Dexter Legaspi dalegaspi

Homebrew and Symlink Issues

if you see this:

Error: An unexpected error occurred during the `brew link` step
The formula built, but is not symlinked into /usr/local
Permission denied @ dir_s_mkdir - /usr/local/Frameworks
Error: Permission denied @ dir_s_mkdir - /usr/local/Frameworks

As my first major project using Scala hits production without too much of a hitch, I present a small list of notes that summarizes my experience with the language:

Learn to embrace functional programming. I learned this early on that while Scala is able handle object-oriented programming quite well, it's built with functional programming in mind. This means learning flatMap/map, immutables, et cetera.
Learn the thin cake pattern.
For every XML library available, there are about 3,781 JSON libraries available. If you're using XML a lot in your project, I feel that Java 8/10 is still easier.
Jackson is still king of JSON serialization. Is it the fastest? No. Is it the leanest? Not by a long shot. We used a lot of POJOs that rely on serialization, and we've tried a lot of JSON libraries, and while others are faster, they have soul-crushing limitations that were not present with Jackson.
blocking > no

Preface

Listen, if you want to use Zuul, the easiest path is just use the Spring Boot version. I'm not exactly a fan of Spring in general, but this is the only solution available that's the least painful. Recent release of Spring Cloud Gateway looks promising but I'm satisfied so far with Spring Boot + Zuul that I'm not that interested in evaluating Spring Cloud Gateway,

We are also talking about Zuul 1.x here. Zuul 2.x is a whole can of worms I'd rather not rant about here.

This applies to Finchley (2.x) from SR1 release (may apply to other releases but not guaranteed)

Re: using logback. The documentation says it's recommended to use logback-spring.xml in your classpath, but it doesn't work. You should actually use logback.xml or your logging configuration is not recognized at all. While at the subject, it is recommended not to use asynchronous logging at all. U

Moving From IntelliJ to Emacs ENSIME for Scala Development on MacOS

Preface

I have been dependent on the use of IntelliJ for most of my development tasks but unfortunately it has given me an abundant source of frustration: despite its recent updates and fixes, it has become progressively slow and a resource hog even with conservative use of plugins and fancy features, which sometimes gets in the way of productivity.

Over the last few months, I have been purposely avoiding IntelliJ for some simple code edits if I don't have to.

Now I wouldn't exactly say Emacs with ENSIME is walk in a park--far from it. If you've been spoiled with the convenience of what IntelliJ provides out of the box, I would say switching to a mostly text editor would be a challenging endeavor.

UPDATE December 2022

**Unless you have an older Apple TV 4K and/or a TV that's not manufactured from the last 5 years or so, do NOT follow the instructions below.

With the newer 4K TVs and especially paired with the newer generation of AppleTV, leave the Dolby Vision ON and leave Frame Rate Matching OFF or you will have audio sync issues. Newer TVs are already smart enough to detect frame rate so no need to do any of the instructions below.**

Prologue

The most prevalent issues as of this writing when using an Apple TV 4K with an A/V receiver:

	import java.io.{ByteArrayInputStream, ByteArrayOutputStream, ObjectInputStream, ObjectOutputStream}
	import java.util.zip.{GZIPInputStream, GZIPOutputStream}
	import scala.util.Try

	/**
	* aped from https://gist.github.com/owainlewis/1e7d1e68a6818ee4d50e (gzip compression)
	* and from https://stackoverflow.com/a/39371571/918858 (ser/deser from Array[Byte])
	*/
	object Gzip {

	import com.typesafe.scalalogging.StrictLogging
	import monix.eval.Task
	import monix.execution.Scheduler

	import scala.concurrent.Future

	object Stopwatch extends StrictLogging {

	/**
	* http://biercoff.com/easily-measuring-code-execution-time-in-scala/

	import java.io.{ByteArrayInputStream, ByteArrayOutputStream, ObjectInputStream, ObjectOutputStream}
	import java.util.zip.{GZIPInputStream, GZIPOutputStream}

	import scala.util.Try

	/**
	* aped from https://gist.github.com/owainlewis/1e7d1e68a6818ee4d50e (gzip compression)
	* and from https://stackoverflow.com/a/39371571/918858 (ser/deser from Array[Byte])
	*/
	object CompressionUtils {

	<?xml version="1.0" encoding="UTF-8"?>
	<configuration>
	<property name="LOGS" value="./logs" />

	<appender name="Console"
	class="ch.qos.logback.core.ConsoleAppender">
	<encoder class="ch.qos.logback.classic.encoder.PatternLayoutEncoder">
	<pattern>%black(%d{ISO8601}) %highlight(%-5level) [%blue(%t)] [%blue(%contextName)] %yellow(%logger{36}): %msg%n%throwable</pattern>
	</encoder>
	</appender>

	import akka.http.scaladsl.marshalling.{Marshaller, ToEntityMarshaller}
	import akka.http.scaladsl.model.MediaType.{NotCompressible, applicationBinary}
	import akka.http.scaladsl.model.{ContentTypeRange, MediaType}
	import akka.http.scaladsl.unmarshalling.{FromEntityUnmarshaller, Unmarshaller}
	import pbdirect._

	/**
	* This allows you to have a mixin trait for use with Akka HTTP endpoints that accept/return Protobuf.
	* This has been patterned after https://github.com/hseeberger/akka-http-json and uses
	* https://github.com/btlines/pbdirect for "proto-less" (i.e. no .proto schemas) serialization/