第一次提交

2026-08-01 16:50:55 +08:00
commit c8f9e39039
68 changed files with 6016 additions and 0 deletions
@@ -0,0 +1,32 @@
+"""Pix2Text 独立微服务的 HTTP 客户端（后续阶段部署 OCR 服务时启用）。"""
+import httpx
+
+from app.services.ocr.base import OcrResult
+
+
+class Pix2TextClient:
+    def __init__(self, service_url: str, timeout: float = 60.0):
+        self._url = service_url.rstrip("/")
+        self._timeout = timeout
+
+    async def healthy(self) -> bool:
+        try:
+            async with httpx.AsyncClient(timeout=5.0) as c:
+                r = await c.get(f"{self._url}/health")
+                return r.status_code == 200
+        except httpx.HTTPError:
+            return False
+
+    async def recognize(self, image_bytes: bytes, mime: str) -> OcrResult:
+        async with httpx.AsyncClient(timeout=self._timeout) as c:
+            r = await c.post(
+                f"{self._url}/ocr",
+                files={"file": ("image", image_bytes, mime)},
+            )
+            r.raise_for_status()
+            data = r.json()
+        return OcrResult(
+            markdown=data["markdown"],
+            confidence=float(data.get("confidence", 1.0)),
+            engine="pix2text",
+        )