Showing revision 8

JSONLab

An open-source MATLAB/Octave JSON/UBJSON encoder and decoder

Table of Content:

1. Introduction
2. Installation
3. Using JSONLab
4. Known Issues and TODOs
5. Contribution and feedback
6. Acknowledgement

0 What's New

JSONLab v1.9.8 is the beta release of the next milestone - code named "Magnus".

Starting from this release, JSONLab supports encoding/decoding MessagePack, a widely-used binary JSON-like data format. Via ZMat v0.9, JSONLab v1.9.8 also supports LZMA/LZ4/LZ4HC data compression/decompression. More importantly, JSONLab is now the official reference implementation for JData Specification (Draft 2) as defined in http://github.com/fangq/jdata, the foundation for the OpenJData Project (http://openjdata.org).

There have been numerous major updates to this toolbox since the previous release v1.9 in May 2019. A list of the major changes are summarized below with key features marked by *:

  • 2019-10-22*[650b5ec] enable preencode by default for savejson and saveubjson
  • 2019-10-21*[874945f] decode graph data, encode non-char-keyed map data
  • 2019-10-18 [11712b7] add any2jd, pass opt to name check, add more options
  • 2019-10-18*[f97de9b] extract name encoding/decoding to separate function, like in easyh5
  • 2019-10-17*[9d0fd4a] rewrite jdataencode
  • 2019-10-15 [23f14d6] minor updates to make msgpack to work on octave
  • 2019-09-16*[689cb40] support lz4 and lz4hc compression via zmat v0.9
  • 2019-07-11*[06d33aa] update zmat test to support zmat v0.8 mox-the-fox
  • 2019-06-24*[eba4078] saving table objects with new syntax
  • 2019-06-12 [3eb6d56] change ArrayCompression keywords to ArrayZip to be short
  • 2019-06-12*[e5f2ffb] complete saveubjson debug mode, add compression example
  • 2019-06-11 [ebbcfd2] pass formatversion tag to jdatadecode
  • 2019-06-10*[95b2eb0] add jdataencode and jdatadecode
  • 2019-06-10*[f86219d] major update: use row-major for N-D array, incompatible with old JSONLab
  • 2019-05-31*[0c467ee] support lzma and lzip compression decompression via zmat toolbox (http://github.com/fangq/zmat)
  • 2019-05-31 [599ee4c] support categorical data
  • 2019-05-30*[d47be45] fast bracket matching
  • 2019-05-24*[0ec2d01] rewriting fastarrayparser, 10x faster for Octave, close #4 with fast bracket matching
  • 2019-05-22*[d8c19b8] add support to MessagePack, close #53, add NestArray option, close #6
  • 2019-05-19*[c87e7d2] support containers.Map

Please note that JSONLab v1.9.8 is compliant with JData Spec Draft 2, while v1.9 and previous releases are compatible with Draft 1. The main differences are

  • _ArrayCompressionMethod_, _ArrayCompressionSize_</tt> and _ArrayCompressedData_ are replaced by _ArrayZipType_, <tt>_ArrayZipSize_ and _ArrayZipData_, respectively
  • The serialization of N-D array data stored in <tt>_ArrayData_is changed from column-major to row-major

To read data files generated by JSONLab v1.9 or older versions, you need to attach option <tt>'FormatVersion', 1.9 in all the loadjson/savejson function calls. To convert an older file (JSON/UBJSON) to the new format, you should run 

   data=loadjson('my_old_data_file.json','FormatVersion',1.9)
   savejson('',data,'FileName','new_file.json')

You are strongly encouraged to convert all previously generated data files using the new format.

1. Introduction

JSONLab is a free and open-source implementation of a JSON/UBJSON/MessagePack encoder and a decoder in the native MATLAB language. It can be used to convert a MATLAB data structure (array, struct, cell, struct array, cell array, and objects) into JSON/UBJSON/MessagePack formatted strings, or to decode a JSON/UBJSON/MessagePack file into MATLAB data structure. JSONLab supports both MATLAB and GNU Octave (a free MATLAB clone).

JSON (JavaScript Object Notation) is a highly portable, human-readable and "fat-free" text format to represent complex and hierarchical data. It is as powerful as XML but less verbose. JSON format is widely used for data-exchange in applications.

UBJSON (Universal Binary JSON) is a binary JSON format, specifically specifically optimized for compact file size and better performance while keeping the semantics as simple as the text-based JSON format. Using the UBJSON format allows to wrap complex binary data in a flexible and extensible structure, making it possible to process complex and large dataset without accuracy loss due to text conversions. MessagePack is another binary JSON-like data format widely used in data exchange in web/native applications. It is slightly more compact than UBJSON, but is not directly readable compared to UBJSON.

