I need to parse the following string:
"Apple-sweet#tangy@Bannana@Orange-citrusy@Pear-crispy#green/yellow"
Each item except for the first that starts with "@" is the key, "-" denotes the start of subitems for that key. "#" separates each subitem with that same key.
Into an object similar to this:
{
"Apple": ["sweet", "tangy"],
"Bannana":[],
"Orange": ["citrusy"],
"Pear": ["crispy", "green/yellow"]
}
What is the best way to parse this?
You need to split the string by the guard symbols several times:
var str = 'Apple-sweet#tangy@Bannana@Orange-citrusy@Pear-crispy#green/yellow';
var result = {};
str.split('@').forEach(function(item){
var split = item.split('-');
result[split[0]] = split[1] && split[1].split('#') || [];
});
console.dir(result);
First you split by @ to get each item from the object as a separate array value. Then you split each of the items in that array by - to get the key and the values. After that you need to make sure the values exist and if they do, split them by # to get an array of them.
You may try this (Example Here)
var str = "Apple-sweet#tangy@Bannana@Orange#citrusy@Pear#crispy#green/yellow";
var arr = str.split('@'), i = 0, l= arr.length, obj = {} ;
for(; i< l; i++){
var a = arr[i].split(/-|#/);
obj[a[0]] = a.splice(1);
}
console.log(obj);
Also, you can use (to make your keys lowercase, so you can use apple instead of Apple)
obj[a[0].toLowerCase()] = a.splice(1);
You just need a lot of string manipulation functions
var inputString = "Apple-sweet#tangy@Bannana@Orange-citrusy@Pear-crispy#green/yellow";
var keyValues = inputString.split("@");
var obj = {};
for (var i = 0; i < keyValues.length; ++i) {
var index = keyValues[i].indexOf("-");
if (index == -1) {
obj[keyValues[i]] = [];
} else {
var key = keyValues[i].substring(0, index);
var values = keyValues[i].substring(index + 1).split("#");
obj[key] = values;
}
}
Just for fun, I tried to implement your "grammar" within PEG.js
Go there http://pegjs.majda.cz/online and paste the following grammar on the grammar part (1).
start
= fruits
fruits
= ft:fruit "@" fts:fruits {fts[ft[0]] = ft[1]; return fts;}
/ ft:fruit {var fruits = {}; fruits[ft[0]] = ft[1]; return fruits;}
fruit
= name:word "-" flavs:flavors {var ft = []; ft.push(name); ft.push(flavs); return ft;}
/ name:word {var ft = []; ft.push(name); ft.push([]); return ft;}
flavors
= flavor:word "#" flavs:flavors {flavs.unshift(flavor); return flavs;}
/ flavor:word { var tab = []; tab.push(flavor); return tab;}
word
= value:[a-zA-Z/]+ {return value.join("");}
Then put the sentence in the input section (2)
Apple-sweet#tangy@Bannana@Orange-citrusy@Pear-crispy#green/yellow
And here comes the result :
{
"Pear": [
"crispy",
"green/yellow"
],
"Orange": [
"citrusy"
],
"Bannana": [],
"Apple": [
"sweet",
"tangy"
]
}
Then embed the resulting javascript parser into a file you won't have to bother about the lexical/syntaxic analysis job. Better maintenability to come only working at semantic level IMHO :)
Quite powerful API, although in this particular case it is a bit overkill ;)
