Parsing potentially nonexistant JSON fields in JavaScript

Context: I'm currently toying with the Google Geocode API to break down address strings into a more machine-friendly format. The problem is that neither the user input or API output are 100% consistent, and I would receive less entries in the JSON response than expected. For example, requesting "Times Square NY" gives me 6 address components

"address_components" : [
        {
           "long_name" : "Manhattan",
           "short_name" : "Manhattan",
           "types" : [ "political", "sublocality", "sublocality_level_1" ]
        },
        {
           "long_name" : "New York",
           "short_name" : "New York",
           "types" : [ "locality", "political" ]
        },
        {
           "long_name" : "New York County",
           "short_name" : "New York County",
           "types" : [ "administrative_area_level_2", "political" ]
        },
        {
           "long_name" : "New York",
           "short_name" : "NY",
           "types" : [ "administrative_area_level_1", "political" ]
        },
        {
           "long_name" : "United States",
           "short_name" : "US",
           "types" : [ "country", "political" ]
        },
        {
           "long_name" : "10036",
           "short_name" : "10036",
           "types" : [ "postal_code" ]
        }

But, for some reason, "Times Circle NY" gives me 8 address components (notice, for example, how "Times Square NY" does not contain a route entry):

{
           "long_name" : "10",
           "short_name" : "10",
           "types" : [ "street_number" ]
        },
        {
           "long_name" : "Columbus Circle",
           "short_name" : "Columbus Cir",
           "types" : [ "route" ]
        },
        {
           "long_name" : "Manhattan",
           "short_name" : "Manhattan",
           "types" : [ "political", "sublocality", "sublocality_level_1" ]
        },
        {
           "long_name" : "New York",
           "short_name" : "New York",
           "types" : [ "locality", "political" ]
        },
        {
           "long_name" : "New York County",
           "short_name" : "New York County",
           "types" : [ "administrative_area_level_2", "political" ]
        },
        {
           "long_name" : "New York",
           "short_name" : "NY",
           "types" : [ "administrative_area_level_1", "political" ]
        },
        {
           "long_name" : "United States",
           "short_name" : "US",
           "types" : [ "country", "political" ]
        },
        {
           "long_name" : "10019",
           "short_name" : "10019",
           "types" : [ "postal_code" ]
        }

My code: To work with the JSON in a more reasonable format, for ex. getting rid of the short_name or accessing a field by component type, I've written an object constructor that goes like this:

class Address {
constructor(googleJson) {
    try {
        this.state= googleJson["results"][0]["address_components"].filter(el => JSON.stringify(el["types"]) === JSON.stringify(["administrative_area_level_1", "political"]))[0]["short_name"];
    }
    catch (e) {
        this.state= "";
    }
    try {
        this.city= googleJson["results"][0]["address_components"].filter(el => JSON.stringify(el["types"]) === JSON.stringify(["administrative_area_level_2", "political"]))[0]["long_name"];
    }
    catch (e) {
        this.city= "";
    }
    try {
        this.lat = googleJson["results"][0]["geometry"]["location"]["lat"].toString();
    }
    catch (e) {
        this.lat = "";
    }
    // And so on...

And it works perfectly fine, does its job, but half of my code is just these try/catch pairs. Is there any smarter way to attempt to read multiple potentially empty JSON fields without so much code repetition?

Also, since I'm here, I found the filter I wrote inside each try/catch to be quite legible, but I fear the constructor as a whole would be \$O(n \times m)\$. What other way could I read the JSON in \$O(n + m)\$ and maintain legibility?