Switch to a different algroithm.

about/scraping_line_today_home.md

@@ -106,4 +106,8 @@ Apperently, there is something called a "hybrid listing" which is s simple recom
 ```
 This is 100% easier to work with, and with a another extra, I can easily search shitty news terms. Also there is as category type??? What?
 
-Also the id can just work with the following pattern in regex: `news_cat:[a-zA-Z0-9]{24}`
+Also the id can just work with the following pattern in regex: `news_cat:[a-zA-Z0-9]{24}`, there is also `top_foryou:[a-zA-Z0-9]{24}`
+
+### Hybrid listings?
+- news_cat
+- top_foryou

bun.lock

@@ -17,6 +17,7 @@
         "@tailwindcss/vite": "^4.1.5",
         "@uploadthing/nuxt": "^7.1.7",
         "@vueuse/core": "^13.1.0",
+        "ahocorasick": "^1.0.2",
         "animate.css": "^4.1.1",
         "argon2": "^0.43.0",
         "axios": "^1.9.0",
@@ -884,6 +885,8 @@
 
     "agentkeepalive": ["agentkeepalive@4.6.0", "", { "dependencies": { "humanize-ms": "^1.2.1" } }, "sha512-kja8j7PjmncONqaTsB8fQ+wE2mSU2DJ9D4XKoJ5PFWIdRMa6SLSN1ff4mOr4jCbfRSsxR4keIiySJU0N9T5hIQ=="],
 
+    "ahocorasick": ["ahocorasick@1.0.2", "", {}, "sha512-hCOfMzbFx5IDutmWLAt6MZwOUjIfSM9G9FyVxytmE4Rs/5YDPWQrD/+IR1w+FweD9H2oOZEnv36TmkjhNURBVA=="],
+
     "ajv": ["ajv@6.12.6", "", { "dependencies": { "fast-deep-equal": "^3.1.1", "fast-json-stable-stringify": "^2.0.0", "json-schema-traverse": "^0.4.1", "uri-js": "^4.2.2" } }, "sha512-j3fVLgvTo527anyYyJOGTYJbG+vnnQYvE0m5mmkc1TK+nxAppkCLMIL0aZ4dblVCNoGShhm+kzE4ZUykBoMg4g=="],
 
     "algoliasearch": ["algoliasearch@5.25.0", "", { "dependencies": { "@algolia/client-abtesting": "5.25.0", "@algolia/client-analytics": "5.25.0", "@algolia/client-common": "5.25.0", "@algolia/client-insights": "5.25.0", "@algolia/client-personalization": "5.25.0", "@algolia/client-query-suggestions": "5.25.0", "@algolia/client-search": "5.25.0", "@algolia/ingestion": "1.25.0", "@algolia/monitoring": "1.25.0", "@algolia/recommend": "5.25.0", "@algolia/requester-browser-xhr": "5.25.0", "@algolia/requester-fetch": "5.25.0", "@algolia/requester-node-http": "5.25.0" } }, "sha512-n73BVorL4HIwKlfJKb4SEzAYkR3Buwfwbh+MYxg2mloFph2fFGV58E90QTzdbfzWrLn4HE5Czx/WTjI8fcHaMg=="],

components/checks/checkKidUnfriendlyContent.ts

@@ -1,14 +1,16 @@
-import NewsAnalyzer from "~/components/newsAnalyzer";
-const newsAnalyzer = new NewsAnalyzer();
+// Trying out the ahocorasick algorithm
+// Recommended by: https://www.threads.com/@hsinspeng/post/DJ3yVGQxBg7
+import AhoCorasick from "ahocorasick";
+
 async function checkUnsafeContent(title: string) {
   try {
     const req = await fetch("/api/contentcheck/kidunfriendlycontent");
     const res = await req.json();
-    const patterns = res.words.map((word) => new RegExp(word, "i"));
-    console.log(patterns);
-    newsAnalyzer.setSensitivePatterns(patterns);
-    const kidfriendly = newsAnalyzer.isKidFriendly(title);
-    return !kidfriendly;
+    console.log(res.words);
+    const ac = new AhoCorasick(res.words);
+    const kidfriendly = ac.search(title);
+    console.log(kidfriendly);
+    return kidfriendly;
   } catch (e) {
     console.log(e);
   }

database/kidunfriendlycontent.json

@@ -20,6 +20,7 @@
