fix: duplicate-URL tooltip now says the path is lowercased too

apps/desktop/src/renderer/src/components/SettingsDialog.tsx

@@ -3826,7 +3826,7 @@ function DuplicatesPanel({ form, update }: PanelProps) {
           label={t('settingsPanels.duplicates.preNormalize', { defaultValue: 'Normalise URLs before duplicate-URL check' })}
           checked={form.dedupePreNormalize}
           onChange={(v) => update('dedupePreNormalize', v)}
-          info="When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time)."
+          info="When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time)."
           example="ON for SEO audits. OFF only when you specifically need to inspect raw-URL collisions (e.g. case-sensitive filesystem CMSes)."
         />
       </div>

apps/desktop/src/renderer/src/i18n/info-tips/az.ts

@@ -484,8 +484,8 @@ export const AZ_INFO_TIPS: Record<string, string> = {
     'Söndürülü olduqda büdcə qiymətləndirməsi işləmir və qərar sütunu təmizlənir. Aktiv olduqda tarama sonrası mərhələ hər daxili 200 HTML səhifəni aşağıdakı tavanlara görə qiymətləndirir.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Aktiv olduqda (standart) pagination_next + pagination_prev URL-ləri yükləmədən sonra növbəyə qoyulur. Yalnız bütün keçid izləməni söndürmədən yalnız-səhifələmə dövrlərini sazlamaq üçün söndürün.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    'AKTİV olduqda (standart) Təkrar URL filtri URL-ləri hostu kiçik hərfə çevirib, sorğu sətrini atıb və sonluq slaşını kəsib müqayisə edir — kanonik SEO davranışı. SÖNDÜRÜLÜ olduqda müqayisə bayt-dəqiqdir, ona görə filtr yalnız eyni xam URL sətrini paylaşan sətirlərdə işləyir (nadirdir, çünki URL-lər daxil edilərkən təkrarsızlaşdırılır).',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    'AKTİV olduqda (standart) Təkrar URL filtri bütün URL-i (host və yolu) kiçik hərfə çevirib, sorğu sətrini atıb və sonluq slaşını kəsib müqayisə edir — /Page və /page kimi yalnız hərf registri ilə fərqlənən ünvanlar təkrar sayılır. SÖNDÜRÜLÜ olduqda müqayisə bayt-dəqiqdir, ona görə filtr yalnız eyni xam URL sətrini paylaşan sətirlərdə işləyir (nadirdir, çünki URL-lər daxil edilərkən təkrarsızlaşdırılır).',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Aktiv olduqda `<meta http-equiv='refresh'>` content URL-ləri yönləndirmə hədəfi kimi növbəyə qoyulur. window.location gövdə yönləndirmələri yalnız evristikdir və hazırda əhatə xaricindədir.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/es.ts

@@ -487,8 +487,8 @@ export const ES_INFO_TIPS: Record<string, string> = {
     'Cuando está desactivado, no se evalúa ningún presupuesto y la columna de veredicto se vacía. Cuando está activado, la pasada posterior al rastreo puntúa cada página HTML 200 interna contra los techos de abajo.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Cuando está activado (por defecto), las URL pagination_next + pagination_prev se encolan tras la obtención. Desactívalo solo para depurar bucles exclusivos de paginación sin desactivar todo el seguimiento de enlaces.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    'Cuando está ACTIVADO (por defecto), el filtro URL duplicada compara las URL tras pasar el host a minúsculas, eliminar la cadena de consulta y recortar la barra final — el comportamiento SEO canónico. Cuando está DESACTIVADO, la comparación es byte a byte, así que el filtro solo salta en filas que comparten una cadena de URL idéntica sin procesar (raro, ya que las URL se deduplican al insertarse).',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    'Cuando está ACTIVADO (por defecto), el filtro URL duplicada compara las URL tras pasar la URL completa (host y ruta) a minúsculas, eliminar la cadena de consulta y recortar la barra final — variantes que solo difieren en mayúsculas, como /Page y /page, cuentan como duplicadas. Cuando está DESACTIVADO, la comparación es byte a byte, así que el filtro solo salta en filas que comparten una cadena de URL idéntica sin procesar (raro, ya que las URL se deduplican al insertarse).',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Cuando está activado, las URL del content de `<meta http-equiv='refresh'>` se encolan como un destino de redirección. Las redirecciones window.location del cuerpo son solo heurísticas y actualmente están fuera del alcance.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/fr.ts