We envision that both JSON and its binary counterparts will play important roles as mainstream data-exchange formats for scientific research. It has both the flexibility and generality as offered by other popular general-purpose file specifications, such as HDF5 but with significantly reduced complexity and excellent readability.

Towards this goal, we have developed the JData Specification (http://github.com/fangq/jdata) to standardize serializations of complex scientific data structures, such as N-D arrays, sparse/complex-valued arrays, trees, maps, tables and graphs using JSON/binary JSON constructs. The text and binary formatted JData files are syntactically compatible with JSON/UBJSON formats, and can be readily parsed using existing JSON and UBJSON parsers.

Please note that data files produced by saveubjson may utilize a special "optimized header" to store N-D (N>1) arrays, as defined in the JData Specification Draft 2. This feature is not supported by UBJSON Specification Draft 12. To produce UBJSON files that can be parsed by UBJSON-Draft-12 compliant parsers, you must add the option 'NestArray',1 in the call to saveubjson.

2. Installation

The installation of JSONLab is no different from installing any other MATLAB toolbox. You only need to download/unzip the JSONLab package to a folder, and add the folder's path to MATLAB/Octave's path list by using the following command: 

    addpath('/path/to/jsonlab');

If you want to add this path permanently, you need to type "pathtool", browse to the zmat root folder and add to the list, then click "Save". Then, run "rehash" in MATLAB, and type "which savejson", if you see an output, that means JSONLab is installed for MATLAB/Octave.

If you use MATLAB in a shared environment such as a Linux server, the best way to add path is to type 

   mkdir ~/matlab/
   nano ~/matlab/startup.m

and type addpath('/path/to/jsonlab') in this file, save and quit the editor. MATLAB will execute this file every time it starts. For Octave, the file you need to edit is ~/.octaverc , where "~" is your home directory.

Install JSONLab on Fedora 24 or later

JSONLab has been available as an official Fedora package since 2015. You may install it directly using the below command 

   sudo dnf install octave-jsonlab

To enable data compression/decompression, you are encouraged to install <tt>octave-zmat using 

   sudo dnf install octave-zmat

Install JSONLab on Arch Linux

JSONLab is also available on Arch Linux. You may install it using the below command 

   sudo pacman -S jsonlab

3. Using JSONLab

JSONLab provides two functions, loadjson -- a MATLAB->JSON decoder, and savejson -- a MATLAB->JSON encoder, for the text-based JSON, and two equivallent function pairs -- loadubjson and saveubjson for binary JSON and loadmsgpack and savemsgpack for MessagePack. The load functions for the 3 supported data formats share almost the same input parameters; similarly for the 3 save functions (savejson/saveubjson/savemsgpack) The detailed help information can be found in the Contents.m file.

In the below section, we simply provide a few examples on how to use each of the core functions for encoding/decoding JSON/UBJSON/MessagePack data

savejson.m

       jsonmesh=struct('MeshNode',[0 0 0;1 0 0;0 1 0;1 1 0;0 0 1;1 0 1;0 1 1;1 1 1],... 
                'MeshElem',[1 2 4 8;1 3 4 8;1 2 6 8;1 5 6 8;1 5 7 8;1 3 7 8],...
                'MeshSurf',[1 2 4;1 2 6;1 3 4;1 3 7;1 5 6;1 5 7;...
                           2 8 4;2 8 6;3 8 4;3 8 7;5 8 6;5 8 7],...
                'MeshCreator','FangQ','MeshTitle','T6 Cube',...
                'SpecialData',[nan, inf, -inf]);
       savejson(jsonmesh)
       savejson('jmesh',jsonmesh)
       savejson('',jsonmesh,'compact',1)
       savejson('jmesh',jsonmesh,'outputfile.json')
       savejson('',jsonmesh,'ArrayIndent',0,'FloatFormat','\t%.5g','FileName','outputfile2.json')
       savejson('cpxrand',eye(5)+1i*magic(5))
       savejson('ziparray',eye(10),'Compression','zlib','CompressArraySize',1)
       savejson('',jsonmesh,'ArrayToStruct',1)

loadjson.m

       loadjson('{}')
       dat=loadjson('{"obj":{"string":"value","array":[1,2,3]}}')
       dat=loadjson(['examples' filesep 'example1.json'])
       dat=loadjson(['examples' filesep 'example1.json'],'SimplifyCell',1)

saveubjson.m

       a={single(rand(2)), struct('va',1,'vb','string'), 1+2i};
       saveubjson(a)
       saveubjson('rootname',a,'testdata.ubj')
       saveubjson('zeros',zeros(100),'Compression','gzip')

loadubjson.m

       obj=struct('string','value','array',single([1 2 3]),'empty',[],'magic',uint8(magic(5)));
       ubjdata=saveubjson('obj',obj);
       dat=loadubjson(ubjdata)
       class(dat.obj.array)
       isequaln(obj,dat.obj)
       dat=loadubjson(saveubjson('',eye(10),'Compression','zlib','CompressArraySize',1))

