From e7607d0d3d0509af594bff4c57b7c98bf0c5f51a Mon Sep 17 00:00:00 2001
From: Artem Sapegin <artem@sapegin.ru>
Date: Fri, 28 Jun 2024 07:29:49 +0200
Subject: [PATCH] fix: Ignore file extensions by default

Thus not require custom look behinds in the terms file.

Another approach to https://github.com/jargonLint/jargonLint/issues/24
---
 index.js    | 22 +++++++++------
 terms.jsonc | 14 +++++-----
 test.js     | 80 ++++++++++++++++++-----------------------------------
 3 files changed, 48 insertions(+), 68 deletions(-)

diff --git a/index.js b/index.js
index 76ea1dd..eec7219 100644
--- a/index.js
+++ b/index.js
@@ -12,6 +12,7 @@ const DEFAULT_OPTIONS = {
 	exclude: [],
 };
 const sentenceStartRegExp = /\w+[.?!]\)? $/;
+const punctuation = '[\\.,;:!?\'"’”)]';
 
 function reporter(context, opts = {}) {
 	const options = { ...DEFAULT_OPTIONS, ...opts };
@@ -54,7 +55,7 @@ function reporter(context, opts = {}) {
 
 						let replacement = getReplacement(pattern, replacements, matched);
 
-						// Capitalize word in the beginning of a sentense if the original word was capitalized
+						// Capitalize word in the beginning of a sentence if the original word was capitalized
 						const textBeforeMatch = text.substring(0, index);
 						const isSentenceStart =
 							index === 0 || sentenceStartRegExp.test(textBeforeMatch);
@@ -90,7 +91,8 @@ function getTerms(defaultTerms, terms, exclude) {
 		? loadJson(path.resolve(__dirname, 'terms.jsonc'))
 		: [];
 	const extras = typeof terms === 'string' ? loadJson(terms) : terms;
-	// Order matters, the first term to match is used. We prioritize user 'extras' before defaults
+	// Order matters, the first term to match is used. We prioritize user
+	// 'extras' before defaults
 	const listTerms = [...(Array.isArray(extras) ? extras : []), ...defaults];
 
 	// Filter on all terms
@@ -133,12 +135,16 @@ function readTermsFile(filepath) {
  * @param {string} pattern
  */
 function getExactMatchRegExp(pattern) {
-	const punctuation = '[\\.,;:!?\'"’”)]';
 	return new RegExp(
-		// 1. Beginning of the string, or any character that isn't "-" or alphanumeric
-		// 2. Exact match of the pattern
-		// 3. Space, punctuation + space, punctuation + punctuation, or punctuation at the end of the string, end of the string
-		`(?<=^|[^-\\w])\\b${pattern}\\b(?= |${punctuation} |${punctuation}${punctuation}|${punctuation}$|$)`,
+		// 1. Beginning of the string, or any character that isn't "-"
+		//    or alphanumeric
+		// 2. Not a dot "." (to make it ignore file extensions)
+		// 3. Word boundary
+		// 4. Exact match of the pattern
+		// 5. Word boundary
+		// 6. Space, punctuation + space, punctuation + punctuation,
+		//    or punctuation at the end of the string, end of the string
+		`(?<=^|[^-\\w])(?<!\\.)\\b${pattern}\\b(?= |${punctuation} |${punctuation}${punctuation}|${punctuation}$|$)`,
 		'igm'
 	);
 }
@@ -153,7 +159,7 @@ function getMultipleWordRegExp(words) {
 
