fix: add new Liquid corruption patterns for 8 translation languages (#60599)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Kevin Heis <heiskr@users.noreply.github.com>

Author: 3 people

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

@@ -66,6 +66,12 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% nota %}', '{% note %}')
     content = content.replaceAll('{%- nota %}', '{%- note %}')
     content = content.replaceAll('{%- nota -%}', '{%- note -%}')
+    // `{% otra %}` / `{%- otra %}` — "another/other" = else
+    content = content.replaceAll('{% otra %}', '{% else %}')
+    content = content.replaceAll('{%- otra %}', '{%- else %}')
+    // `{% encabezados de fila %}` — "row headers" = rowheaders
+    content = content.replaceAll('{% encabezados de fila %}', '{% rowheaders %}')
+    content = content.replaceAll('{%- encabezados de fila %}', '{%- rowheaders %}')
   }
 
   if (context.code === 'ja') {
@@ -104,6 +110,14 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% 終了コメント %}', '{% endcomment %}')
     content = content.replaceAll('{% エンドビジュアルスタジオ %}', '{% endvisualstudio %}')
     content = content.replaceAll('{% エクリプス %}', '{% eclipse %}')
+    // `{% それ以外の %}` — truncated form of "in the other case" = else
+    content = content.replaceAll('{% それ以外の %}', '{% else %}')
+    content = content.replaceAll('{%- それ以外の %}', '{%- else %}')
+    // `{% それ以外の場合 ifversion X %}` → `{% elsif X %}` (confused elsif + ifversion)
+    content = content.replace(
+      /\{% それ以外の場合 ifversion\s+(.+?)\s*%\}/g,
+      '{% elsif $1 %}',
+    )
     // `{%- "supported" %}` → `{%- when "supported" %}` (missing `when`)
     // Preserves original trim syntax (`{%-` vs `{%`)
     content = content.replace(/\{%-?\s*"(supported|not_supported|preview)"\s*%\}/g, (match) => {
@@ -220,6 +234,9 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% 数据可重用', '{% data reusables')
     content = content.replaceAll('{% 其他 %}', '{% else %}')
     content = content.replaceAll('{% 原始 %}', '{% raw %}')
+    // `{% 否则 %}` — "otherwise" = else (different Chinese word than 其他)
+    content = content.replaceAll('{% 否则 %}', '{% else %}')
+    content = content.replaceAll('{%- 否则 %}', '{%- else %}')
     // Chinese `如果` = "if": `{ 如果 X %}` → `{% if X %}`
     content = content.replace(/\{ 如果 /g, '{% if ')
     // Stray Chinese `，则为` ("then") merged with `{%` before HTML: `，则为 {%<tag>` → `<tag>`
@@ -305,13 +322,17 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% данных для повторного использования.', '{% data reusables.')
     content = content.replaceAll('{% еще %}', '{% else %}')
     content = content.replaceAll('{% ещё %}', '{% else %}')
+    // `{% иначе %}` — "otherwise" = else
+    content = content.replaceAll('{% иначе %}', '{% else %}')
+    content = content.replaceAll('{%- иначе %}', '{%- else %}')
     content = content.replaceAll('{% необработанные %}', '{% raw %}')
     content = content.replaceAll('{% необработанный %}', '{% raw %}')
     content = content.replaceAll('{% сырой %}', '{% raw %}')
     content = content.replaceAll('{% нарисовать %}', '{% endraw %}')
     content = content.replaceAll('{% эндкёрл %}', '{% endcurl %}')
     content = content.replaceAll('{% запроса %}', '{% endraw %}')
-
+    // `{% Mac %}` — capitalized mac platform tag
+    content = content.replaceAll('{% Mac %}', '{% mac %}')
     // Fix double quotes in Russian YAML files that cause parsing errors
     content = content.replace(/href=""https:\/\//g, 'href="https://')
 
