fix: naverPlace 오매핑 수정 + naverBlog RSS 공식 블로그 스크래핑

- naverPlace: 매칭 우선순위 재정의 — 전체 클리닉명 정확 매칭 > 짧은명+성형 카테고리 > 성형 카테고리 순으로 변경 (기존 로직은 피부과까지 매칭되어 데이뷰의원 오매핑 발생)
- naverBlog: Firecrawl 공식 블로그 스크래핑 → Naver RSS 피드로 교체 (blog.naver.com은 Firecrawl 차단, RSS는 공개 엔드포인트)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

supabase/functions/collect-channel-data/index.ts

@@ -473,41 +473,39 @@ Deno.serve(async (req) => {
         if (!res.ok) throw new Error(`Naver Blog API returned ${res.status}`);
         const data = await res.json();
 
-        // ─── 5b. Firecrawl: Official blog recent posts ───
+        // ─── 5b. Naver RSS: Official blog recent posts ───
-        // Registry always provides the official blog URL — scrape it for real content metrics.
+        // blog.naver.com is blocked by Firecrawl. Use the public RSS feed instead:
+        // https://rss.blog.naver.com/{blogId}.xml — no auth required.
         let officialBlogContent: Record<string, unknown> | null = null;
-        if (officialBlogUrl) {
+        if (officialBlogHandle) {
-          const FIRECRAWL_KEY = Deno.env.get("FIRECRAWL_API_KEY");
+          try {
-          if (FIRECRAWL_KEY) {
+            const rssRes = await fetchWithRetry(
-            try {
+              `https://rss.blog.naver.com/${officialBlogHandle}.xml`,
-              const blogScrape = await fetchWithRetry(`https://api.firecrawl.dev/v1/scrape`, {
+              {},
-                method: "POST",
+              { label: "naver-rss", timeoutMs: 15000 }
-                headers: { "Content-Type": "application/json", Authorization: `Bearer ${FIRECRAWL_KEY}` },
+            );
-                body: JSON.stringify({
+            if (rssRes.ok) {
-                  url: officialBlogUrl,
+              const xml = await rssRes.text();
-                  formats: ["json"],
+              // Parse RSS items: <item><title>...</title><link>...</link><pubDate>...</pubDate><description>...</description></item>
-                  jsonOptions: {
+              const items: Array<{ title: string; link: string; date: string; excerpt: string }> = [];
-                    prompt: "Extract the blog's recent posts: title, date, excerpt. Also total post count visible on the page, and the blog category/tag list.",
+              const itemMatches = xml.matchAll(/<item>([\s\S]*?)<\/item>/g);
-                    schema: {
+              for (const m of itemMatches) {
-                      type: "object",
+                const block = m[1];
-                      properties: {
+                const title = (block.match(/<title><!\[CDATA\[(.*?)\]\]><\/title>/) || block.match(/<title>(.*?)<\/title>/))?.[1] || "";
-                        totalPosts: { type: "number" },
+                const link = (block.match(/<link>(.*?)<\/link>/))?.[1] || "";
-                        recentPosts: { type: "array", items: { type: "object", properties: { title: { type: "string" }, date: { type: "string" }, excerpt: { type: "string" } } } },
+                const date = (block.match(/<pubDate>(.*?)<\/pubDate>/))?.[1] || "";
-                        categories: { type: "array", items: { type: "string" } },
+                const desc = (block.match(/<description><!\[CDATA\[(.*?)\]\]><\/description>/) || block.match(/<description>(.*?)<\/description>/))?.[1] || "";
-                      },
+                items.push({ title, link, date, excerpt: desc.replace(/<[^>]*>/g, "").trim().slice(0, 150) });
-                    },
-                  },
-                  waitFor: 3000,
-                }),
-              }, { label: "firecrawl-naver-blog", timeoutMs: 45000 });
-              if (blogScrape.ok) {
-                const blogData = await blogScrape.json();
-                officialBlogContent = blogData.data?.json || null;
-                console.log(`[naverBlog] Official blog scraped: ${officialBlogContent?.totalPosts ?? 0} posts`);
               }
-            } catch (e) {
+              const totalPosts = (xml.match(/<totalCount>(\d+)<\/totalCount>/) || xml.match(/<managedCount>(\d+)<\/managedCount>/))?.[1];
-              console.warn(`[naverBlog] Official blog Firecrawl failed (non-critical):`, e);
+              officialBlogContent = {
+                totalPosts: totalPosts ? Number(totalPosts) : items.length,
+                recentPosts: items.slice(0, 10),
+              };
+              console.log(`[naverBlog] RSS fetched: ${items.length} posts from ${officialBlogHandle}`);
             }
+          } catch (e) {
+            console.warn(`[naverBlog] RSS fetch failed (non-critical):`, e);
           }
         }
 
@@ -533,23 +531,36 @@ Deno.serve(async (req) => {
           `${clinicName} 성형`,
           clinicName,
         ];
+        // Core name without type suffixes (e.g. "뷰성형외과" → "뷰성형외과", trimmed for short names)
+        const cleanedName = clinicName.replace(/성형외과|병원|의원/g, '').trim().toLowerCase();
+
         for (const q of queries) {
           const query = encodeURIComponent(q);
           const res = await fetchWithRetry(`https://openapi.naver.com/v1/search/local.json?query=${query}&display=5&sort=comment`, { headers: naverHeaders }, { label: "naver-place" });
           if (!res.ok) continue;
           const data = await res.json();
-          // Find the best match: prefer category containing 성형 or 피부
           const items = (data.items || []) as Record<string, string>[];
-          const match = items.find(i =>
+
-            (i.category || '').includes('성형') || (i.category || '').includes('피부')
+          const normalize = (s: string) => (s || '').replace(/<[^>]*>/g, '').toLowerCase();
-          ) || items.find(i => {
+
-            const name = (i.title || '').replace(/<[^>]*>/g, '').toLowerCase();
+          // Priority 1: name contains full clinicName (exact match, ignoring HTML tags)
-            return name.includes(clinicName.replace(/성형외과|병원|의원/g, '').trim().toLowerCase());
+          let match = items.find(i => normalize(i.title).includes(clinicName.toLowerCase())) || null;
-          }) || null;
+
+          // Priority 2: name contains cleaned short name AND category is 성형 (plastic surgery only)
+          if (!match && cleanedName.length >= 2) {
+            match = items.find(i =>
+              normalize(i.title).includes(cleanedName) && (i.category || '').includes('성형')
+            ) || null;
+          }
+
+          // Priority 3: category is 성형 (plastic surgery) — not 피부 to avoid skin clinics
+          if (!match) {
+            match = items.find(i => (i.category || '').includes('성형')) || null;
+          }
 
           if (match) {
             channelData.naverPlace = {
-              name: (match.title || "").replace(/<[^>]*>/g, ""),
+              name: normalize(match.title),
               category: match.category, address: match.roadAddress || match.address,
               telephone: match.telephone, link: match.link, mapx: match.mapx, mapy: match.mapy,
             };