Add translation corrections for es, zh, and ko Liquid tags (#63831)

Author: V-halfaro

src/languages/lib/correct-translation-content.ts

@@ -94,6 +94,13 @@ export function correctTranslatedContentStrings(
   if (context.code === 'es') {
     content = content.replace(/\{%:/g, '{%')
 
+    // The translator replaced the space after "data" with a dot, as in {% datos.variables.x %},
+    // or inserted a stray word, as in {% datos existente variables.x %}.
+    content = content.replace(
+      /\{%(-?)\s*datos(?:\.|\s+existente\s+)(variables|reusables)\./g,
+      '{%$1 data $2.',
+    )
+
     content = content.replaceAll('{% advertencia %}', '{% warning %}')
     content = content.replaceAll('{% siVersion ', '{% ifversion ')
     content = content.replaceAll('{%- siVersion ', '{%- ifversion ')
@@ -671,6 +678,9 @@ export function correctTranslatedContentStrings(
   }
 
   if (context.code === 'zh') {
+    // Chinese "变量" ("variables") glued to "data" with no space.
+    // Also the space-separated form, as in {% data变量 x.y %}, which loses the dot.
+    content = content.replace(/\{%(-?)\s*data变量(?:\.|\s+)/g, '{%$1 data variables.')
     content = content.replaceAll('{% 数据variables', '{% data variables')
     content = content.replaceAll('{% 数据 variables', '{% data variables')
     content = content.replaceAll('{%数据variables', '{% data variables')
@@ -1247,6 +1257,13 @@ export function correctTranslatedContentStrings(
   }
 
   if (context.code === 'ko') {
+    // Misspelled "data" keyword and translated IDE/app tag names
+    content = content.replaceAll('{% dada variables', '{% data variables')
+    content = content.replaceAll('{%- dada variables', '{%- data variables')
+    content = content.replaceAll('{% 이클립스 %}', '{% eclipse %}')
+    content = content.replaceAll('{%- 이클립스 %}', '{%- eclipse %}')
+    content = content.replaceAll('{% 데스크톱 %}', '{% desktop %}')
+    content = content.replaceAll('{%- 데스크톱 %}', '{%- desktop %}')
     content = content.replaceAll('{% 경고 %}', '{% warning %}')
     content = content.replaceAll('{% datda variables', '{% data variables')
     content = content.replaceAll('{%- datda variables', '{%- data variables')
@@ -1611,6 +1628,9 @@ export function correctTranslatedContentStrings(
     return match.replace(/[«»“”„]/g, '"').replace(/[‘’‚]/g, "'")
   })
 
+  // The translator dropped the space between octicon and its quoted name.
+  content = content.replace(/\{%(-?)\s*octicon(?=")/g, '{%$1 octicon ')
+
   // French « check » becomes " check " after quote normalization.
   content = content.replace(
     /\{%(-?)\s*octicon\s+"\s*([\w-]+)\s*"([^%]*?)(-?)%\}/g,

src/languages/tests/correct-translation-content.ts

@@ -3405,4 +3405,76 @@ Para más información, consulta "[AUTOTITLE](/path)".
       expect(fix(blank, 'fr')).toBe(blank)
     })
   })
+
+  describe('data keyword glued to variables or reusables', () => {
+    test('es: {% datos.variables. and a stray word between datos and variables', () => {
+      expect(fix('{% datos.variables.product.company_short %}', 'es')).toBe(
+        '{% data variables.product.company_short %}',
+      )
+      expect(fix('un {% datos existente variables.copilot.x %} de su', 'es')).toBe(
+        'un {% data variables.copilot.x %} de su',
+      )
+      const ok = '{% data variables.product.company_short %}'
+      expect(fix(ok, 'es')).toBe(ok)
+    })
+
+    test('zh: {% data变量. becomes {% data variables.', () => {
+      expect(fix('{% data变量.dependabot.github_presets %} 是规则', 'zh')).toBe(
+        '{% data variables.dependabot.github_presets %} 是规则',
+      )
+      const ok = '{% data variables.dependabot.github_presets %}'
+      expect(fix(ok, 'zh')).toBe(ok)
+    })
+
+    test('zh: space-separated data变量 restores the dot', () => {
+      expect(fix('{% data变量 dependabot.github_presets %} 是规则', 'zh')).toBe(
+        '{% data variables.dependabot.github_presets %} 是规则',
+      )
+      expect(
+        fix(
+          '{% data变量 dependabot.github_presets %} 是规则',
+          'zh',
+          '{% data variables.dependabot.github_presets %} are rules',
+        ),
+      ).toBe('{% data variables.dependabot.github_presets %} 是规则')
+    })
+
+    test('zh: space-separated data变量 with only whitespace-control data tags', () => {
+      expect(
+        fix(
+          '{%- data变量 dependabot.github_presets %} 是规则',
+          'zh',
+          '{%- data variables.dependabot.github_presets %} are rules',
+        ),
+      ).toBe('{%- data variables.dependabot.github_presets %} 是规则')
+    })
+  })
+
+  describe('octicon missing space before quoted name', () => {
+    test('restores the space, including after curly-quote normalization', () => {
+      expect(fix('{% octicon“kebab-horizontal” aria-label=“图标” %}', 'zh')).toBe(
+        '{% octicon "kebab-horizontal" aria-label="图标" %}',
+      )
+      expect(fix('{% octicon"pulse" aria-hidden="true" %}', 'ja')).toBe(
+        '{% octicon "pulse" aria-hidden="true" %}',
+      )
+    })
+
+    test('leaves a correct octicon unchanged', () => {
+      const ok = '{% octicon "pulse" aria-hidden="true" %}'
+      expect(fix(ok, 'ja')).toBe(ok)
+    })
+  })
+
+  describe('Korean data keyword and tag names', () => {
+    test('fixes {%- dada and translated eclipse and desktop tags', () => {
+      expect(fix('{%- dada variables.product.prodname_actions %}', 'ko')).toBe(
+        '{%- data variables.product.prodname_actions %}',
+      )
+      expect(fix('{% 이클립스 %}x{% endeclipse %}', 'ko')).toBe('{% eclipse %}x{% endeclipse %}')
+      expect(fix('{%- 데스크톱 %}x{% enddesktop %}', 'ko')).toBe('{%- desktop %}x{% enddesktop %}')
+      const ok = '{% data variables.product.prodname_actions %}'
+      expect(fix(ok, 'ko')).toBe(ok)
+    })
+  })
 })