     "大露",
     "色誘",
     "死亡",
-    "撩妹"
+    "撩妹",
+    "裸上身"
   ]
 }

package.json

@@ -29,6 +29,7 @@
     "@tailwindcss/vite": "^4.1.5",
     "@uploadthing/nuxt": "^7.1.7",
     "@vueuse/core": "^13.1.0",
+    "ahocorasick": "^1.0.2",
     "animate.css": "^4.1.1",
     "argon2": "^0.43.0",
     "axios": "^1.9.0",

pages/test.vue

@@ -0,0 +1,24 @@
+<script setup lang="ts">
+import CheckKidUnfriendlyContent from "~/components/checks/checkKidUnfriendlyContent";
+const title = ref("");
+const system = ref("");
+const checkTitle = async () => {
+  if (!title.value) return;
+  system.value = await CheckKidUnfriendlyContent(title.value);
+};
+</script>
+<template>
+  <div
+    class="flex flex-col absolute h-screen w-full inset-0 justify-center align-center text-center"
+  >
+    <h1 class="text-4xl m-2">標體？</h1>
+    <div class="flex flex-row justify-center align-center gap-2">
+      <input type="text" class="text-black rounded-xl p-2 w-[300px]" />
+      <button @click="checkTitle" class="rounded-xl bg-sky-600 p-2">
+        Find
+      </button>
+    </div>
+    <div>{{ system }}</div>
+  </div>
+  <div class="h-screen"></div>
+</template>

pages/tools/checkweirdkeywords.vue

@@ -2,9 +2,10 @@
 import CheckKidUnfriendlyContent from "~/components/checks/checkKidUnfriendlyContent";
 const title = ref("");
 const system = ref(false);
+const testingReturn = ref("");
 const checkTitle = async () => {
   if (!title.value) return;
-  system.value = await CheckKidUnfriendlyContent(title.value);
+  testingReturn.value = await CheckKidUnfriendlyContent(title.value);
 };
 useSeoMeta({
   title: "這個文章是不是使用偏色情的標體？",
@@ -16,11 +17,7 @@ useSeoMeta({
   >
     <h1 class="text-4xl m-2">這個文章是不是使用偏色情的標體？</h1>
     <div class="flex flex-row justify-center align-center gap-2">
-      <input
-        type="text"
-        v-model="title"
-        class="text-black rounded-xl p-2 w-[300px]"
-      />
+      <input type="text" class="text-black rounded-xl p-2 w-[300px]" />
       <button @click="checkTitle" class="rounded-xl bg-sky-600 p-2">
         Find
       </button>
@@ -28,5 +25,6 @@ useSeoMeta({
     <span v-if="system" class="text-red-400 text-8xl m-8">是</span>
     <span v-else class="text-white">不是</span>
   </div>
+  <div>{{ testingReturn }}</div>
   <div class="h-screen"></div>
 </template>

server/api/contentcheck/kidunfriendlycontent.ts

@@ -17,6 +17,13 @@ export default defineEventHandler(async (event) => {
       "裸照",
       "性感",
       "找妹",
+      "肉蹼",
+      "超兇北半球",
+      "大露",
+      "色誘",
+      "死亡",
+      "撩妹",
+      "裸上身",
     ],
   };
 });

server/api/home/uuid_lt/action.ts

@@ -73,7 +73,23 @@ export default defineEventHandler(async (event) => {
     if (noDup.includes(key)) {
       return;
     } else {
-      noDup.push(key);
+      noDup.push({
+        type: "nuuid",
+        content: key,
+      });
+    }
+  });
+  const nonUUIDbutValidLinks = data.filter((id) =>
+    /.*:[a-zA-Z0-9]{24}/g.test(id),
+  );
+  nonUUIDbutValidLinks.forEach((key) => {
+    if (noDup.includes(key)) {
+      return;
+    } else {
+      noDup.push({
+        type: "vUUID",
+        content: key,
+      });
     }
   });
   return {