初始化红餐观察库项目

2026-07-19 18:11:08 +08:00
commit c663539548
22 changed files with 1261 additions and 0 deletions
@@ -0,0 +1,220 @@
+#!/usr/bin/env python3
+"""Local Hongcan article library and Qwen-assisted editorial analysis."""
+
+from __future__ import annotations
+
+import json
+import os
+import sqlite3
+import urllib.error
+import urllib.request
+from datetime import datetime, timezone
+from pathlib import Path
+
+from dotenv import load_dotenv
+from flask import Flask, abort, jsonify, render_template, request
+
+
+BASE_DIR = Path(__file__).resolve().parent
+DB_PATH = Path(os.getenv("HONGCAN_DB", BASE_DIR / "data" / "hongcan.db"))
+load_dotenv(BASE_DIR / ".env")
+
+app = Flask(__name__)
+app.secret_key = os.getenv("FLASK_SECRET_KEY", "hongcan-local-only")
+
+
+def now_iso() -> str:
+    return datetime.now(timezone.utc).astimezone().isoformat(timespec="seconds")
+
+
+def db() -> sqlite3.Connection:
+    conn = sqlite3.connect(DB_PATH)
+    conn.row_factory = sqlite3.Row
+    conn.execute("PRAGMA foreign_keys = ON")
+    return conn
+
+
+def ensure_schema() -> None:
+    with db() as conn:
+        conn.executescript((BASE_DIR / "schema.sql").read_text(encoding="utf-8"))
+
+
+def clamp_int(value: str | None, default: int, low: int, high: int) -> int:
+    try:
+        return max(low, min(high, int(value or default)))
+    except ValueError:
+        return default
+
+
+TREND_CN = {"emerging": "新兴", "accelerating": "加速", "stable": "稳定", "declining": "下行"}
+
+
+@app.template_filter("date_cn")
+def date_cn(value: str | None) -> str:
+    if not value:
+        return "时间未知"
+    try:
+        return datetime.fromisoformat(value).strftime("%Y.%m.%d %H:%M")
+    except ValueError:
+        return value
+
+
+@app.get("/")
+def index():
+    query = (request.args.get("q") or "").strip()
+    page = clamp_int(request.args.get("page"), 1, 1, 10_000)
+    per_page = 18
+    where = "WHERE crawl_status='success'"
+    params: list[object] = []
+    if query:
+        where += " AND (title LIKE ? OR author LIKE ? OR content LIKE ?)"
+        needle = f"%{query}%"
+        params.extend([needle, needle, needle])
+    with db() as conn:
+        total = conn.execute(f"SELECT COUNT(*) FROM articles {where}", params).fetchone()[0]
+        rows = conn.execute(
+            f"""
+            SELECT a.id,a.title,a.author,a.published_at,a.summary,a.canonical_url,
+                   substr(a.content,1,180) AS excerpt,
+                   json_extract(x.result_json,'$.summary') AS ai_summary,
+                   CASE WHEN x.id IS NULL THEN 0 ELSE 1 END AS analyzed
+            FROM articles a
+            LEFT JOIN ai_analyses x ON x.article_id=a.id AND x.analysis_type='editorial'
+            {where.replace('crawl_status', 'a.crawl_status')}
+            ORDER BY a.published_at DESC,a.id DESC LIMIT ? OFFSET ?
+            """,
+            [*params, per_page, (page - 1) * per_page],
+        ).fetchall()
+        stats = conn.execute(
+            """
+            SELECT COUNT(*) total,
+                   SUM(crawl_status='success') success,
+                   MAX(published_at) latest,
+                   (SELECT COUNT(DISTINCT article_id) FROM ai_analyses) analyzed
+            FROM articles
+            """
+        ).fetchone()
+        signal_rows = conn.execute(
+            "SELECT * FROM industry_signals ORDER BY signal_date DESC,confidence DESC LIMIT 8"
+        ).fetchall()
+        signals = []
+        for signal in signal_rows:
+            item = dict(signal)
+            item["tags"] = json.loads(item["tags_json"])
+            item["implications"] = json.loads(item["implications_json"])
+            item["article_ids"] = json.loads(item["article_ids_json"])
+            item["trend"] = TREND_CN.get(item["trend"], item["trend"])
+            signals.append(item)
+    return render_template(
+        "index.html", articles=rows, query=query, page=page, total=total,
+        pages=max(1, (total + per_page - 1) // per_page), stats=stats, signals=signals,
+    )
+
+
+@app.get("/signals/<int:signal_id>")
+def signal_detail(signal_id: int):
+    with db() as conn:
+        signal = conn.execute("SELECT * FROM industry_signals WHERE id=?", (signal_id,)).fetchone()
+        if not signal:
+            abort(404)
+        signal = dict(signal)
+        signal["trend"] = TREND_CN.get(signal["trend"], signal["trend"])
+        article_ids = json.loads(signal["article_ids_json"])
+        articles = []
