From 212f0d3c3f250817cd20ce0798b537954ba219d9 Mon Sep 17 00:00:00 2001
From: Kurt McKee <contactme@kurtmckee.org>
Date: Wed, 26 Jul 2023 19:28:58 -0500
Subject: [PATCH] Use pre-commit to sort the dictionaries

This changes the sort order of specific characters (like `"_"`)
but is cross-platform and can be fixed by pre-commit.ci.

Closes #2689
---
 .pre-commit-config.yaml                |  3 +++
 Makefile                               |  8 +-------
 codespell_lib/data/dictionary.txt      | 14 +++++++-------
 codespell_lib/tests/test_dictionary.py | 15 +++++++++++++++
 4 files changed, 26 insertions(+), 14 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 491371d902..08f82000d9 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -46,6 +46,9 @@ repos:
       - id: fix-byte-order-marker
       - id: check-case-conflict
       - id: check-toml
+      - id: file-contents-sorter
+        files: dictionary.*\.txt$
+        args: [--ignore-case]
   - repo: https://github.com/adrienverge/yamllint.git
     rev: v1.32.0
     hooks:
diff --git a/Makefile b/Makefile
index 0c239e5938..18cf94cf2b 100644
--- a/Makefile
+++ b/Makefile
@@ -14,10 +14,6 @@ codespell.1: codespell.1.include Makefile
 
 check-dictionaries:
 	@for dictionary in ${DICTIONARIES}; do \
-		if ! LC_ALL=C sort ${SORT_ARGS} -c $$dictionary; then \
-			echo "Dictionary $$dictionary not sorted. Sort with 'make sort-dictionaries'"; \
-			exit 1; \
-		fi; \
 		if grep -E -n "^\s*$$|\s$$|^\s" $$dictionary; then \
 			echo "Dictionary $$dictionary contains leading/trailing whitespace and/or blank lines.  Trim with 'make trim-dictionaries'"; \
 			exit 1; \
@@ -31,9 +27,7 @@ check-dictionaries:
 	fi
 
 sort-dictionaries:
-	@for dictionary in ${DICTIONARIES}; do \
-		LC_ALL=C sort ${SORT_ARGS} -u -o $$dictionary $$dictionary; \
-	done
+	pre-commit run --all-files file-contents-sorter
 
 trim-dictionaries:
 	@for dictionary in ${DICTIONARIES}; do \
diff --git a/codespell_lib/data/dictionary.txt b/codespell_lib/data/dictionary.txt
index f2bad8f511..a54e70d18c 100644
--- a/codespell_lib/data/dictionary.txt
+++ b/codespell_lib/data/dictionary.txt
@@ -5,6 +5,9 @@
 3rt->3rd
 3st->3rd
 4rd->4th
+__attribyte__->__attribute__
+__cpluspus->__cplusplus
+__cpusplus->__cplusplus
 a-diaerers->a-diaereses
 aaccess->access
 aaccessibility->accessibility
@@ -8230,9 +8233,9 @@ cllouds->clouds
 cloack->cloak
 cloacks->cloaks
 cloberring->clobbering
+clock_getttime->clock_gettime
 clocksourc->clocksource
 clockwíse->clockwise
-clock_getttime->clock_gettime
 clodes->closed, clothes,
 cloding->closing
 cloes->close
@@ -11247,8 +11250,8 @@ cought->caught, cough, fought,
 coul->could
 could'nt->couldn't
 could't->couldn't
-couldent->couldn't
 coulden`t->couldn't
+couldent->couldn't
 couldn->could, couldn't,
 couldn;t->couldn't
 couldnt'->couldn't
@@ -31326,6 +31329,7 @@ phsyically->physically
 phsyics->physics
 phtread->pthread
 phtreads->pthreads
+phy_interace->phy_interface
 phyiscal->physical
 phyiscally->physically
 phyiscs->physics
@@ -31352,7 +31356,6 @@ physisions->physicians
 physisist->physicist
 phython->python
 phyton->python
-phy_interace->phy_interface
 piar->pair, pier, pliers,
 piars->pairs, piers, pliers,
 piblisher->publisher
@@ -47125,8 +47128,8 @@ woudl->would
 woudn't->wouldn't
 would'nt->wouldn't
 would't->wouldn't
-wouldent->wouldn't
 woulden`t->wouldn't
+wouldent->wouldn't
 wouldn;t->wouldn't
 wouldnt'->wouldn't
 wouldnt->wouldn't
@@ -47370,9 +47373,6 @@ zukeeni->zucchini
 zuser->user
 zylophone->xylophone
 zylophones->xylophones
-__attribyte__->__attribute__
-__cpluspus->__cplusplus
-__cpusplus->__cplusplus
 évaluate->evaluate
 сontain->contain
 сontained->contained
diff --git a/codespell_lib/tests/test_dictionary.py b/codespell_lib/tests/test_dictionary.py
index 059b7f46a4..8b41ad52ab 100644
--- a/codespell_lib/tests/test_dictionary.py
+++ b/codespell_lib/tests/test_dictionary.py
@@ -77,6 +77,21 @@ def test_dictionary_formatting(
         raise AssertionError("\n" + "\n".join(errors))
 
 
+@fname_params
+def test_dictionary_sorting(
+    fname: str,
+    in_aspell: Tuple[bool, bool],
+    in_dictionary: Tuple[Iterable[str], Iterable[str]],
+) -> None:
+    previous_line = None
+    with open(fname, encoding="utf-8") as file:
+        for current_line in file:
+            current_line = current_line.strip().lower()
+            if previous_line is not None:
+                assert previous_line < current_line, f"{fname} is not sorted"
+            previous_line = current_line
+
+
 def _check_aspell(
     phrase: str,
     msg: str,