kdr2 / redis-leveldb

A redis-protocol compatible frontend to google's leveldb

Hi there 👋

redis-leveldb's Issues

snappy and recent GNU/Linux distribution

For your information,

It would be wise to include snappy in LDFLAGS for all the operating systems:

 LDFLAGS += -lsnappy ../vendor/libleveldb.a -lm -L$(LIBEV)/lib -lev -L$(GMP)/lib -lgmp 

and not only Darwin. As the snappy compression library is required for LevelDB linking.

Thank you very much for your work.

[问题]空间占用问题

这个项目非常的酷，我非常喜欢，从年初开始使用。

看到最近增加了hash,set,list,等功能。此前我使用的是将一个object用msgpack后，然后存储为一个kv类型的值，然后在取出时，再decode。

现在有了hash，set，list之后，我的代码可以简化，但是不知道hash、set、list的空间占用是怎样计算的？

encoding with sadd

Hi

I use redis-leveldb with pyredis. I noticed that the encoding is wrong for sadd statements but good for others. Here is the code:

leveldb_con = redis.StrictRedis(host=leveldb_host, port=leveldb_port, db=0, charset='utf-8')
pipe = leveldb_con.pipeline(transaction=False)

key='toto'
val=9999

skey='tutu'
sval='127.0.0.1'

pipe.set(key, val)
pipe.sadd(skey, sval)
pipe.execute()

And the output:

127.0.0.1:8324> KEYS *
1) "\x02\x03size\x04tutu"
2) "\x03\x04tutu\t127.0.0.1"
3) "toto"
127.0.0.1:8324>

I am wondering if I missed something of if it is a known bug?

consecutive delete and set

got this error during a consecutive delete and set

After a period of time of executing these commands I get a segfault

key = "test"
for (i = 0; i< 100; i++) {
conn.delete(key)
conn.set(key, "test")
}

Note: multiple threads

terminate called after throwing an instance of 'std::length_error'
what(): basic_string::_S_create

Program received signal SIGABRT, Aborted.
[Switching to Thread 801c07400 (LWP 100406/redis-leveldb)]
0x00000008016e0a7c in thr_kill () from /lib/libc.so.7
(gdb)
(gdb) bt
#0 0x00000008016e0a7c in thr_kill () from /lib/libc.so.7
#1 0x000000080177dd3b in abort () from /lib/libc.so.7
#2 0x0000000800fc4c94 in __gnu_cxx::__verbose_terminate_handler () from /usr/lib/libstdc++.so.6
#3 0x0000000800fc9193 in std::set_unexpected () from /usr/lib/libstdc++.so.6
#4 0x0000000800fc91d3 in std::terminate () from /usr/lib/libstdc++.so.6
#5 0x0000000800fc913a in __cxa_throw () from /usr/lib/libstdc++.so.6
#6 0x000000080101a972 in std::__throw_length_error () from /usr/lib/libstdc++.so.6
#7 0x0000000800fca5d9 in std::string::_Rep::_S_create () from /usr/lib/libstdc++.so.6
#8 0x0000000800fcb305 in std::string::_S_copy_chars () from /usr/lib/libstdc++.so.6
#9 0x0000000800fcb4ba in std::basic_string<char, std::char_traits, std::allocator >::basic_string () from /usr/lib/libstdc++.so.6
#10 0x00000000004072fe in RLConnection::do_read ()
#11 0x0000000000407739 in RLConnection::on_readable ()
#12 0x00000008008aac52 in ev_invoke_pending (loop=0x801c6f600) at ev.c:2994
#13 0x00000008008af37c in ev_run (loop=0x801c6f600, flags=Variable "flags" is not available.

) at ev.c:3394
#14 0x00000000004049fb in RLServer::start ()

missing license for the repo

👋 Just found this repo does not have any license, it would be good to add OSI-approved one, https://spdx.org/licenses/

Some popular license choice would be:

• MIT
• Apache-2.0
• BSD-3-Clause

sort()

Is there a way to use something like the sort() command or a workaround?

Actually I would not like to load all the 10k+ items into my nodejs to fetch 10 with the highest value.

Or any tips on that?

Fixed buffer produces livelock

When a request that's too big to fit in the buffer is issued RLConnection::on_readable will livelock.

I can propose a fix for this issue by using a growing buffer.

(error) ERR redis-leveldb is running in single-db mode

(error) ERR redis-leveldb is running in single-db mode

但是我已经使用了multiple模式，数据库目录也已经生成了。

ev.h and leveldb/c.h Not Found

cd src && make all
make[1]: “/home/test/Downloads/redis-leveldb-1.4/src”
g++ -Wall -I/usr/include -I/usr/include -I../vendor/leveldb/include -c -o rl_util.o rl_util.cpp
In file included from rl_util.cpp:22:
rl_server.h:14:10: fatal error: ev.h: Not Found
#include <ev.h>
^~~~~~
compilation terminated.

ev.h and leveldb/c.h not in code ？how to run this project

Benchmarks?

Cool project! Any benchmarks?

Scard seems to not work

This is a little example as a "proof of bug" In normal redis this command (r_serv.scard("i")) would return a "0" if not existing: http://redis.io/commands/scard

In [1]: import redis
In [3]: r_serv = redis.StrictRedis("127.0.0.1", 8323, 10)
In [4]: r_serv.smembers("i")
Out[4]: set([])
In [5]: r_serv.scard("i")

TypeError Traceback (most recent call last)
in ()
----> 1 r_serv.scard("i")

/usr/local/lib/python2.7/dist-packages/redis/client.pyc in scard(self, name)
1250 def scard(self, name):
1251 "Return the number of elements in set name"
-> 1252 return self.execute_command('SCARD', name)
1253
1254 def sdiff(self, keys, *args):

/usr/local/lib/python2.7/dist-packages/redis/client.pyc in execute_command(self, _args, *_options)
459 try:
460 connection.send_command(_args)
--> 461 return self.parse_response(connection, command_name, *_options)
462 except ConnectionError:
463 connection.disconnect()

/usr/local/lib/python2.7/dist-packages/redis/client.pyc in parse_response(self, connection, command_name, *_options)
471 response = connection.read_response()
472 if command_name in self.response_callbacks:
--> 473 return self.response_callbacks command_name (response, *_options)
474 return response
475

TypeError: int() argument must be a string or a number, not 'NoneType'

只输入incr 不带任何参数Segmentation fault

Redis其他命令的支持问题

这个项目真的非常好，打算用它来替换redis。
请问Redis其他的命令比如hincr/hincrby和sorted set类型后面会支持吗？

std::length_error

During our batch imports redis-leveldb goes 100% CPU, not even a restart helps. Before that (or maybe unrelated - not sure) it throws a lot of these:

terminate called after throwing an instance of 'std::length_error'
  what():  basic_string::_S_create

I do not know any C/C++. How could I help debug this error?