File Exchange

image thumbnail

Highly portable JSON-input parser

version (3.21 KB) by Nedialko
For any version of Matlab that may still be in use by the fileexchange community.

1 Download

Updated 03 Nov 2009

View Version History

View License

The JSON quickly gains popularity for a good reason. To paraphrase Christophe Delord (
JSON is born from a simple statement:
A simple human-like meta-language will quickly
justify its use in such cases as data exchange over the web (e.g. Google translate), configuration files, etc.


for an example of AUTOMATIC, PORTABLE & REUSABLE creation of GUI sets for Matlab.
(Note: p_json.m is at the core of this example too)

Hence it is worth the while to provide this novel facility to the widest possible audience of Matlab users.

Based on work from:
& ::

F.Glineur 23393-another-json-parser: Faster, Clearer & More robust than #20565
J.Feenstra 20565-json-parser

This submission also provides a very simple (yet meaningful) use example (see also the screenshot).

Your test script (called a0t on the screenshot) may look like:

% ---------------------------------------------------------

fName = 'test02.jon';

fid = fopen(fName,'rt'); inString = fscanf(fid,'%c'); fclose(fid);

data = p_json(inString)

% ---------------------------------------------------------

and hence

inString == verbatim the 'test02.jon' TEXT input file,

provided in the .zip (see links below for more detail and more definitions):

% =========================================================
inString =

"name": {"type":"string"},
"age" : {"type":"integer",
"type" : "object"
"name" : "John Doe",
"age" : 30

% =========================================================

*** See also:

Given the topical nature of the JSO notation we invite the user to take a look at places like:

Cite As

Nedialko (2021). Highly portable JSON-input parser (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (4)

Ilya Belevich

Qianqian Fang

FYI, an optimized JSON parser, JSONlab, based on this work is available at

It is 10x to 100x faster. A JSON encoder is also provided in JSONlab.

Tanyer Alan


I kinda wish there were a more straight-forward way of accessing structs that live inside an array. For example, in a JSON string that has a series of unkeyed objects living in an array (i.e., "arrayName": [ {"color": "red"}, {"color": "green"}, {"color": "blue"} ]), accessing those objects is odd. Instead of using arrayName(i).color you must use arrayName{1,i}.color. Not a huge deal, just different than what I was expecting.

Other than that, it is very particular about properly-formatted JSON string inputs, and that's a good thing. Nice job.

MATLAB Release Compatibility
Created with R10
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!