 /**
  * Match pattern on word boundaries in the middle of the text unless the pattern
- * has look behinds
+ * has look behinds or look aheads
  * @param {string} pattern
  */
 function getAdvancedRegExp(pattern) {
diff --git a/terms.jsonc b/terms.jsonc
index 3e82cba..af091c5 100644
--- a/terms.jsonc
+++ b/terms.jsonc
@@ -112,12 +112,12 @@
   ["API['’]?s", "APIs"],
 	"CDN",
 	"CORS",
-  ["(?<![\\.-])css\\b", "CSS"],
+  "CSS",
 	"DNS",
 	"DOM",
-  ["(?<![\\.-])gif\\b", "GIF"],
+  "GIF",
   "GUI",
-  ["(?<![\\.-])html\\b", "HTML"],
+  "HTML",
   ["HTTP[ /]2(?:\\.0)?", "HTTP/2"],
   ["\bHTTP(?!:)", "HTTP"],
   ["\bHTTPS(?!:)", "HTTPS"],
@@ -125,12 +125,12 @@
   "IoT",
   "I/O",
   ["I-O", "I/O"],
-  ["(?<![\\.-])jpeg\\b", "JPEG"],
+  "JPEG",
   "MIME",
   "OK",
   "PaaS",
-  ["(?<![\\.-])pdf\\b", "PDF"],
-  ["(?<![\\.-])png\\b", "PNG"],
+  "PDF",
+  "PNG",
   "SaaS",
   "URI",
   "URIs",
@@ -144,7 +144,7 @@
   ["wi[- ]?fi", "Wi-Fi"],
   "WYSIWYG",
   "XML",
-  ["(?<![\\.-])yaml\\b", "YAML"],
+  "YAML",
   "ZIP",
 
   // Names
diff --git a/test.js b/test.js
index 880d54f..e53c444 100644
--- a/test.js
+++ b/test.js
@@ -243,61 +243,35 @@ describe('getReplacement', () => {
 
 tester.run('textlint-rule-terminology', rule, {
 	valid: [
-		{
-			text: 'My JavaScript is good',
-		},
-		{
-			// Should skip code examples
-			text: 'My `javascript` is good',
-		},
-		{
-			// Should skip URLs
-			text: 'My [code](http://example.com/javascript) is good',
-		},
-		{
-			// Should keep a capital letter at the beginning of a sentense
-			text: 'Webpack is good',
-		},
+		{ text: 'My JavaScript is good' },
+		// Should skip code examples
+		{ text: 'My `javascript` is good' },
+		// Should skip URLs
+		{ text: 'My [code](http://example.com/javascript) is good' },
+		// Should keep a capital letter at the beginning of a sentense
+		{ text: 'Webpack is good' },
 		// Should not warn when incorrect term is used as a part of another word
-		{
-			text: 'Your javascriptish code',
-		},
-		{
-			text: 'javascriptish',
-		},
-		{
-			text: 'Your uberjavascript code',
-		},
-		{
-			text: 'uberjavascript',
-		},
+		{ text: 'Your javascriptish code' },
+		{ text: 'javascriptish' },
+		{ text: 'Your uberjavascript code' },
+		{ text: 'uberjavascript' },
 		// Should not warn when incorrect term is used as a part of a hyphenates word
-		{
-			text: 'Install javascript-some-plugin here',
-		},
-		{
-			text: 'javascript-some-plugin',
-		},
-		{
-			text: 'Install some-plugin-javascript here',
-		},
-		{
-			text: 'some-plugin-javascript',
-		},
-		{
-			// Should not warn about file names
-			text: 'Filetype.md',
-		},
-		{
-			text: 'I think Internet Explorer 6 is the best browser!',
-		},
-		{
-			// Should ignore `http` in the middle of a word
-			text: 'We should all use XMLHttpRequest everywhere',
-		},
-		{
-			text: 'foo.yaml',
-		},
+		{ text: 'Install javascript-some-plugin here' },
+		{ text: 'javascript-some-plugin' },
+		{ text: 'Install some-plugin-javascript here' },
+		{ text: 'some-plugin-javascript' },
+		{ text: 'I think Internet Explorer 6 is the best browser!' },
+		// Should ignore `http` in the middle of a word
+		{ text: 'We should all use XMLHttpRequest everywhere' },
+		// Filenames with .yaml extensions are valid
+		{ text: 'foo.yaml' },
+		// Filenames with `yaml` inside are valid
+		{ text: 'foo-yaml.txt' },
+		{ text: 'foo_yaml.txt' },
+		{ text: 'foo-yaml-bar.txt' },
+		{ text: 'foo_yaml_bar.txt' },
+		{ text: 'fooyaml.txt' },
+		{ text: 'fooyaml.txt' },
 	],
 	invalid: [
 		{