Lessons Learned from Effective Go, Part 2

Reading time ~2 minutes

This second post on Effective Go, captures my notes on reading the second half of the official documentation (from initialization through panic and recovery).

Notes on the first half here.

More Takeaways from Effective Go

Declaring Constants and Setting State

  • enum groups related constants together in one type - timezones, weekdays, etc. (more here)
  • iota automates enum values with a numeric universal counter (more here).
  • Use iota + 1 to be sure that the enum type is initialized.

     const (
         Sleeping = iota + 1
         Walking
         Running
     )
    
     func main() {
         var activity Activity
         // activity will be zero,
         // so it's not initialized
    
         activity = Sleeping
         // now you know that it's been
         // initialized
     }
    
  • Unknown state pattern: Start with an Unknown to be sure of enum’s initialization.

     const (
         Unknown = iota
         Sleeping
         Walking
         Running
     )
    
  • Use init (no args) in a source file to set up whatever state is required

Pointers

  • A pointer holds the memory address of a value.
  • Type *T is a pointer to a T value.
  • The & operator generates a pointer to its operand.
  • The * operator denotes the pointer’s underlying value.
  • pointers vs. values
    • for receivers, value methods can be invoked on pointers and values, but pointer methods can only be invoked on pointers.
  • More here.

Interfaces

  • Interfaces in Go provide a way to specify the behavior of an object; if the object can do X, then it can be used in Y.
  • For example, if you create a type Z and want to be able to sort it, you have to implement each of the required functions for the sort interface; telling sort what len, less, and swap mean in the context of a Z.
  • A type can implement multiple interfaces.

Blank Identifier in Multiple Assignment

  • Always check error returns; it is bad practice to discard the error value in order to ignore the error.

Concurrency

Do not communicate by sharing memory; instead, share memory by communicating.

  • A goroutine is a function executing concurrently with other goroutines in the same address space.
  • Function literals make sure the variables referred to by the function survive as long as they are active.
  • Unbuffered channels give a way to exchange values while guaranteeing that two goroutines are in a known state.
  • Receivers block until there is data to receive. If the channel is unbuffered, the sender blocks until the receiver has received the value.
  • If the channel is buffered, the sender blocks only until the value has been copied to the buffer; if the buffer is full, this means waiting until some receiver has retrieved a value.
  • If the calculation can be broken into separate pieces that can execute independently, it can be parallelized, with a channel to signal when each piece completes.

Errors

  • The built-in function panic creates a run-time error that will stop the program.
  • Real library functions should avoid panic. If the problem can be masked or worked around, it’s always better to let things continue to run rather than taking down the whole program.
  • A call to recover stops the unwinding and returns the argument passed to panic.
  • recover always returns nil unless called directly from a deferred function; deferred code can call library routines that themselves use panic, and recover without failing.

A Parrot Trainer Eats Crow

In this post, we'll consider how it is that models trained on massive datasets using millions of parameters can be both "low bias" and al...… Continue reading

Embedded Binaries for Go

Published on February 06, 2021