+        if article_ids:
+            marks = ",".join("?" for _ in article_ids)
+            articles = conn.execute(
+                f"SELECT id,title,published_at FROM articles WHERE id IN ({marks}) ORDER BY published_at DESC",
+                article_ids,
+            ).fetchall()
+    return render_template(
+        "signal.html", signal=signal, articles=articles,
+        implications=json.loads(signal["implications_json"]), tags=json.loads(signal["tags_json"]),
+    )
+
+
+@app.get("/article/<int:article_id>")
+def article(article_id: int):
+    with db() as conn:
+        row = conn.execute("SELECT * FROM articles WHERE id=?", (article_id,)).fetchone()
+        if not row:
+            abort(404)
+        analysis = conn.execute(
+            "SELECT * FROM ai_analyses WHERE article_id=? AND analysis_type='editorial' ORDER BY updated_at DESC LIMIT 1",
+            (article_id,),
+        ).fetchone()
+    parsed = json.loads(analysis["result_json"]) if analysis else None
+    return render_template("article.html", article=row, analysis=parsed, analysis_row=analysis)
+
+
+def qwen_analyze(title: str, content: str) -> dict:
+    api_key = os.getenv("DASHSCOPE_API_KEY", "").strip()
+    if not api_key:
+        raise RuntimeError("尚未配置 DASHSCOPE_API_KEY，请在 .env 中设置")
+    model = os.getenv("QWEN_MODEL", "qwen-plus")
+    prompt = f"""你是一名资深餐饮行业研究编辑。分析下面这篇文章，只返回合法 JSON，不要 Markdown 代码块。
+JSON 字段必须为：summary（120字内摘要）、key_points（3-6条字符串）、industry_signal（行业信号）、brands（品牌数组）、numbers（关键数据数组）、content_angles（3条可延展选题）、risk_notes（事实或表述风险数组）、tags（5-10个标签）、sentiment（positive/neutral/negative）。
+文章标题：{title}
+正文：{content[:24000]}"""
+    payload = json.dumps({
+        "model": model,
+        "messages": [
+            {"role": "system", "content": "你输出严谨、简洁、可供编辑部直接使用的结构化行业分析。"},
+            {"role": "user", "content": prompt},
+        ],
+        "temperature": 0.25,
+        "response_format": {"type": "json_object"},
+    }, ensure_ascii=False).encode("utf-8")
+    req = urllib.request.Request(
+        "https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions",
+        data=payload,
+        headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
+        method="POST",
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=90) as response:
+            body = json.loads(response.read().decode("utf-8"))
+    except urllib.error.HTTPError as exc:
+        detail = exc.read().decode("utf-8", errors="replace")[:500]
+        raise RuntimeError(f"千问接口返回 {exc.code}：{detail}") from exc
+    text = body["choices"][0]["message"]["content"].strip()
+    if text.startswith("```"):
+        text = text.strip("`").removeprefix("json").strip()
+    return json.loads(text)
+
+
+@app.post("/api/articles/<int:article_id>/analyze")
+def analyze(article_id: int):
+    force = bool((request.get_json(silent=True) or {}).get("force"))
+    model = os.getenv("QWEN_MODEL", "qwen-plus")
+    with db() as conn:
+        row = conn.execute("SELECT * FROM articles WHERE id=?", (article_id,)).fetchone()
+        if not row:
+            return jsonify({"ok": False, "error": "文章不存在"}), 404
+        cached = conn.execute(
+            "SELECT * FROM ai_analyses WHERE article_id=? AND model=? AND analysis_type='editorial'",
+            (article_id, model),
+        ).fetchone()
+        if cached and cached["content_hash"] == row["content_hash"] and not force:
+            return jsonify({"ok": True, "cached": True, "analysis": json.loads(cached["result_json"])})
+        try:
+            result = qwen_analyze(row["title"], row["content"])
+        except Exception as exc:
+            return jsonify({"ok": False, "error": str(exc)}), 502
+        timestamp = now_iso()
+        conn.execute(
+            """
+            INSERT INTO ai_analyses(article_id,model,analysis_type,result_json,content_hash,created_at,updated_at)
+            VALUES (?,?, 'editorial', ?,?,?,?)
+            ON CONFLICT(article_id,model,analysis_type) DO UPDATE SET
+                result_json=excluded.result_json,content_hash=excluded.content_hash,updated_at=excluded.updated_at
+            """,
+            (article_id, model, json.dumps(result, ensure_ascii=False), row["content_hash"], timestamp, timestamp),
+        )
+        conn.commit()
+    return jsonify({"ok": True, "cached": False, "analysis": result})
+
+
+if __name__ == "__main__":
+    ensure_schema()
+    app.run(host="127.0.0.1", port=int(os.getenv("PORT", "8766")), debug=False)