{"id":8017,"date":"2016-11-11T09:00:46","date_gmt":"2016-11-11T14:00:46","guid":{"rendered":"https:\/\/blogs.mathworks.com\/pick\/?p=8017"},"modified":"2016-11-09T01:05:05","modified_gmt":"2016-11-09T06:05:05","slug":"r2016b-features-markerindices-jsonencode-jsondecode","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/pick\/2016\/11\/11\/r2016b-features-markerindices-jsonencode-jsondecode\/","title":{"rendered":"R2016b Features: MarkerIndices, jsonencode, jsondecode"},"content":{"rendered":"
\n

Jiro<\/a>‘s picks this week are two of the many new features of R2016b: MarkerIndices<\/tt> property for specifying marker locations<\/a> and jsonencode<\/tt><\/a>\/ jsondecode<\/tt><\/a>.<\/p>\n

In the past, I’ve highlighted a couple of new product features which were related to previous Picks: “Interactive Legend in R 2016a”<\/a> and “Does your origin go through zero?”<\/a>. This week’s post is of the same nature.<\/p>\n

MarkerIndices<\/tt><\/b><\/p>\n

Back in 2013, I posted this blog on creating line plots with fewer markers<\/a>. This is especially useful when you have a dense (high-sampled) signal that you want to plot with markers. In R2016b, you can use the new MarkerIndices<\/tt> property to specify which points to add a marker to.<\/p>\n

For example, if you want 10 evenly-spaced markers, you can do the following.<\/p>\n

t  = 0:0.005:pi;\nplot(t, sin(3*t).*cos(t\/2), 'p-'<\/span>, 'MarkerIndices'<\/span>, round(linspace(1,length(t),10)))\n<\/pre>\n
\"\" <\/p>\n
jsonencode<\/tt> \/ jsondecode<\/tt><\/b><\/p>\n
In April 2015, I posted another blog on parsing JSON files<\/a>. In R2016b, we now have jsonencode<\/tt> and jsondecode<\/tt>. I’ll use the same example I used in the 2015 blog post. Let’s say we have example.json<\/tt> the has the following content.<\/p>\n
{\n    \"firstName\": \"John\",\n    \"lastName\": \"Smith\",\n    \"age\": 25,\n    \"address\":\n    {\n        \"streetAddress\": \"3 Apple Hill Dr\",\n        \"city\": \"Natick\",\n        \"state\": \"MA\",\n        \"postalCode\": \"01760\"\n    },\n    \"phoneNumber\":\n    [\n        {\n          \"type\": \"home\",\n          \"number\": \"123 456 7890\"\n        },\n        {\n          \"type\": \"cell\",\n          \"number\": \"098 765 4321\"\n        }\n    ]\n}<\/pre>\n
jsondecode<\/tt> works on character arrays, so we will first read in the text data.<\/p>\n
str = fileread('example.json'<\/span>);\n<\/pre>\n
Then, simply call jsondecode<\/tt>.<\/p>\n
data = jsondecode(str)\n<\/pre>\n
data = \n  struct with fields:\n\n      firstName: 'John'\n       lastName: 'Smith'\n            age: 25\n        address: [1×1 struct]\n    phoneNumber: [2×1 struct]\n<\/pre>\n
The output is a structure.<\/p>\n
disp(data.address)\n<\/pre>\n
    streetAddress: '3 Apple Hill Dr'\n             city: 'Natick'\n            state: 'MA'\n       postalCode: '01760'\n<\/pre>\n
disp(data.phoneNumber(1))\n<\/pre>\n
      type: 'home'\n    number: '123 456 7890'\n<\/pre>\n
Comments<\/b><\/p>\n
Give these new features a try. If you’re using pre-R2016b MATLAB, be sure to check out the blog posts I mentioned above. Let us know what you think here<\/a>.<\/p>\n