feat: support full click gap history generation

scripts/af_click_gap_pipeline.py

@@ -103,21 +103,33 @@ def validate_existing_schema(path: str) -> None:
         raise RuntimeError("click_gap 主键必须为 (pid, dt, pkg_name)")
 
 
-def read_rows(path: str, start: str, end: str, pid: str = PID) -> list[dict[str, Any]]:
+def read_rows(path: str, start: str | None, end: str | None, pid: str = PID) -> list[dict[str, Any]]:
     validate_existing_schema(path)
     try:
         with sqlite3.connect(f"file:{path}?mode=ro", uri=True) as conn:
             conn.row_factory = sqlite3.Row
-            result = conn.execute(
+            query = (
                 "SELECT dt, attr_platform, pkg_name, shareit_clicks, af_clicks, installs_ua, clicks_diff, clicks_diff_rate "
-                "FROM click_gap WHERE pid=? AND dt BETWEEN ? AND ? ORDER BY dt DESC, shareit_clicks DESC, pkg_name",
-                (pid, start, end),
+                "FROM click_gap WHERE pid=?"
             )
+            parameters: list[str] = [pid]
+            if start is not None and end is not None:
+                query += " AND dt BETWEEN ? AND ?"
+                parameters.extend((start, end))
+            query += " ORDER BY dt DESC, shareit_clicks DESC, pkg_name"
+            result = conn.execute(query, parameters)
             return [dict(row) for row in result]
     except sqlite3.Error as exc:
         raise RuntimeError(f"SQLite 查询失败: {exc}") from exc
 
 
+def available_dates(rows: list[dict[str, Any]]) -> list[str]:
+    dates = set()
+    for row in rows:
+        dates.add(date8(str(row["dt"])))
+    return sorted(dates, reverse=True)
+
+
 def validate_rows(rows: list[dict[str, Any]], start: str, end: str) -> list[dict[str, Any]]:
     normalized = []
     for row in rows:
@@ -213,6 +225,7 @@ def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(description="Read SQLite click-gap data and publish JSON")
     parser.add_argument("--start-date", help="读取开始日期 YYYY-MM-DD")
     parser.add_argument("--end-date", help="读取结束日期 YYYY-MM-DD")
+    parser.add_argument("--all", action="store_true", help="读取数据库中所有实际有数据的日期")
     parser.add_argument("--dry-run", action="store_true", help="只读取、校验并生成临时 JSON，不写文件或 Git")
     group = parser.add_mutually_exclusive_group()
     group.add_argument("--no-push", action="store_true", help="写文件但不 push")
@@ -226,11 +239,19 @@ def main() -> int:
     try:
         if args.dry_run and (args.no_push or args.push):
             raise ValueError("--dry-run 不可与 --push/--no-push 同时使用")
-        start, end = window(args.start_date, args.end_date)
-        rows = validate_rows(read_rows(args.db_path, start, end), start, end)
-        first = dt.datetime.strptime(start, "%Y%m%d").date()
-        last = dt.datetime.strptime(end, "%Y%m%d").date()
-        dates = [(first + dt.timedelta(days=i)).strftime("%Y%m%d") for i in range((last - first).days + 1)]
+        if args.all and (args.start_date or args.end_date):
+            raise ValueError("--all 不可与 --start-date/--end-date 同时使用")
+        if args.all:
+            rows = validate_rows(read_rows(args.db_path, None, None), "00010101", "99991231")
+            dates = available_dates(rows)
+            start = dates[-1] if dates else "无日期"
+            end = dates[0] if dates else "无日期"
+        else:
+            start, end = window(args.start_date, args.end_date)
+            rows = validate_rows(read_rows(args.db_path, start, end), start, end)
+            first = dt.datetime.strptime(start, "%Y%m%d").date()
+            last = dt.datetime.strptime(end, "%Y%m%d").date()
+            dates = [(first + dt.timedelta(days=i)).strftime("%Y%m%d") for i in range((last - first).days + 1)]
         repo = Path(__file__).resolve().parents[1]
         with tempfile.TemporaryDirectory(prefix="af-click-gap-") as temp:
             generated = Path(temp) / "data"