README.md

A.L.F.R.E.D - JSON Client

Compilation

alfred-json depends on:

• libjansson

• zlib

  $ apt-get install cmake libjansson-dev zlib1g-dev

Prepare the build using:

 $ mkdir build
 $ cd build/
 $ cmake ../

To compile, simply type:

 $ make

To install, use:

 $ make install

Usage

First, an alfred daemon must be running. Data can be retrieved like this:

 $ alfred-json -r 64
{
  "fe:f1:00:00:01:01": {
    "firmware": "0.3.2.1-1",
    "hostname": "alfredtest1"
  },
  "fe:f1:00:00:02:01": {
    "key": "value"
  }
}

There are three output formats available: json, string (UTF-8) and binary. These can be selected using the -f switch. "json" is default.

 $ alfred-json -r 64 -f string
{
  "fe:f1:00:00:01:01": "{ \"hostname\": \"alfredtest1\", \"firmware\": \"0.3.2.1-1\" }\n",
  "fe:f1:00:00:02:01": "{ \"key\": \"value\" }\n"
}

String output is wrapped in an JSON object. You'll get a warning if the string is invalid UTF-8.

JSON output tries to interprete the string as an JSON object. You'll get a warning if in case the string does not represent a valid JSON object.

 $ alfred-json -r 64 -f json
{
  "fe:f1:00:00:01:01": {
    "firmware": "0.3.2.1-1",
    "hostname": "alfredtest1"
  },
  "fe:f1:00:00:02:01": {
    "key": "value"
  }
}

Binary output is meant to be parsed by other programs only. The format is:

  struct output {
    struct item items[n];
  };

  struct item {
    uint16_t id_len; // big-endian
    uint8_t id[id_len];

    uint16_t data_len; // big-endian
    uint8_t data[data_len];
  };

Transparent GZIP decompression

alfred-json will automatically and transparently decompress GZIP compressed data when the -z option is given:

 $ echo Hello World | gzip | alfred -s 150
 $ alfred-json -z -r 150 -f string

License

alfred-json is licensed under the terms of version 2 of the GNU General Public License (GPL). Please see the LICENSE file.