URL Rewriting in Cloudfront with Lambda@Edge functions

So I previously put a Lambda@Edge function in place to fix the \blog\post\ requests to become \blog\post\index.html which is what the files are, and I also put a 403/404 error handler page to catch those 404 and log then to Analytics events, and sure enough, there where a number of pages that had been have errors.

I fixed the internal page links, assuming those where the primary dirvers, and moved on.

Those failures just keep happening though

So in the weekend I rewrote the lambda@edge function from

function handler(event) {
    var request = event.request;
    var uri = request.uri;
    
    // Check whether the URI is missing a file name.
    if (uri.endsWith('/')) {
        request.uri += 'index.html';
    } 
    // Check whether the URI is missing a file extension.
    else if (!uri.includes('.')) {
        request.uri += '/index.html';
    }

    return request;
}

to a new flashy one, that has a list of urls to rewrite. I was originally doing to use a javascript Set but those are not supported, so went for a dictionary.

'use strict';

var patch_url = 'https://simeonpilgrim.com/nikon-patch/nikon-patch.html';
var patch_response = {
	statusCode: 302,
	statusDescription: 'Found',
	headers: 
		{ "location": { "value": patch_url } }
};

var patch_redirect_set = {'/nikon-patch/Nikon-Patch.html': true,
	'/nikon-patch/nikon-patch-beta.html': true,
	'/nikon-patch-test/index.html': true,
	'/nikon-patch': true,
	'/nikon-patch/': true,
	};

function handler(event) {
    var request = event.request;
    var uri = request.uri;
    
    if( patch_redirect_set[uri] ){
		return patch_response;
	}
	
	if( patch_redirect_set[uri] ){
		return patch_response;
	}
    // Check whether the URI is missing a file name.
    if (uri.endsWith('/')) {
        request.uri += 'index.html';
    } 
    // Check whether the URI is missing a file extension.
    else if (!uri.includes('.')) {
        request.uri += '/index.html';
    }
    
    return request;
}

I rather like the Test page on the lambda pages, so once it was not failing, I published it, and tested. And it worked, so there we are. A framework to fix urls. It might make more sense to put the mapping in the dictionary, and if I end up wanting to fix non-patch pages I might go that route. But it’s more less wrong. So that’s a win.