如何在RIak-erlang客户端从位串内的元组中获取数据

Question

我正在做riak-erlang客户端。在执行Mapreduce时，我得到了以下输出:

现在我想从我得到的结果集中获得数据。我想要age/name，也想通过特定的name得到age

我怎么能那样做呢?请帮帮我

{ok,[{0,R}]} = riakc_pb_socket:mapred(Pid,<<"test">>,[{map,{qfun,Mapf},none,true}]).
{ok,[{0,
  [<<"{"age": 24, "name": "krishna"}">>,
   <<"{"age": 29, "name": "sharat"}">>,
   <<"{"age": 25, "name": "ramesh"}">>,
   <<"{"age": 28, "name": "kumar"}">>,
   <<"{"age": 24, "name": "gopi"}">>,
   <<"{"age": 27, "name": "anil"}">>]}]}

:年龄:24

或

name: "krishna"

或者如果我给出名字:krishna

age:24

如何获取数据

Answer 1

基本上，你得到json编码的数据，所以首先你需要解码它，然后实现一些过滤/查找机制。我的方法是使用jiffy json解析器:

首先，克隆并构建jiffy:

git clone [e-mail]:davisp/jiffy.git;
cd jiffy; make

您必须在运行erlang命令行客户端时将jiffy添加到代码路径:

erl -pa Private/jiffy/ebin -pa Private/jiffy/deps

最后是基于name获取年龄的实现:

-module(test).
-compile(export_all).
decode(Results) ->
   [jiffy:decode(E)||E<-Results].
get_age(_, []) ->
    erlang:throw(name_not_found);
get_age(Name, [{H}|T]) ->
    case proplists:get_value(<<"name">>, H) of
        Name -> proplists:get_value(<<"age">>, H);
        _ -> get_age(Name, T)
    end.

用法:

erl -pa Private/jiffy/ebin -pa Private/jiffy/deps
Erlang R16B03 (erts-5.10.4) [source] [64-bit] [smp:8:8] [async-threads:10] [hipe] [kernel-poll:false]
Eshell V5.10.4  (abort with ^G)
1> application:start(jiffy).
ok
2> c(test).
{ok,test}
3> {ok, [{0, Results}]} = {ok,[{0,
3>   [<<"{"age": 24, "name": "krishna"}">>,
3>    <<"{"age": 29, "name": "sharat"}">>,
3>    <<"{"age": 25, "name": "ramesh"}">>,
3>    <<"{"age": 28, "name": "kumar"}">>,
3>    <<"{"age": 24, "name": "gopi"}">>,
3>    <<"{"age": 27, "name": "anil"}">>]}]}.
{ok,[{0,
      [<<"{"age": 24, "name": "krishna"}">>,
       <<"{"age": 29, "name": "sharat"}">>,
       <<"{"age": 25, "name": "ramesh"}">>,
       <<"{"age": 28, "name": "kumar"}">>,
       <<"{"age": 24, "name": "gopi"}">>,
       <<"{"age": 27, "name": "anil"}">>]}]}
4> Decoded = test:decode(Results).
[{[{<<"age">>,24},{<<"name">>,<<"krishna">>}]},
 {[{<<"age">>,29},{<<"name">>,<<"sharat">>}]},
 {[{<<"age">>,25},{<<"name">>,<<"ramesh">>}]},
 {[{<<"age">>,28},{<<"name">>,<<"kumar">>}]},
 {[{<<"age">>,24},{<<"name">>,<<"gopi">>}]},
 {[{<<"age">>,27},{<<"name">>,<<"anil">>}]}]
5> test:get_age(<<"krishna">>, Decoded).
24
6>

Answer 2

如果您不局限于使用JSON，还有另一种方法:如果您的应用程序使用Erlang，请考虑在存储对象时使用proplist。如果要像
这样存储每个值[{<<"age">>,24},{<<"name">>,<<"krishna">>}],
从您的MR返回的结果可能看起来像这样:

{ok,[{0, 
     [{<<"age">>, 24}, {<<"name">>, <<"krishna">>}],
     [{<<"age">>, 29}, {<<"name">>, <<"sharat">>}],
     [{<<"age">>, 25}, {<<"name">>, <<"ramesh">>}]}]}

那么你可以用
[ {proplists:get_value(<<"name">>,V),proplists:get_value(<<"age">>,V) || V <- R ].
返回:

[{<<"krishna">>,24},
 {<<"sharat">>,29},
 {<<"ramesh">>,25}]

要从特定记录中提取特定字段，可以使用:

[ "age: " ++ integer_to_list(proplists:get_value(<<"age">>,Record) || 
      Record <- R, proplists:get_value(<<"name">>,Record) =:= <<"krishna">> ].

还要留意R17兼容的客户端，您可能会发现Erlang的新映射数据类型很有用