Let’s say your web-application communicates with a 3rd party service and this service is somewhat fragile. Either because of its fragility – or for some other reason – under no circumstances do you want your application to bombard this service with concurrent requests. Whenever a component needs to make a request to this service, and there is already another request waiting for a response, the new request should wait until the previous request completes.
Continue reading
Back in September we blogged about the GA of Amazon’s Serverless Aurora and how it really simplifies scaling database servers in the cloud – mainly because.. well basically you don’t have to. At AWS re:Invent 2018 last month, Amazon has taken Aurora Serverless to the next level with its ‘Data API’.
If you care about Web Accessibility (I know you do) then some of the accessibility related debugging tools in the Chrome browser may come in handy.
As you know there are several ways to speed up your website like:
– HTTP/2
– Brotli compression
– Webp image format
– HTTP Caching
but they are still separate. There is no one ready solution that would unite them all.
Lets try to combine all we need in one place!
Sometimes we need to make calls to some RESTful APIs from an AWS Lamda function. Let’s say we use Node.js as our platform. On the surface, there are two ways to do it: Continue reading
Building custom themes for Drupal can be done with tons of approaches and many different ways of automating routine CSS operations. Here is the short list of tools I use with Gulp that I find solve close to 100% of the most common needs for Drupal (as well as for other CMS custom development). Feel free to share your thoughts on this list in the comments below. Continue reading
Chrome’s default tab page – that handful of thumbnail images of your most often visited websites – is useful to a certain extent only. There are easy ways to make that page more useful.
In Drupal 8, sometimes we have a lot of levels of taxonomy terms and we need to manage this in our code. For example, what if we need to get a specific level of the taxonomy?
So let’s imagine the following taxonomy tree:
DynamoDB is a non-blocking asynchronous storage, and it is great. Mostly. However, sometimes you just need to ensure that only one instance of script accesses a certain share of information in your table. In my case it was operation or purgin+rebuilding records, which under race conditions would result in a massive data loss.
So, I decided to build a semaphore. Continue reading
