* followings are now throwing JedisDataException: it was uncontrolled or controlled by Redis itself
** exec without multi
** discard without multi
** multi within multi
* updates unit test
actually Redis returns ERR and we can pick, but Pipeline + multi has some complex sequence
so it can just throw NPE without ERR
This reverts commit ac53759f97.
Sorry for double reverting, I should revert "merged commit", not origin commit
Conflicts:
src/main/java/redis/clients/jedis/BinaryClient.java
src/main/java/redis/clients/jedis/Jedis.java
src/test/java/redis/clients/jedis/tests/commands/HyperLogLogCommandsTest.java
* method signatures are a bit similar to Jedis's *SCAN
** but it takes parameters to byte[] instead of String
* ScanParams : allow match pattern with byte[]
* ScanResult : add method to get cursor with byte[] type
* *SCAN for BinaryJedis unit tests included
* Implement Closeable from Jedis, ShardedJedis with Pooled
** resources from JedisPool, JedisSentinelPool, ShardedJedis, ShardedJedisPool
* Connection class : check whether Jedis Connection is broken
** when it's time to throw JedisConnectionException, mark Connection to broken
first if Response's dependency found and not built
* there's some dependency with exec response and command responses
within multi
* if command responses's get() called before exec response's build(), it
calls exec response's build() first
* unit test included
* remove all unused methods
* move JedisSentinelPoolTest.waitForJedisSentinelPoolRecognizeNewMaster to JedisSentin
** both JedisSentinelTest and JedisSentinelPoolTest can use this implementation
* introduce FailoverAbortedException
** throws when we subscribe sentinel channels and got message by "-failover-abort-*" c
* respect Source Format to Java Convention (by Eclipse -> Source -> Format)
* remove all unused methods
* move JedisSentinelPoolTest.waitForJedisSentinelPoolRecognizeNewMaster to JedisSentinelTestUtil
** both JedisSentinelTest and JedisSentinelPoolTest can use this implementation
* introduce FailoverAbortedException
** throws when we subscribe sentinel channels and got message by "-failover-abort-*" channel
* respect Source Format to Java Convention (by Eclipse -> Source -> Format)
* let unit test fail if -failover-abort-* message is produced while waiting to failover
** waiting more is meaningless, and there is nothing we can do
** we can request "sentinel failover" to failover again, but it may can fail again
* Use "sentinel failover" to force failover
** faster than kill redis instances
* set failover timeout to 1 min
** It makes sense with failover within localhost
* reduce instances : 1 Redis Server and 1 Sentinel
** port is not changed -> I'll changed later at end of refactoring