@@ -487,8 +487,8 @@ export const FR_INFO_TIPS: Record<string, string> = {
     "Désactivé, aucune évaluation de budget n'a lieu et la colonne de verdict est vidée. Activé, la passe post-exploration note chaque page HTML 200 interne par rapport aux plafonds ci-dessous.",
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Activé (par défaut), les URL pagination_next + pagination_prev sont mises en file après récupération. Désactivez uniquement pour déboguer des boucles de pagination sans désactiver tout le suivi de liens.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    "ACTIVÉ (par défaut), le filtre URL dupliquée compare les URL après mise en minuscules de l'hôte, suppression de la chaîne de requête et retrait de la barre finale — le comportement SEO canonique. DÉSACTIVÉ, la comparaison est octet pour octet, donc le filtre ne se déclenche que sur des lignes partageant une chaîne d'URL brute identique (rare puisque les URL sont dédupliquées à l'insertion).",
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    "ACTIVÉ (par défaut), le filtre URL dupliquée compare les URL après mise en minuscules de l'URL entière (hôte et chemin), suppression de la chaîne de requête et retrait de la barre finale — des variantes de casse comme /Page et /page comptent comme doublons. DÉSACTIVÉ, la comparaison est octet pour octet, donc le filtre ne se déclenche que sur des lignes partageant une chaîne d'URL brute identique (rare puisque les URL sont dédupliquées à l'insertion).",
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Activé, les URL du content de `<meta http-equiv='refresh'>` sont mises en file comme une cible de redirection. Les redirections window.location dans le corps sont purement heuristiques et actuellement hors périmètre.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/hi.ts

@@ -482,8 +482,8 @@ export const HI_INFO_TIPS: Record<string, string> = {
     'बंद होने पर कोई बजट मूल्यांकन नहीं चलता और निर्णय कॉलम साफ़ हो जाता है। चालू होने पर क्रॉल के बाद का चरण हर आंतरिक 200 HTML पेज को नीचे की सीमाओं के विरुद्ध स्कोर करता है।',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'चालू (डिफ़ॉल्ट) होने पर pagination_next + pagination_prev URL फ़ेच के बाद कतार में डाले जाते हैं। पूरा लिंक फ़ॉलो बंद किए बिना केवल-पेजिनेशन लूप डीबग करने के लिए ही बंद करें।',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    'चालू (डिफ़ॉल्ट) होने पर डुप्लिकेट URL फ़िल्टर होस्ट लोअरकेस करके, क्वेरी स्ट्रिंग हटाकर और अंतिम स्लैश काटकर URL की तुलना करता है — कैनोनिकल SEO व्यवहार। बंद होने पर तुलना बाइट-सटीक है, इसलिए फ़िल्टर केवल उन पंक्तियों पर चलता है जिनकी कच्ची URL स्ट्रिंग समान हो (दुर्लभ, क्योंकि URL इन्सर्ट के समय डीडुप्लिकेट होते हैं)।',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    'चालू (डिफ़ॉल्ट) होने पर डुप्लिकेट URL फ़िल्टर पूरा URL (होस्ट और पथ) लोअरकेस करके, क्वेरी स्ट्रिंग हटाकर और अंतिम स्लैश काटकर URL की तुलना करता है — /Page और /page जैसे केवल केस में भिन्न URL भी डुप्लिकेट गिने जाते हैं। बंद होने पर तुलना बाइट-सटीक है, इसलिए फ़िल्टर केवल उन पंक्तियों पर चलता है जिनकी कच्ची URL स्ट्रिंग समान हो (दुर्लभ, क्योंकि URL इन्सर्ट के समय डीडुप्लिकेट होते हैं)।',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "चालू होने पर `<meta http-equiv='refresh'>` content URL रीडायरेक्ट लक्ष्य की तरह कतार में डाले जाते हैं। window.location बॉडी रीडायरेक्ट केवल ह्यूरिस्टिक हैं और फ़िलहाल दायरे से बाहर।",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/it.ts

@@ -487,8 +487,8 @@ export const IT_INFO_TIPS: Record<string, string> = {
     'Se disattivato, non viene eseguita alcuna valutazione del budget e la colonna del verdetto viene svuotata. Se attivato, la passata post-scansione valuta ogni pagina HTML 200 interna rispetto ai tetti sottostanti.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Se attivato (predefinito), gli URL pagination_next + pagination_prev vengono accodati dopo il recupero. Disattiva solo per fare debug di cicli di sola paginazione senza disabilitare tutto il follow dei link.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    "Se ATTIVATO (predefinito), il filtro URL duplicato confronta gli URL dopo aver reso minuscolo l'host, rimosso la query string e tolto la barra finale — il comportamento SEO canonico. Se DISATTIVATO, il confronto è byte per byte, quindi il filtro scatta solo su righe che condividono una stringa URL grezza identica (raro, perché gli URL vengono deduplicati all'inserimento).",
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    "Se ATTIVATO (predefinito), il filtro URL duplicato confronta gli URL dopo aver reso minuscolo l'intero URL (host e percorso), rimosso la query string e tolto la barra finale — varianti che differiscono solo per maiuscole, come /Page e /page, contano come duplicati. Se DISATTIVATO, il confronto è byte per byte, quindi il filtro scatta solo su righe che condividono una stringa URL grezza identica (raro, perché gli URL vengono deduplicati all'inserimento).",
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Se attivato, gli URL nel content di `<meta http-equiv='refresh'>` vengono accodati come una destinazione di reindirizzamento. I reindirizzamenti window.location nel corpo sono solo euristici e attualmente fuori ambito.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/ko.ts

