feat(baoyu-fetch): add URL reader CLI with Chrome CDP and site adapters

2026-07-21 01:39:48 +08:00 · 2026-03-27 14:11:00 -05:00
parent 2c14872e88
commit d0764c2739
65 changed files with 10235 additions and 0 deletions
@@ -0,0 +1,51 @@
+export type ContentBlock =
+  | {
+      type: "paragraph";
+      text: string;
+    }
+  | {
+      type: "heading";
+      depth: number;
+      text: string;
+    }
+  | {
+      type: "list";
+      ordered: boolean;
+      items: string[];
+    }
+  | {
+      type: "quote";
+      text: string;
+    }
+  | {
+      type: "code";
+      code: string;
+      language?: string;
+    }
+  | {
+      type: "image";
+      url: string;
+      alt?: string;
+    }
+  | {
+      type: "html";
+      html: string;
+    }
+  | {
+      type: "markdown";
+      markdown: string;
+    };
+
+export interface ExtractedDocument {
+  url: string;
+  requestedUrl?: string;
+  canonicalUrl?: string;
+  title?: string;
+  author?: string;
+  siteName?: string;
+  publishedAt?: string;
+  summary?: string;
+  content: ContentBlock[];
+  metadata?: Record<string, unknown>;
+  adapter?: string;
+}