jdataencode.m

      jd=jdataencode(struct('a',rand(5)+1i*rand(5),'b',[],'c',sparse(5,5)))
      savejson('',jd)

jdatadecode.m

      rawdata=struct('a',rand(5)+1i*rand(5),'b',[],'c',sparse(5,5));
      jd=jdataencode(rawdata)
      newjd=jdatadecode(jd)
      isequaln(newjd,rawdata)

examples

Under the "examples" folder, you can find several scripts to demonstrate the basic utilities of JSONLab. Running the "demo_jsonlab_basic.m" script, you will see the conversions from MATLAB data structure to JSON text and backward. In "jsonlab_selftest.m", we load complex JSON files downloaded from the Internet and validate the loadjson/savejson functions for regression testing purposes. Similarly, a "demo_ubjson_basic.m" script is provided to test the saveubjson and loadubjson functions for various matlab data structures.

Please run these examples and understand how JSONLab works before you use it to process your data.

4. Known Issues and TODOs

JSONLab has several known limitations. We are striving to make it more general and robust. Hopefully in a few future releases, the limitations become less.

Here are the known issues:

  1. 3D or higher dimensional cell/struct-arrays will be converted to 2D arrays;
  2. When processing names containing multi-byte characters, Octave and MATLAB can give different field-names; you can use feature('DefaultCharacterSet','latin1') in MATLAB to get consistant results
  3. savejson can only export the properties from MATLAB classes, but not the methods
  4. saveubjson converts a logical array into a uint8 ([U]) array
  5. a special N-D array format, as defined in the JData specification, is implemented in saveubjson. You may use saveubjson(...,'NestArray',1) to create UBJSON Draft-12 compliant files
  6. loadubjson can not parse all UBJSON Specification (Draft 9) compliant files, however, it can parse all UBJSON files produced by saveubjson.

5. Contribution and feedback

JSONLab is an open-source project. This means you can not only use it and modify it as you wish, but also you can contribute your changes back to JSONLab so that everyone else can enjoy the improvement. For anyone who want to contribute, please download JSONLab source code from its source code repositories by using the following command: 

      git clone https://github.com/fangq/jsonlab.git jsonlab

or browsing the github site at 

      https://github.com/fangq/jsonlab

Please report any bugs or issues to the below URL: 

      https://github.com/fangq/jsonlab/issues

Sometimes, you may find it is necessary to modify JSONLab to achieve your goals, or attempt to modify JSONLab functions to fix a bug that you have encountered. If you are happy with your changes and willing to share those changes to the upstream author, you are recommended to create a pull-request on github.

To create a pull-request, you first need to "fork" jsonlab on Github by clicking on the "fork" button on top-right of jsonlab's github page. Once you forked jsonlab to your own directory, you should then implement the changes in your own fork. After thoroughly testing it and you are confident the modification is complete and effective, you can then click on the "New pull request" button, and on the left, select fangq/jsonlab as the "base". Then type in the description of the changes. You are responsible to format the code updates using the same convention (tab-width: 8, indentation: 4 spaces) as the upstream code.

We appreciate any suggestions and feedbacks from you. Please use the following mailing list to report any questions you may have regarding JSONLab: 

      https://github.com/fangq/jsonlab/issues

(Subscription to the mailing list is needed in order to post messages).

6. Acknowledgement

This toolbox contains modified functions from the below toolboxes:

loadjson.m

The loadjson.m function was significantly modified from the earlier parsers (BSD 3-clause licensed) written by the below authors 

    created on 2009/11/02

    created on  2009/03/22
  • Joel Feenstra: 
    http://www.mathworks.com/matlabcentral/fileexchange/20565
    created on 2008/07/03

loadmsgpack.m

Copyright (c) 2014,2016 Bastian Bechtold All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of source code must retain the above copyright notice, this 
  list of conditions and the following disclaimer.

  • Redistributions in binary form must reproduce the above copyright notice, 
  this list of conditions and the following disclaimer in the documentation 
  and/or other materials provided with the distribution.

  • Neither the name of the copyright holder nor the names of its contributors 
  may be used to endorse or promote products derived from this software without 
  specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

zlibdecode.m, zlibencode.m, gzipencode.m, gzipdecode.m, base64encode.m, base64decode.m

Copyright (c) 2012, Kota Yamaguchi All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of source code must retain the above copyright notice, this 
  list of conditions and the following disclaimer.

  • Redistributions in binary form must reproduce the above copyright notice, 
  this list of conditions and the following disclaimer in the documentation
  and/or other materials provided with the distribution

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Powered by Habitat