scraping and improvements

frontend/src/app/scrape/story/route.ts

@@ -0,0 +1,85 @@
+import { NextRequest, NextResponse } from 'next/server';
+
+export async function POST(request: NextRequest) {
+  try {
+    const body = await request.json();
+    const { url } = body;
+
+    if (!url || typeof url !== 'string') {
+      return NextResponse.json(
+        { error: 'URL is required and must be a string' },
+        { status: 400 }
+      );
+    }
+
+    // Dynamic import to prevent client-side bundling
+    const { StoryScraper } = await import('@/lib/scraper/scraper');
+    const { ScraperError } = await import('@/lib/scraper/types');
+    
+    const scraper = new StoryScraper();
+    const story = await scraper.scrapeStory(url);
+
+    // Debug logging
+    console.log('Scraped story data:', {
+      url: url,
+      title: story.title,
+      author: story.author,
+      summary: story.summary,
+      contentLength: story.content?.length || 0,
+      contentPreview: story.content?.substring(0, 200) + '...',
+      tags: story.tags,
+      coverImage: story.coverImage
+    });
+
+    return NextResponse.json(story);
+  } catch (error) {
+    console.error('Story scraping error:', error);
+
+    // Check if it's a ScraperError without importing at module level
+    if (error && typeof error === 'object' && error.constructor.name === 'ScraperError') {
+      return NextResponse.json(
+        { 
+          error: (error as any).message,
+          url: (error as any).url 
+        },
+        { status: 400 }
+      );
+    }
+
+    if (error instanceof Error) {
+      // Handle specific error types
+      if (error.message.includes('Invalid URL')) {
+        return NextResponse.json(
+          { error: 'Invalid URL provided' },
+          { status: 400 }
+        );
+      }
+
+      if (error.message.includes('Unsupported site')) {
+        return NextResponse.json(
+          { error: 'This website is not supported for scraping' },
+          { status: 400 }
+        );
+      }
+
+      if (error.message.includes('HTTP 404')) {
+        return NextResponse.json(
+          { error: 'Story not found at the provided URL' },
+          { status: 404 }
+        );
+      }
+
+      if (error.message.includes('timeout')) {
+        return NextResponse.json(
+          { error: 'Request timed out while fetching content' },
+          { status: 408 }
+        );
+      }
+    }
+
+    return NextResponse.json(
+      { error: 'Failed to scrape story. Please try again.' },
+      { status: 500 }
+    );
+  }
+}