init

core/actions.py

@@ -0,0 +1,87 @@
+from __future__ import annotations
+
+import time
+from dataclasses import dataclass
+from typing import Optional
+
+import pyautogui
+
+from core.ocr_client import UmiClient
+from utils.screenshot import capture_screen
+
+
+@dataclass(frozen=True)
+class MatchResult:
+    # 屏幕坐标（可直接点击）
+    x: int
+    y: int
+    # 命中的文字（当前实现为目标文字本身；若后续做模糊匹配可返回实际匹配串）
+    text: str
+
+
+class ActionRunner:
+    def __init__(self, ocr: UmiClient, *, prefer_mss: bool = True, default_region: Optional[tuple[int, int, int, int]] = None):
+        self.ocr = ocr
+        self.prefer_mss = prefer_mss
+        self.default_region = default_region
+
+    def locate_text(
+        self,
+        text: str,
+        *,
+        region: Optional[tuple[int, int, int, int]] = None,
+        exact: bool = True,
+        case_sensitive: bool = False,
+    ) -> Optional[MatchResult]:
+        cap = capture_screen(region or self.default_region, prefer_mss=self.prefer_mss)
+        items = self.ocr.ocr_bytes(cap.image_bytes)
+        pt = self.ocr.find_text(text, items, exact=exact, case_sensitive=case_sensitive)
+        if pt is None:
+            return None
+
+        img_x, img_y = pt
+        left, top, _, _ = cap.region
+        # OCR 坐标是“截图图片像素坐标”，需先缩放到屏幕坐标，再加上截图区域偏移
+        scr_x = int(left + img_x * cap.scale_x)
+        scr_y = int(top + img_y * cap.scale_y)
+        return MatchResult(x=scr_x, y=scr_y, text=text)
+
+    def click_text(
+        self,
+        text: str,
+        *,
+        region: Optional[tuple[int, int, int, int]] = None,
+        exact: bool = True,
+        case_sensitive: bool = False,
+        clicks: int = 1,
+        interval: float = 0.05,
+        button: str = "left",
+        move_duration: float = 0.0,
+        pause: float = 0.05,
+    ) -> MatchResult:
+        pyautogui.PAUSE = pause
+        m = self.locate_text(text, region=region, exact=exact, case_sensitive=case_sensitive)
+        if m is None:
+            raise TimeoutError(f"未找到文字：{text}")
+        pyautogui.moveTo(m.x, m.y, duration=move_duration)
+        pyautogui.click(x=m.x, y=m.y, clicks=clicks, interval=interval, button=button)
+        return m
+
+    def wait_for_text(
+        self,
+        text: str,
+        *,
+        timeout: float = 20.0,
+        interval: float = 0.5,
+        region: Optional[tuple[int, int, int, int]] = None,
+        exact: bool = True,
+        case_sensitive: bool = False,
+    ) -> MatchResult:
+        end = time.time() + timeout
+        while time.time() < end:
+            m = self.locate_text(text, region=region, exact=exact, case_sensitive=case_sensitive)
+            if m is not None:
+                return m
+            time.sleep(interval)
+        raise TimeoutError(f"等待超时：{timeout}s 内未出现文字：{text}")
+