Dev/no can read (#2360)

* make can_read() destructive for simplicity, and rename the method.
Remove timeout argument, always timeout immediately.

* don't use can_read in pubsub

* connection.connect() now has its own retry, don't need it inside a retry loop

1 files changed, 15 insertions, 7 deletions

diff --git a/redis/asyncio/client.py b/redis/asyncio/client.py
index 9c8caae..c13054b 100644
--- a/redis/asyncio/client.py
+++ b/redis/asyncio/client.py
@@ -24,6 +24,8 @@ from typing import (
     cast,
 )
 
+import async_timeout
+
 from redis.asyncio.connection import (
     Connection,
     ConnectionPool,
@@ -754,15 +756,21 @@ class PubSub:
 
         await self.check_health()
 
-        async def try_read():
-            if not block:
-                if not await conn.can_read(timeout=timeout):
+        if not conn.is_connected:
+            await conn.connect()
+
+        if not block:
+
+            async def read_with_timeout():
+                try:
+                    async with async_timeout.timeout(timeout):
+                        return await conn.read_response()
+                except asyncio.TimeoutError:
                     return None
-            else:
-                await conn.connect()
-            return await conn.read_response()
 
-        response = await self._execute(conn, try_read)
+            response = await self._execute(conn, read_with_timeout)
+        else:
+            response = await self._execute(conn, conn.read_response)
 
         if conn.health_check_interval and response == self.health_check_response:
             # ignore the health check message as user might not expect it