be inclined to keep it.

reverted all inferrable -> inferable changes

include/swift/DependencyScan/DependencyScanningTool.h

beccadax · 2024-07-05T21:37:18Z

include/swift/SILOptimizer/Analysis/RegionAnalysis.h

@@ -357,15 +357,15 @@ class RegionAnalysisValueMap {
 public:
  RegionAnalysisValueMap(SILFunction *fn) : fn(fn) { }

-  /// Returns the value for this instruction if it isn't a fake "represenative
+  /// Returns the value for this instruction if it isn't a "fake" representative


The change to quotation marks here is incorrect, although the change to spelling is fine. (The same goes for later changes in this file.)

It is consistent with the rest in the file. Many "fake" occurrences. But i will revert/take another look!

compnerd · 2024-07-05T21:43:41Z

docs/Generics/chapters/completion.tex

@@ -742,7 +742,7 @@ \section{Associated Types}\label{critical pairs}
 \arrow[u, Rightarrow, "\ttgp{0}{0}.(\assocsym{P}{A}.\assocsym{Q}{B}\Rightarrow\assocsym{P}{A}.\texttt{B})"']&&
 \end{tikzcd}
 \]
-Note that this was an overlap of the first kind, so rule (7) is now marked \index{left-simplified rule}\textbf{left-simplified}. Thus, rule (*8) completely supercedes rule (7). Rule (*8) survives minimization and maps to the requirement $\ConfReq{\ttgp{0}{0}.[P]A.[Q]B}{R}$, which appears in the generic signature that we output for this protocol extension:
+Note that this was an overlap of the first kind, so rule (7) is now marked \index{left-simplified rule}\textbf{left-simplified}. Thus, rule (*8) completely supersedes rule (7). Rule (*8) survives minimization and maps to the requirement $\ConfReq{\ttgp{0}{0}.[P]A.[Q]B}{R}$, which appears in the generic signature that we output for this protocol extension:


OED indicates that supercede is permissible. It is no longer "standard" but it is still a correct spelling.

I'm fine with treating this as a typo. There's no good reason to use non-standard spellings, even if they were previously accepted.

beccadax

These changes are good, but they aren't so good that they don't need to be reviewed carefully, and this PR is much too large to review in one go.

I'd recommend you try to break this up into individual PRs by compiler subsystem (e.g. each include/ folder and its matching lib/ folder, plus any changes in other files that were forced by changes in the include/ folder) so that each PR is focused enough to find an appropriate set of reviewers for it.

Sajjon · 2024-07-06T11:42:49Z

@rjmccall

I would suggest breaking it up into groups that a reviewer could reasonably scan through in about 5 minutes. I got down to RegionAnalysis.h before I started losing focus and needed a break.

@beccadax

I'd recommend you try to break this up into individual PRs by compiler subsystem (e.g. each include/ folder and its matching lib/ folder, plus any changes in other files that were forced by changes in the include/ folder) so that each PR is focused enough to find an appropriate set of reviewers for it.

I'm doing just that, it will be ~20 PRs. I'm dividing them up in:

Benchmark
docs
stdlib
SwiftCompilerSources
test, multiple batches:
- test/ide
- test/SourceKit
- test/SILOptimizer
- test/SILGen
- test/IRGen
- test/Interop
- test/MISC
lib, multiple batches, work in progress
include, multiple batches, work in progress

While doing so, I'm applying the suggestions/corrections you have already reviewed.

I need one more day.

If it is helpful, I can create an "epic issue" for this, with links to all PRs?

I will probably close this PR once I've opened the other PRs.

Any input? Or sounds good?

compnerd · 2024-07-06T17:20:32Z

I think that doing the lib and include simultaneously is better - the headers are associated with the libraries, and so the reviewers are going to be the same and sending them together makes more sense as the same area is being looked at.

Sajjon · 2024-07-07T13:26:57Z

@compnerd @beccadax @rjmccall @AnthonyLatsis @Lancelotbronner I'm closing this PR in favour of tracking issue: #75022 which links to 14 PRs

Fix thousands of typos across hundreds of files.

e49e1c5

Lancelotbronner reviewed Jul 5, 2024

View reviewed changes

Sajjon commented Jul 5, 2024

View reviewed changes

rjmccall reviewed Jul 5, 2024

View reviewed changes

beccadax reviewed Jul 5, 2024

View reviewed changes

include/swift/ABI/Task.h Show resolved Hide resolved

beccadax reviewed Jul 5, 2024

View reviewed changes

include/swift/AST/Decl.h Show resolved Hide resolved

beccadax reviewed Jul 5, 2024

View reviewed changes

include/swift/DependencyScan/DependencyScanningTool.h Show resolved Hide resolved

beccadax reviewed Jul 5, 2024

View reviewed changes

compnerd reviewed Jul 5, 2024

View reviewed changes

beccadax requested changes Jul 5, 2024

View reviewed changes

This was referenced Jul 7, 2024

Typos batch: lib and include folders #75020

Closed

Fix many typos #75022

Open

Sajjon closed this Jul 7, 2024

Fix thousands of typos across hundreds of files. #74983

Fix thousands of typos across hundreds of files. #74983

Uh oh!

Conversation

Sajjon commented Jul 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

AnthonyLatsis commented Jul 5, 2024

Uh oh!

Sajjon commented Jul 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Sajjon commented Jul 5, 2024

Uh oh!

AnthonyLatsis commented Jul 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Sajjon Jul 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

rjmccall left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

beccadax Jul 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

beccadax left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Sajjon commented Jul 6, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

compnerd commented Jul 6, 2024

Uh oh!

Sajjon commented Jul 7, 2024

Uh oh!

Uh oh!

Sajjon commented Jul 5, 2024 •

edited

Loading

Sajjon commented Jul 5, 2024 •

edited

Loading

AnthonyLatsis commented Jul 5, 2024 •

edited

Loading

Sajjon Jul 5, 2024 •

edited

Loading

beccadax Jul 5, 2024 •

edited

Loading

beccadax left a comment •

edited

Loading

Sajjon commented Jul 6, 2024 •

edited

Loading