feat: tag-aware daily digest - tags in AI prompt, grouping, frontmatter, cross-day continuity

tools/memos_client.py

@@ -1,12 +1,14 @@
 """Memos API client - fetch memos using Connect RPC protocol."""
 
 import logging
+from collections import Counter
 from datetime import datetime, timezone, timedelta
-
-TZ_BEIJING = timezone(timedelta(hours=8))
+from re import findall
 
 import requests
 
+TZ_BEIJING = timezone(timedelta(hours=8))
+
 logger = logging.getLogger(__name__)
 
 
@@ -45,13 +47,11 @@ class MemosClient:
     def list_memos(self, days=1, page_size=100):
         """Fetch memos from the last N days (Beijing time reference) via Connect RPC."""
         user = self.get_user_id()
-        # Use Beijing time as reference for "last N days"
         now_bj = datetime.now(TZ_BEIJING)
         since_bj = now_bj - timedelta(days=days)
         since_utc = since_bj.astimezone(timezone.utc)
         since = since_utc.isoformat()
 
-        # Connect RPC uses POST with JSON body
         payload = {
             "pageSize": page_size,
             "filter": f"creator == '{user}'",
@@ -68,7 +68,6 @@ class MemosClient:
         memos = data.get("memos", [])
         logger.info("Memos API OK | fetched %d memos", len(memos))
 
-        # Filter by time client-side (Beijing time reference)
         results = []
         for m in memos:
             created = m.get("createTime", "")
@@ -88,9 +87,7 @@ class MemosClient:
                 "visibility": m.get("visibility", ""),
             })
 
-        logger.info(
-            "Filtered to %d memos since %s", len(results), since[:10]
-        )
+        logger.info("Filtered to %d memos since %s", len(results), since[:10])
         return results
 
     def list_all_memos_from_range(self, start_date, end_date, page_size=200):
@@ -110,7 +107,6 @@ class MemosClient:
         memos = data.get("memos", [])
         logger.info("Memos API OK | fetched %d memos total", len(memos))
 
-        # Client-side date range filtering
         results = []
         for m in memos:
             created = m.get("createTime", "")
@@ -136,8 +132,76 @@ class MemosClient:
         )
         return results
 
+    def list_memos_by_tag(self, tag, page_size=200):
+        """Fetch all memos containing a specific tag.
+
+        Useful for retrieving memos across multiple days related to a specific
+        topic (e.g. a training course that spans a full week).
+        """
+        payload = {"pageSize": page_size}
+
+        resp = self.session.post(
+            f"{self.base_url}/memos.api.v1.MemoService/ListMemos",
+            json=payload,
+            timeout=15
+        )
+        resp.raise_for_status()
+        data = resp.json()
+        memos = data.get("memos", [])
+
+        results = []
+        for m in memos:
+            content_text = m.get("content", "")
+            tags = self._extract_tags(content_text)
+            if tag not in tags:
+                continue
+
+            results.append({
+                "id": m.get("name", "").split("/")[-1],
+                "content": content_text,
+                "created_at": m.get("createTime", ""),
+                "tags": tags,
+                "visibility": m.get("visibility", ""),
+            })
+
+        logger.info("Found %d memos with tag #%s", len(results), tag)
+        return results
+
+    @staticmethod
+    def aggregate_tags(memos):
+        """Count tag frequency across a list of memos.
+
+        Returns dict of {tag: count}, sorted by frequency descending.
+        """
+        counter = Counter()
+        for m in memos:
+            for tag in m.get("tags", []):
+                counter[tag] += 1
+        return dict(counter.most_common())
+
+    @staticmethod
+    def group_memos_by_tag(memos):
+        """Group memos by their tags.
+
+        Returns dict of {tag: [memo_list]}, with "无标签" key for memos
+        without any tags. A memo can appear under multiple tags.
+        """
+        groups = {}
+        untagged = []
+        for m in memos:
+            tags = m.get("tags", [])
+            if not tags:
+                untagged.append(m)
+                continue
+            for tag in tags:
+                if tag not in groups:
+                    groups[tag] = []
+                groups[tag].append(m)
+        if untagged:
+            groups["无标签"] = untagged
+        return groups
+
     @staticmethod
     def _extract_tags(content):
         """Extract #tags from memo content."""
-        import re
-        return re.findall(r"#(\w[\w\-]*)", content)
+        return findall(r"#(\w[\w\-]*)", content)