@@ -386,6 +407,9 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% conseil %}', '{% tip %}')
     content = content.replaceAll('{%- conseil %}', '{%- tip %}')
     content = content.replaceAll('{%- conseil -%}', '{%- tip -%}')
+    // `{% sinon %}` / `{%- sinon %}` — French "otherwise" = else
+    content = content.replaceAll('{% sinon %}', '{% else %}')
+    content = content.replaceAll('{%- sinon %}', '{%- else %}')
     // Remove orphaned {% endif %} tags when no ifversion/elsif opener exists in the content.
     // Caused by translations where only the closing tag survived (e.g. user-api.md reusable).
     if (
@@ -416,6 +440,11 @@ export function correctTranslatedContentStrings(
     content = content.replace(/\{%-? (?:ifversion|elsif|if) [^%]*?또는[^%]*?%\}/g, (match) => {
       return match.replace(/ 또는 /g, ' or ')
     })
+    // `{% 그렇지 않으면 %}` — "otherwise" = else
+    content = content.replaceAll('{% 그렇지 않으면 %}', '{% else %}')
+    content = content.replaceAll('{%- 그렇지 않으면 %}', '{%- else %}')
+    // `{% 옥티콘` — Korean transliteration of "octicon"
+    content = content.replaceAll('{% 옥티콘 ', '{% octicon ')
 
     // Korean translation of github-glossary.md
     content = content.replaceAll('{{ 용어집.term }}', '{{ glossary.term }}')
@@ -431,6 +460,8 @@ export function correctTranslatedContentStrings(
     content = content.replaceAll('{% data wiederverwendbare.', '{% data reusables.')
     content = content.replaceAll('{% Daten wiederverwendbare.', '{% data reusables.')
     content = content.replaceAll('{% Data wiederverwendbare.', '{% data reusables.')
+    // `wiederverwendbar.` (without trailing 'e') — alternate German form
+    content = content.replaceAll('{% Daten wiederverwendbar.', '{% data reusables.')
     content = content.replaceAll('{%-Daten variables', '{%- data variables')
     content = content.replaceAll('{%-Daten-variables', '{%- data variables')
     content = content.replaceAll('{%- ifversion fpt oder ghec %}', '{%- ifversion fpt or ghec %}')
@@ -453,6 +484,12 @@ export function correctTranslatedContentStrings(
     content = content.replace(/\{%-? für (\w+) in /g, (match) => {
       return match.replace('für', 'for')
     })
+    // `{% ansonsten %}` / `{%- ansonsten %}` — "otherwise" = else
+    content = content.replaceAll('{% ansonsten %}', '{% else %}')
+    content = content.replaceAll('{%- ansonsten %}', '{%- else %}')
+    // `{% Zeilenkopfzeilen %}` — "row headers" = rowheaders
+    content = content.replaceAll('{% Zeilenkopfzeilen %}', '{% rowheaders %}')
+    content = content.replaceAll('{%- Zeilenkopfzeilen %}', '{%- rowheaders %}')
   }
 
   // --- Generic fixes (all languages) ---

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

@@ -81,6 +81,16 @@ describe('correctTranslatedContentStrings', () => {
       expect(fix('{%- nota -%}', 'es')).toBe('{%- note -%}')
     })
 
+    test('fixes otra → else', () => {
+      expect(fix('{% otra %}', 'es')).toBe('{% else %}')
+      expect(fix('{%- otra %}', 'es')).toBe('{%- else %}')
+    })
+
+    test('fixes encabezados de fila → rowheaders', () => {
+      expect(fix('{% encabezados de fila %}', 'es')).toBe('{% rowheaders %}')
+      expect(fix('{%- encabezados de fila %}', 'es')).toBe('{%- rowheaders %}')
+    })
+
     test('fixes multiple or-translations in single ifversion', () => {
       expect(fix('{% ifversion fpt o ghec o ghes %}', 'es')).toBe(
         '{% ifversion fpt or ghec or ghes %}',
@@ -216,6 +226,21 @@ describe('correctTranslatedContentStrings', () => {
       // `{ endif% %}` — percent appears after "endif" instead of after the opening brace
       expect(fix('some content\n{ endif% %}\nmore', 'ja')).toBe('some content\n{% endif %}\nmore')
     })
+
+    test('fixes truncated それ以外の → else', () => {
+      expect(fix('{% それ以外の %}', 'ja')).toBe('{% else %}')
+      expect(fix('{%- それ以外の %}', 'ja')).toBe('{%- else %}')
+    })
+
+    test('fixes それ以外の場合 ifversion X → elsif X', () => {
+      expect(fix('{% それ以外の場合 ifversion codeql-rust-public-preview %}', 'ja')).toBe(
+        '{% elsif codeql-rust-public-preview %}',
+      )
+      // no space before closing tag
+      expect(fix('{% それ以外の場合 ifversion codeql-rust-public-preview%}', 'ja')).toBe(
+        '{% elsif codeql-rust-public-preview %}',
+      )
+    })
   })
 
   // ─── PORTUGUESE (pt) ───────────────────────────────────────────────
@@ -320,6 +345,11 @@ describe('correctTranslatedContentStrings', () => {
       expect(fix('{% ifversion fpt 或 ghec %}', 'zh')).toBe('{% ifversion fpt or ghec %}')
       expect(fix('{%- elsif fpt 或 ghec %}', 'zh')).toBe('{%- elsif fpt or ghec %}')
     })
+
+    test('fixes 否则 → else', () => {
+      expect(fix('{% 否则 %}', 'zh')).toBe('{% else %}')
+      expect(fix('{%- 否则 %}', 'zh')).toBe('{%- else %}')
+    })
   })
 
   // ─── RUSSIAN (ru) ──────────────────────────────────────────────────
@@ -474,6 +504,15 @@ describe('correctTranslatedContentStrings', () => {
         fix('{% octicon "организация" aria-hidden="true" aria-label="organization" %}', 'ru'),
       ).toBe('{% octicon "organization" aria-hidden="true" aria-label="organization" %}')
     })
+
+    test('fixes иначе → else', () => {
+      expect(fix('{% иначе %}', 'ru')).toBe('{% else %}')
+      expect(fix('{%- иначе %}', 'ru')).toBe('{%- else %}')
+    })
+
+    test('fixes capitalized Mac → mac platform tag', () => {
+      expect(fix('{% Mac %}', 'ru')).toBe('{% mac %}')
+    })
   })
 
   // ─── FRENCH (fr) ───────────────────────────────────────────────────
@@ -548,6 +587,11 @@ describe('correctTranslatedContentStrings', () => {
       const input = '{% ifversion fpt %}a{% elsif ghec %}b{% endif %}'
       expect(fix(input, 'fr')).toBe(input)
     })
+
+    test('fixes sinon → else', () => {
+      expect(fix('{% sinon %}', 'fr')).toBe('{% else %}')
+      expect(fix('{%- sinon %}', 'fr')).toBe('{%- else %}')
+    })
   })
 
   // ─── KOREAN (ko) ──────────────────────────────────────────────────
@@ -592,6 +636,17 @@ describe('correctTranslatedContentStrings', () => {
     test('fixes Korean glossary template', () => {
       expect(fix('{{ 용어집.term }}', 'ko')).toBe('{{ glossary.term }}')
     })
+
+    test('fixes 그렇지 않으면 → else', () => {
+      expect(fix('{% 그렇지 않으면 %}', 'ko')).toBe('{% else %}')
+      expect(fix('{%- 그렇지 않으면 %}', 'ko')).toBe('{%- else %}')
+    })
+
+    test('fixes 옥티콘 → octicon', () => {
+      expect(fix('{% 옥티콘 "check" aria-label="Supported" %}', 'ko')).toBe(
+        '{% octicon "check" aria-label="Supported" %}',
+      )
+    })
   })
 
   // ─── GERMAN (de) ──────────────────────────────────────────────────
@@ -657,9 +712,22 @@ describe('correctTranslatedContentStrings', () => {
         fix('{% Data wiederverwendbare.audit_log.referenz-nach-kategorie-gruppiert %}', 'de'),
       ).toBe('{% data reusables.audit_log.referenz-nach-kategorie-gruppiert %}')
     })
-  })
+    test('fixes wiederverwendbar (without trailing e) reusables path', () => {
+      expect(fix('{% Daten wiederverwendbar.user-settings.access_settings %}', 'de')).toBe(
+        '{% data reusables.user-settings.access_settings %}',
+      )
+    })
 
-  // ─── GENERIC FIXES ────────────────────────────────────────────────
+    test('fixes ansonsten → else', () => {
+      expect(fix('{% ansonsten %}', 'de')).toBe('{% else %}')
+      expect(fix('{%- ansonsten %}', 'de')).toBe('{%- else %}')
+    })
+
+    test('fixes Zeilenkopfzeilen → rowheaders', () => {
+      expect(fix('{% Zeilenkopfzeilen %}', 'de')).toBe('{% rowheaders %}')
+      expect(fix('{%- Zeilenkopfzeilen %}', 'de')).toBe('{%- rowheaders %}')
+    })
+  })
 
   describe('Generic fixes (all languages)', () => {
     test('strips LLM sentinel markers and preserves word boundaries', () => {