@@ -477,8 +477,8 @@ export const KO_INFO_TIPS: Record<string, string> = {
     '끄면 예산 평가가 실행되지 않고 판정 열이 비워집니다. 켜면 크롤링 후 단계가 모든 내부 200 HTML 페이지를 아래 상한에 대해 채점합니다.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     '켜면(기본값) pagination_next + pagination_prev URL이 가져온 뒤 큐에 들어갑니다. 모든 링크 추적을 끄지 않고 페이지네이션 전용 루프만 디버깅할 때만 끄세요.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    '켜면(기본값) 중복 URL 필터가 호스트를 소문자로 바꾸고 쿼리 문자열을 버리고 후행 슬래시를 제거한 뒤 URL을 비교합니다 — 표준적인 SEO 동작입니다. 끄면 바이트 단위로 정확히 비교하므로 동일한 원시 URL 문자열을 공유하는 행에서만 필터가 발동합니다(URL은 삽입 시 중복 제거되므로 드뭅니다).',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    '켜면(기본값) 중복 URL 필터가 URL 전체(호스트와 경로)를 소문자로 바꾸고 쿼리 문자열을 버리고 후행 슬래시를 제거한 뒤 URL을 비교합니다 — /Page와 /page처럼 대소문자만 다른 주소도 중복으로 집계됩니다. 끄면 바이트 단위로 정확히 비교하므로 동일한 원시 URL 문자열을 공유하는 행에서만 필터가 발동합니다(URL은 삽입 시 중복 제거되므로 드뭅니다).',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "켜면 `<meta http-equiv='refresh'>` content의 URL이 리디렉션 대상처럼 큐에 들어갑니다. 본문의 window.location 리디렉션은 휴리스틱에 불과하며 현재 범위 밖입니다.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/pt-BR.ts

@@ -485,8 +485,8 @@ export const PT_BR_INFO_TIPS: Record<string, string> = {
     'Quando desligado, nenhuma avaliação de orçamento roda e a coluna de veredito é limpa. Quando ligado, a etapa pós-rastreamento pontua cada página HTML 200 interna contra os tetos abaixo.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Quando ligado (padrão), as URLs pagination_next + pagination_prev são enfileiradas após a busca. Desligue apenas para depurar loops exclusivos de paginação sem desativar todo o seguimento de links.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    'Quando LIGADO (padrão), o filtro URL duplicada compara as URLs após converter o host para minúsculas, remover a query string e cortar a barra final — o comportamento SEO canônico. Quando DESLIGADO, a comparação é byte a byte, então o filtro só dispara em linhas que compartilham uma string de URL bruta idêntica (raro, já que as URLs são desduplicadas na inserção).',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    'Quando LIGADO (padrão), o filtro URL duplicada compara as URLs após converter a URL inteira (host e caminho) para minúsculas, remover a query string e cortar a barra final — variantes que só diferem em maiúsculas, como /Page e /page, contam como duplicadas. Quando DESLIGADO, a comparação é byte a byte, então o filtro só dispara em linhas que compartilham uma string de URL bruta idêntica (raro, já que as URLs são desduplicadas na inserção).',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Quando ligado, as URLs do content de `<meta http-equiv='refresh'>` são enfileiradas como destino de redirecionamento. Redirecionamentos window.location no corpo são apenas heurísticos e atualmente fora do escopo.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/ru.ts

@@ -486,8 +486,8 @@ export const RU_INFO_TIPS: Record<string, string> = {
     'Если выключено, оценка бюджета не выполняется, а колонка вердикта очищается. Если включено, проход после сканирования оценивает каждую внутреннюю HTML-страницу с кодом 200 по порогам ниже.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     'Если включено (по умолчанию), URL из pagination_next + pagination_prev ставятся в очередь после загрузки. Выключайте только для отладки петель пагинации, не отключая переходы по ссылкам целиком.',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    'Если ВКЛ. (по умолчанию), фильтр «Дубликат URL» сравнивает URL после приведения хоста к нижнему регистру, отбрасывания строки запроса и удаления завершающего слэша — каноническое SEO-поведение. Если ВЫКЛ., сравнение побайтовое, поэтому фильтр срабатывает только на строках с идентичной сырой строкой URL (редко, так как URL дедуплицируются при вставке).',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    'Если ВКЛ. (по умолчанию), фильтр «Дубликат URL» сравнивает URL после приведения всего URL (хоста и пути) к нижнему регистру, отбрасывания строки запроса и удаления завершающего слэша — адреса, отличающиеся только регистром, как /Page и /page, считаются дубликатами. Если ВЫКЛ., сравнение побайтовое, поэтому фильтр срабатывает только на строках с идентичной сырой строкой URL (редко, так как URL дедуплицируются при вставке).',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Если включено, URL из content `<meta http-equiv='refresh'>` ставятся в очередь как цель перенаправления. Перенаправления window.location в теле — только эвристика и сейчас вне области.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/tr.ts

@@ -486,8 +486,8 @@ export const TR_INFO_TIPS: Record<string, string> = {
     'Kapalıyken bütçe değerlendirmesi çalışmaz ve sonuç kolonu temizlenir. Açıkken crawl sonrası geçiş, 200 dönen her iç HTML sayfasını aşağıdaki tavanlara göre puanlar.',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     "Açıkken (varsayılan), pagination_next ve pagination_prev URL'leri çekim sonrası kuyruğa alınır. Yalnızca tüm bağlantı takibini kapatmadan sayfalama döngülerini incelemek için kapatın.",
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    "AÇIK olduğunda (varsayılan), Yinelenen URL filtresi URL'leri host küçük harfe çevrilmiş, sorgu dizesi atılmış ve sondaki eğik çizgi kırpılmış halde karşılaştırır — standart SEO davranışı. KAPALI olduğunda karşılaştırma bayt bazındadır, bu yüzden filtre yalnızca birebir aynı ham URL dizesini paylaşan satırlarda tetiklenir (URL'ler ekleme sırasında zaten tekilleştirildiği için nadirdir).",
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    "AÇIK olduğunda (varsayılan), Yinelenen URL filtresi URL'leri tamamı (host ve yol) küçük harfe çevrilmiş, sorgu dizesi atılmış ve sondaki eğik çizgi kırpılmış halde karşılaştırır — /Page ve /page gibi yalnızca büyük/küçük harfte ayrılan adresler yinelenen sayılır. KAPALI olduğunda karşılaştırma bayt bazındadır, bu yüzden filtre yalnızca birebir aynı ham URL dizesini paylaşan satırlarda tetiklenir (URL'ler ekleme sırasında zaten tekilleştirildiği için nadirdir).",
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "Açıkken `<meta http-equiv='refresh'>` içeriğindeki URL'ler bir yönlendirme hedefi gibi kuyruğa alınır. window.location ile yapılan gövde yönlendirmeleri yalnızca sezgiseldir ve şu an kapsam dışıdır.",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']:

apps/desktop/src/renderer/src/i18n/info-tips/zh-CN.ts

@@ -473,8 +473,8 @@ export const ZH_CN_INFO_TIPS: Record<string, string> = {
     '关闭时不进行预算评估，判定列会被清空。开启时抓取后处理会按下方上限为每个内部 200 HTML 页面评分。',
   ['When on (default), pagination_next + pagination_prev URLs are post-fetch enqueued. Off only to debug pagination-only loops without disabling all link follow.']:
     '开启（默认）时，pagination_next + pagination_prev URL 在抓取后入队。仅在需要调试纯分页循环而不禁用所有链接跟踪时关闭。',
-  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the host, dropping the query string, and trimming the trailing slash — the canonical SEO behaviour. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
-    '开启（默认）时，“重复 URL”筛选器在主机转小写、去掉查询字符串、去除尾部斜杠后比较 URL——这是规范的 SEO 行为。关闭时按字节精确比较，因此筛选器只对原始 URL 字符串完全相同的行生效（因为 URL 在插入时已去重，所以很少见）。',
+  ['When ON (default), the Duplicate URL filter compares URLs after lowercasing the whole URL (host and path), dropping the query string, and trimming the trailing slash — case variants such as /Page and /page count as duplicates. When OFF, comparison is byte-exact, so the filter only fires on rows that share an identical raw URL string (rare since URLs are deduped at insert time).']:
+    '开启（默认）时，“重复 URL”筛选器在整个 URL（主机和路径）转小写、去掉查询字符串、去除尾部斜杠后比较 URL——/Page 与 /page 这类仅大小写不同的地址会算作重复。关闭时按字节精确比较，因此筛选器只对原始 URL 字符串完全相同的行生效（因为 URL 在插入时已去重，所以很少见）。',
   ["When on, `<meta http-equiv='refresh'>` content URLs are enqueued like a redirect target. window.location body redirects are heuristic-only and currently out of scope."]:
     "开启时，`<meta http-equiv='refresh'>` content 中的 URL 会像重定向目标一样入队。正文中的 window.location 重定向仅为启发式判断，目前不在范围内。",
   ['When on, a 200 page declaring a canonical pointing elsewhere also enqueues that target, so a canonical pointing at a missing or redirected page is caught. Default on.']: