This is How to Build Haskell with GNU Make (and why it's worth trying)

In a previous article I showed the GHC commands you need to compile a basic Haskell executable without explicitly using the source files from its dependencies. But when you're writing your own Haskell code, 99% of the time you want to be using a Haskell build system like Stack or Cabal for your compilation needs instead of writing your own GHC commands. (And you can learn how to use Stack in my new free course, Setup.hs).

But part of my motivation for solving that problem was that I wanted to try an interesting experiment:

How can I build my Haskell code using GNU Make?

GNU Make is a generic build system that allows you to specify components of your project, map out their dependencies, and dictate how your build artifacts are generated and run.

I wanted to structure my source code the same way I would in a Cabal-style application, but rely on GNU Make to chain together the necessary GHC compilation commands. I did this to help gain a deeper understanding of how a Haskell build system could work under the hood.

In a Haskell project, we map out our project structure in the .cabal file. When we use GNU Make, our project is mapped out in the makefile. Here's the Makefile we'll ultimately be constructing:

GHC = ~/.ghcup/ghc/9.2.5/bin/ghc
BIN = ./bin
EXE = ${BIN}/hello

LIB_DIR = ${BIN}/lib
SRCS = $(wildcard src/*.hs)
LIB_OBJS = $(wildcard ${LIB_DIR}/*.o)

library: ${SRCS}
  @mkdir -p ${LIB_DIR}
  @${GHC} ${SRCS} -hidir ${LIB_DIR} -odir ${LIB_DIR}

generate_run: app/Main.hs library
  @mkdir -p ${BIN}
  @cp ${LIB_DIR}/*.hi ${BIN}
  @${GHC} -i${BIN} -c app/Main.hs -hidir ${BIN} -odir ${BIN}
  @${GHC} ${BIN}/Main.o ${LIB_OBJS} -o ${EXE}

run: generate_run
  @${EXE}

TEST_DIR = ${BIN}/test
TEST_EXE = ${TEST_DIR}/run_test

generate_test: test/Spec.hs library
  @mkdir -p ${TEST_DIR}
  @cp ${LIB_DIR}/*.hi ${TEST_DIR}
  @${GHC} -i${TEST_DIR} -c test/Spec.hs -hidir ${TEST_DIR} -odir ${TEST_DIR}
  @${GHC} ${TEST_DIR}/Main.o ${LIB_OBJS} -o ${TEST_EXE}

test: generate_test
  @${TEST_EXE}

clean:
  rm -rf ./bin

Over the course of this article, we'll build up this solution piece-by-piece. But first, let's understand exactly what Haskell code we're trying to build.

Our Source Code

We want to lay out our files like this, separating our source code (/src directory), from our executable code (/app) and our testing code (/test):

.
├── app
│   └── Main.hs
├── makefile
├── src
│   ├── MyFunction.hs
│   └── TryStrings.hs
└── test
    └── Spec.hs

Here's the source code for our three primary files:

-- src/MyStrings.hs
module MyStrings where

greeting :: String
greeting = "Hello"

-- src/MyFunction.hs
module MyFunction where

modifyString :: String -> String
modifyString x = base <> " " <> base
  where
    base = tail x <> [head x]

-- app/Main.hs
module Main where

import MyStrings (greeting)
import MyFunction (modifyString)

main :: IO ()
main = putStrLn (modifyString greeting)

And here's what our simple "Spec" test looks like. It doesn't use a testing library, it just prints different messages depending on whether or not we get the expected output from modifyString.

-- test/Spec.hs
module Main where

import MyFunction (modifyString)

main :: IO ()
main = do
  test "abcd" "bcda bcda"
  test "Hello" "elloH elloH"

test :: String -> String -> IO ()
test input expected = do
  let actual = modifyString input
  putStr $ "Testing case: " <> input <> ": "
  if expected /= actual
    then putStrLn $ "Incorrect result! Expected: " <> expected <> " Actual: " <> actual
    else putStrLn "Correct!"

The files are laid out the way we would expect for a basic Haskell application. We have our "library" code in the src directory. We have a single "executable" in the app directory. And we have a single "test suite" in the test directory. Instead of having a Project.cabal file at the root of our project, we'll have our makefile. (At the end, we'll actually compare our Makefile with an equivalent .cabal file).

But what does the Makefile look like? Well it would be overwhelming to construct it all at once. Let's begin slowly by treating our executable as a single file application.

Running a Single File Application

So for now, let's adjust Main.hs so it's an independent file without any dependencies on our library modules:

-- app/Main.hs
module Main where

main :: IO ()
main = putStrLn "Hello"

The simplest way to run this file is runghc. So let's create our first makefile rule that will do this. A rule has a name, a set of prerequisites, and then a set of commands to run. We'll call our rule run, and have it use runghc on app/Main.hs. We'll also include the app/Main.hs as a prerequisite, since the rule will run differently if that file changes.

run: app/Main.hs
  runghc app/Main.hs

And now we can run this run using make run, and it will work!

$ make run
runghc app/Main.hs
Hello

Notice that it prints the command we're running. We can change this by using the @ symbol in front of the command in our Makefile. We'll do this with almost all our commands:

run: app/Main.hs
  @runghc app/Main.hs

And it now runs our application without printing the command.

Using runghc is convenient, but if we want to use dependencies from different directories, we'll eventually need to use multiple stages of compilation. So we'll want to create two distinct rules. One that generates the executable using ghc, and another that actually runs the generated executable.

So let's create a generate_run rule that will produce the build artifacts, and then run will use them.

generate_run: app/Main.hs
  @ghc app/Main.hs

run: generate_run
  @./app/Main

Notice that run can depend on generate_run as a prerequisite, instead of the source file now. This also generates three build artifacts directly in our app directory: the interface file Main.hi, the object file Main.o, and the executable Main.

It's bad practice to mix build artifacts with source files, so let's use GHC's arguments (-hidir, -odir and -o) to store these artifacts in their own directory called bin.

generate_run: app/Main.hs
  @mkdir -p ./bin
  @ghc app/Main.hs -hidir ./bin -odir ./bin -o ./bin/hello

run: generate_run
  @./bin/hello

We can then add a third rule to "clean" our project. This would remove all binary files so that we can do a fresh recompilation if we want.

clean:
  rm -rf ./bin

For one final flourish in this section, we can use some variables. We can make one for the GHC compiler, referencing its absolute path instead of a symlink. This would make it easy to switch out the version if we wanted. We'll also add a variable for our bin directory and the hello executable, since these are used multiple times.

# Could easily switch versions if desired
# e.g. GHC = ~/.ghcup/ghc/9.4.4/bin/ghc
GHC = ~/.ghcup/ghc/9.2.5/bin/ghc
BIN = ./bin
EXE = ${BIN}/hello

generate_run: app/Main.hs
  @mkdir -p ${BIN}
  @${GHC} app/Main.hs -hidir ${BIN} -odir ${BIN} -o ${EXE}

run: generate_run
  @${EXE}

clean:
  rm -rf ./bin

And all this still works as expected!

$ generate_run
[1 of 1] Compiling Main (app/Main.hs, bin/Main.o)
Linking ./bin/hello
$ make run
Hello
$ make clean
rm -rf ./bin

So we have some basic rules for our executable. But remember our goal is to depend on a library. So let's add a new rule to generate the library objects.

Generating a Library

For this step, we would like to compile src/MyStrings.hs and src/MyFunction.hs. Each of these will generate an interface file (.hi) and an object file (.o). We want to place these artifacts in a specific library directory within our bin folder.

We'll do this by means of a new rule, library, which will use our two source files as its prerequisites. It will start by creating the library artifacts directory:

LIB_DIR = ${BIN}/lib

library: src/MyStrings.hs src/MyFunction.hs
  @mkdir -p ${LIB_DIR}
  ...

But now the only thing we have to do is use GHC on both of our source files, using LIB_DIR as the destination point.

LIB_DIR = ${BIN}/lib

library: src/MyStrings.hs src/MyFunction.hs
  @mkdir -p ${LIB_DIR}
  @ghc src/MyStrings.hs src/MyFunction.hs -hidir ${LIB_DIR} -odir ${LIB_DIR}

Now when we run the target, we'll see that it produces the desired files:

$ make library
$ ls ./bin/lib
MyFunction.hi MyFunction.o MyStrings.hi MyStrings.o

Right now though, if we added a new source file, we'd have to modify the rule in two places. We can fix this by adding a variable that uses wildcard to match all our source files in the directory (src/*.hs).

LIB_DIR = ${BIN}/lib
SRCS = $(wildcard src/*.hs)

library: ${SRCS}
  @mkdir -p ${LIB_DIR}
  @${GHC} ${SRCS} -hidir ${LIB_DIR} -odir ${LIB_DIR}

While we're learning about wildcard, let's make another variable to capture all the produced object files. We'll use this in the next section.

LIB_OBJS = $(wildcard ${LIB_DIR}/*.o)

So great! We're producing our library artifacts. How do we use them?

Linking the Library

In this section, we'll link our library code with our executable. We'll begin by assuming our Main file has gone back to its original form with imports, instead of the simplified form:

-- app/Main.hs
module Main where

import MyStrings (greeting)
import MyFunction (modifyString)

main :: IO ()
main = putStrLn (modifyString greeting)

We when try to generate_run, compilation fails because it cannot find the modules we're trying to import:

$ make generate_run
...
Could not find module 'MyStrings'
...
Could not find module 'MyFunction'

As we went over in the previous article, the general approach to compiling the Main module with its dependencies has two steps:

1. Compile with the -c option (to stop before the linking stage) using -i to point to a directory containing the interface files.

2. Compile the generated Main.o object file together with the library .o files to produce the executable.

So we'll be modifying our generate_main rule with some extra steps. First of course, it must now depend on the library rule. Then our first new command will be to copy the .hi files from the lib directory into the top-level bin directory.

generate_run: app/Main.hs library
  @mkdir -p ./bin
  @cp ${LIB_DIR}/*.hi ${BIN}
  ...

We could have avoided this step by generating the library artifacts in bin directly. I wanted to have a separate location for all of them though. And while there may be some way to direct the next command to find the headers in the lib directory, none of the obvious ways worked for me.

Regardless, our next step will be to modify the ghc call in this rule to use the -c and -i arguments. The rest stays the same:

generate_run: app/Main.hs library
  @mkdir -p ./bin
  @cp ${LIB_DIR}/*.hi ${BIN}
  @${GHC} -i${BIN} -c app/Main.hs -hidir ${BIN} -odir ${BIN}
  ...

Finally, we invoke our final ghc call, linking the .o files together. At the command line, this would look like:

$ ghc ./bin/Main.o ./bin/lib/MyStrings.o ./bin/lib/MyFunction.o -o ./bin/hello

Recalling our LIB_OBJS variable from up above, we can fill in the rule in our Makefile like so:

LIB_OBJS = $(wildcard ${LIB_DIR}/*.o)

generate_run: app/Main.hs library
  @mkdir -p ./bin
  @cp ${LIB_DIR}/*.hi ${BIN}
  @${GHC} -i${BIN} -c app/Main.hs -hidir ${BIN} -odir ${BIN}
  @${GHC} ${BIN}/Main.o ${LIB_OBJS} -o ${EXE}

And now our program will work as expected! We can clean it and jump straight to the make run rule, since this will run its prerequisites make library and make generate_run automatically.

$ make clean
rm -rf ./bin
$ make run
[1 of 2] Compiling MyFunction (src/MyFunction.hs, bin/lib/MyFunction.o)
[2 of 2] Compiling MyStrings (src/MyStrings.hs, bin/lib/MyStrings.o)
elloH elloH

So we've covered the library and an executable, but most Haskell projects have at least one test suite. So how would we implement that?

Adding a Test Suite

Well, a test suite is basically just a special executable. So we'll make another pair of rules, generate_test and test, that will mimic generate_run and run. Very little changes, except that we'll make another special directory within bin for our test artifacts.

TEST_DIR = ${BIN}/test
TEST_EXE = ${TEST_DIR}/run_test

generate_test: test/Spec.hs library
  @mkdir -p ${TEST_DIR}
  @cp ${LIB_DIR}/*.hi ${TEST_DIR}
  @${GHC} -i${TEST_DIR} -c test/Spec.hs -hidir ${TEST_DIR} -odir ${TEST_DIR}
  @${GHC} ${TEST_DIR}/Main.o ${LIB_OBJS} -o ${TEST_EXE}

test: generate_test
  @${TEST_EXE}

Of note here is that at the final step, we're still using Main.o instead of Spec.o. Since it's an executable module, it also compiles as Main.

But we can then use this to run our tests!

$ make clean
$ make test
[1 of 2] Compiling MyFunction (src/MyFunction.hs, bin/lib/MyFunction.o)
[2 of 2] Compiling MyStrings (src/MyStrings.hs, bin/lib/MyStrings.o)
Testing case: abcd: Correct!
Testing case: Hello: Correct!

So now we have all the different components we'd expect in a normal Haskell project. So it's interesting to consider how our makefile definition would compare against an equivalent .cabal file for this project.

Comparing to a Cabal File

Suppose we want to call our project HaskellMake and store its configuration in HaskellMake.cabal. We'd start our Cabal file with four metadata lines:

cabal-version: 1.12
name: HaskellMake
version: 0.1.0.0
build-type: Simple

Now our library would expose its two modules, using the src directory as its root. The only "dependency" is the Haskell base packages. Finally, default-language is a required field.

library
  exposed-modules:
      MyStrings
    , MyFunction
  hs-source-dirs:
      src
  build-depends:
      base
  default-language: Haskell2010

The executable would similarly describe where the files are located and state a base dependency as well as a dependency on the library itself.

executable hello
  main-is: Main.hs
  hs-source-dirs:
      app
  build-depends:
      base
    , HaskellMake
  default-language: Haskell2010

Finally, our test suite would look very similar to the executable, just with a different directory and filename.

test-suite make-test
  type: exitcode-stdio-1.0
  main-is: Spec.hs
  hs-source-dirs:
      test
  build-depends:
      base
    , HaskellMake
  default-language: Haskell2010

And, if we add a bit more boilerplate, we could actually then compile our code with Stack! First we need a stack.yaml specifying the resolver and the package location:

# stack.yaml
resolver: lts-20.12
packages:
  - .

Then we need Setup.hs:

-- Setup.hs

import Distribution.Simple
main = defaultMain

And now we could actually run our code!

$ stack build
$ stack exec hello
elloH elloH
$ stack test
Testing case: abcd: Correct!
Testing case: Hello: Correct!

Now observant viewers will note that we don't use any Hackage dependencies in our code - only base, which GHC always knows how to find. It would require a lot of work for us to replicate dependency management. We could download a .zip file with curl easily enough, but tracking the whole dependency tree would be extremely difficult.

And indeed, many engineers have spent a lot of time getting this process to work well with Stack and Cabal! So while it would be a useful exercise to try to do this manually with a simple dependency, I'll leave that for a future article.

When comparing the two file definitions, Undoubtedly, the .cabal definition is more concise and human readable, but it hides a lot of implementation details. Most of the time, this is a good thing! This is exactly what we expect from tools in general; they should allow us to work more quickly without having to worry about details.

But there are times where we might, on our time, want to occasionally try out a more adventurous path like we've done in this article that avoids relying too much on modern tooling. So why was this article a "useful exercise"™?

What's the Point?

So obviously, there's no chance this Makefile approach is suddenly going to supplant Cabal and Stack for building Haskell projects. Stack and Cabal are "better" for Haskell precisely because they account for the intricacies of Haskell development. In fact, by their design, GHC and Cabal both already incorporate some key ideas and features from GNU Make, especially with avoiding re-work through dependency calculation.

But there's a lot you can learn by trying this kind of exercise.

First of all, we learned about GNU Make. This tool can be very useful if you're constructing a program that combines components from different languages and systems. You could even build your Haskell code with Stack, but combine it with something else in a makefile.

A case and point for this is my recent work with Haskell and AWS. The commands for creating a docker image, authenticating to AWS and deploying it are lengthy and difficult to remember. A makefile can, at the very least, serve as a rudimentary aliasing tool. You could run make deploy and have it automatically rebuild your changes into a Docker image and deploy that to your server.

But beyond this, it's important to take time to deliberately understand how our tools work. Stack and Cabal are great tools. But if they seem like black magic to you, then it can be a good idea to spend some time understanding what is happening at an internal level - like how GHC is being used under the hood to create our build artifacts.

Most of the fun in programming comes in effectively applying tools to create useful programs quickly. But if you ever want to make good tools in the future, you have to understand what's happening at a deeper level! At least a couple times a year, you should strive to go one level deeper in your understanding of your programming stack.

For me this time, it was understanding just a little more about GHC. Next time I might dive into dependency management, or a different topic like the internal workings of Haskell data structures. These kinds of topics might not seem immediately applicable in your day job, but you'll be surprised at the times when deeper knowledge will pay dividends for you.

Getting Better at Haskell

But enough philosophizing. If you're completely new to Haskell, going "one level deeper" might simply mean the practical ability to use these tools at a basic level. If your knowledge is more intermediate, you might want to explore ways to improve your development process. These thoughts can lead to questions like:

1. What's the best way to set up my Haskell toolchain in 2023?

2. How do I get more efficient and effective as a Haskell programmer?

You can answer these questions by signing up for my new free course Setup.hs! This will teach how to install your Haskell toolchain with GHCup and get you started running and testing your code.

Best of all, it will teach you how to use the Haskell Language Server to get code hints in your editor, which can massively increase your rate of progress. You can read more about the course in this blog post.

If you subscribe to our monthly newsletter, you'll also get an extra bonus - a 20% discount on any of our paid courses. This offer is good for two more weeks (until May 1) so don't miss out!

How to Make ChatGPT Go Around in Circles (with GHC and Haskell)

As part of my research for the recently released (and free!) Setup.hs course, I wanted to explore the different kinds of compilation commands you can run with GHC outside the context of a build system.

I wanted to know…

Can I use GHC to compile a Haskell module without its dependent source files?

The answer, obviously, should be yes. When you use Stack or Cabal to get dependencies from Hackage, you aren't downloading and recompiling all the source files for those libraries.

And I eventually managed to do it. It doesn't seem hard once you know the commands already:

$ mkdir bin
$ ghc src/MyStrings.hs src/MyFunction.hs -hidir ./bin -odir ./bin
$ ghc -c app/Main.hs -i./bin -hidir ./bin -odir ./bin
$ ghc bin/Main.o ./bin/MyStrings.o ./bin/MyFunction.o -o ./bin/hello
$ ./bin/hello
...

But, being unfamiliar with the inner workings of GHC, I struggled for a while to find this exact combination of commands, especially with their arguments.

So, like I did last week, I turned to the latest tool in the developer's toolkit: ChatGPT. But once again, everyone's new favorite pair programmer had some struggles of its own on the topic! So let's start by defining exactly the problem we're trying to solve.

The Setup

Let's start with a quick look at our initial file tree.

├── app
│   └── Main.hs
└── src
    ├── MyFunction.hs
    └── MyStrings.hs

This is meant to look the way I would organize my code in a Stack project. We have two "library" modules in the src directory, and one executable module in the app directory that will depend on the library modules. These files are all very simple:

-- src/MyStrings.hs
module MyStrings where

greeting :: String
greeting = "Hello"

-- src/MyFunction.hs
module MyFunction where

modifyString :: String -> String
modifyString x = base <> " " <> base
  where
    base = tail x <> [head x]

-- app/Main.hs
module Main where

import MyStrings (greeting)
import MyFunction (modifyString)

main :: IO ()
main = putStrLn (modifyString greeting)

Our goal is to compile and run the executable with two constraints:

1. Use only GHC (no Stack or Cabal involved)

2. Compile the library separately, so that the Main module could be compiled using only the library's build artifacts, and not the source files.

Trying to Compile

Now, there are two easy ways to compile this code if we're willing to violate our constraints (particularly the second one). If all three files are in the same directory, GHC can immediately find the modules we're importing, so we can just call ghc Main.hs.

└── src
    ├── Main.hs
    ├── MyFunction.hs
    └── MyStrings.hs

...

$ cd src
$ ghc Main.hs
$ ./Main
elloH elloH

If we keep the original file structure, it's still pretty easy if we're willing to violate constraint 2 by using the library source files in the ghc command:

$ ghc app/Main.hs src/MyStrings.hs src/MyFunction.hs
$ ./Main
elloH elloH

However, with my extra constraint of not using the library source files directly, things got a little more difficult. I originally thought "Hey, let's just include the build artifacts in the GHC call." Each library module generates an interface file (.hi) and an object file (.o). So let's try including those in the GHC call:

$ ghc src/MyStrings.hs src/MyFunction.hs
$ ls -l src/
MyFunction.hi MyFunction.hs MyFunction.o MyStrings.hi MyStrings.hs MyStrings.o
$ ghc app/Main.hs src/MyFunction.hi src/MyFunction.o src/MyStrings.hi src/MyStrings.o
...
Could not find module 'MyStrings'
...
Could not find module 'MyFunction'

As we can see, it doesn't work! Even though we're including the build artifacts as direct arguments of ghc, this doesn't help our main module to find them.

So, feeling a little too lazy to dig deep through GHC documentation, and feeling like this problem might have too many arbitrary constraints to phrase in a Google query, I decided to try our aforementioned new favorite tool.

Enter: ChatGPT

The First Session

(Conversation has been abbreviated for relevance)

Me

How do I link existing .o files with a new source module using the 'ghc' command?

Chat GPT

To link existing object (.o) files with a new source module using the ghc command, you need to include the object files in the list of input files passed to the ghc command.

Assuming you have an existing .o file called example.o and a new source module called MyModule.hs, you can link them together into an executable program using the following command:

$ ghc -o MyProgram MyModule.hs example.o

This command tells ghc to compile and link the MyModule.hs source file with the example.o object file and create an executable program called MyProgram. The -o flag specifies the output file name.

Me

Is it possible to use a ghc command to link a Haskell object file (.o) with a Haskell source file that imports the other module?

Chat GPT

Yes, it is possible to use the ghc command to link a Haskell object file (.o) with a Haskell source file that imports the other module. Here's an example:

Suppose you have two files, MyModule.hs and OtherModule.o, and MyModule.hs imports a module that is defined in OtherModule.o. To link the two files into an executable program, you can use the following command:

$ ghc -o MyProgram MyModule.hs OtherModule.o

This command tells ghc to compile and link the MyModule.hs source file with the OtherModule.o object file, and create an executable program called MyProgram.

Me

Hmmm I don't think this works. I created the following file OtherModule.hs:

module OtherModule where

greeting :: String
greeting = "Hello"

And I'm trying to include it in another file:

module Main where

import OtherModule (greeting)

main :: IO ()
main = putStrLn greeting

I started by compiling OtherModule:

$ ghc src/OtherModule.hs

Then I moved OtherModule.o to the app directory (with MyModule.hs), and tried running this command:

$ cd app
$ ghc -o MyProgram MyModule.hs OtherModule.o

I get the error:

"Could not find module 'OtherModule'"

Chat GPT

I apologize for the confusion in my previous answer. You are correct that simply linking the object file with the executable does not work in this case.

When GHC compiles a module, it generates an interface file (.hi) that contains information about the module's name, type signatures, and other details. The compiled object file (.o) contains the compiled machine code for the functions defined in the module.

When you try to import a module in a source file, GHC looks for the interface file (OtherModule.hi) to find information about the module's name and type signatures. If the interface file is not found, GHC will raise an error.

In your case, when you compile OtherModule.hs, GHC generates an interface file called OtherModule.hi. To compile MyModule.hs, GHC needs to be able to find OtherModule.hi in order to resolve the import of OtherModule. Simply linking the OtherModule.o object file is not enough.

To make this work, you need to compile OtherModule.hs and MyModule.hs together, like this:

ghc -o MyProgram MyModule.hs src/OtherModule.hs

This command tells GHC to compile MyModule.hs and OtherModule.hs together, and create an executable program called MyProgram.

I hope this clears up the confusion. Again, I apologize for the confusion in my previous answer.

Me

Ah, so I need to include both source files, and it can't be done by linking an object file with another source file?

Chat GPT

Yes, that's correct. When you import a module in a source file, GHC needs to be able to find the interface file (.hi) for that module in order to resolve the import. If the interface file is not found, GHC will raise an error.

When you compile a source file with GHC, it generates both an object file (.o) and an interface file (.hi). The object file contains the compiled machine code for the functions defined in the module, while the interface file contains information about the module's name, type signatures, and other details.

When you compile a source file that imports another module, GHC needs to be able to find the interface file for that module in order to compile the importing module. If the interface file is not found, GHC will raise an error.

So in your case, you need to compile both MyModule.hs and OtherModule.hs together so that GHC can generate the necessary interface files and link the resulting object files together to produce the executable program.

It is not possible to directly link an object file with a source file that imports it, as GHC needs the interface file to be present in order to compile the importing module.

Analysis

Suspiciously, ChatGPT suggests a less detailed command than the previous one that failed for me!

$ ghc -o MyProgram MyModule.hs OtherModule.o

It includes only the object file, instead of the object and interface file. And, of course this fails, like I showed above.

Now there's an interesting statement that will come into play later, though I didn't realize it!

Yes, that's correct. When you import a module in a source file, GHC needs to be able to find the interface file (.hi) for that module in order to resolve the import. If the interface file is not found, GHC will raise an error.

I assumed that to "find" the interface file meant providing it on the command line (as in my original command). So with that idea having failed, I fixated on one part of the final statement:

It is not possible to directly link an object file with a source file that imports it...

And concluded that getting the right information from ChatGPT might be a lost cause. It seemed like it was sure this was possible, and then just as sure that it was impossible. But details matter, as we'll see!

Working Around It

So feeling slightly defeated, I decided to poke around the areas of the code where Cabal actually saves packages to see if there was a pattern I could follow. And I managed to cobble together a convoluted workaround.

What I observed was that GHC could use any module from a library in the Cabal package database. On my system's setup, this could be found in the .ghcup cache:

$ cd ~/.ghcup/ghc/9.2.5/lib/ghc-9.2.5/
$ ls
array-0.5.4.0/
base-4.16.4.0/
...
package.conf.d/
$ ls package.conf.d
array-0.5.4.0.conf
base-4.16.4.0.conf
...

And indeed, the package directories did not have source files, only various forms of compiled object and interface files. And each package had a corresponding .conf file in the package.conf.d directory.

So I could...

1. Make my own directory in the package database (MyPackage-0.1.0.0)
2. Place the .hi and .o files in there
3. Create a phony MyPackage-0.1.0.0.conf and put it in package.conf.d. (Base it on an existing .conf file but make some substitutions)

And this...technically...works. But it makes me feel a little bit like:

But manually manipulating a package database wasn't the solution I had in mind though. Surely there's something easier, right?

The Second Session

After a week of sitting on this, I decided to come back to ChatGPT and see if it had any other ideas. I decided to frame my question in a different way, stating a lot more details up front. Lo and behold, it had a different answer this time!

Me

Suppose I have a Haskell module called MyStrings:

module MyStrings where

greeting :: String
greeting = "Hello"

I can compile this module:

$ ghc MyStrings.hs

This generates an interface file, MyStrings.hi, and an object file, MyStrings.o.

Now suppose I have an executable file, Main.hs, that imports MyStrings:

module Main where

import MyStrings (greeting)

main :: IO ()
main = putStrLn greeting

Can I use ghc to compile this Main file by using the .hi and .o files, without including the imported source file, MyStrings.hs in the command?

Chat

Yes, you can use the .hi and .o files generated by compiling MyStrings.hs to compile Main.hs without including MyStrings.hs in the command. To do so, you can use the -c option to tell GHC to compile Main.hs to an object file, and then use the -o option to link the object files together into an executable:

$ ghc -c Main.hs
$ ghc Main.o MyStrings.o -o myprogram

The first command compiles Main.hs to an object file Main.o, without attempting to link it. The second command links Main.o with MyStrings.o and creates an executable called myprogram. The .hi files are not needed for linking, so you don't need to include them in the command.

Note that in order for this to work, the interface file MyStrings.hi must be present in the same directory as Main.hs when you compile it. If you move MyStrings.hi to a different directory, you'll need to use the -i option to tell GHC where to find it:

$ ghc -i/path/to/MyStrings -c Main.hs
$ ghc Main.o /path/to/MyStrings/MyStrings.o -o myprogram

Here, -i/path/to/MyStrings tells GHC to look for MyStrings.hi in the directory /path/to/MyStrings.

Analysis

So there are two key ingredients that weren't mentioned at all in our first pass: the -c option and the -i option.

Using -c performs compilation without linking. With this option, GHC is able to produce an object file for our Main module without needing its dependent objects. So .o files are unnecessary in this stage.

We still need the .hi files though. But instead of providing them on the command line, we use the -i argument to supply them. It's an odd argument, because we put the path right after the i, without any spacing.

After we're done with the first phase, then we can link all our object files together.

Solving It

And sure enough, this approach works!

$ ghc src/MyStrings.hs src/MyFunction.hs
$ ghc -c app/Main.hs -i./src
$ ghc app/Main.o ./src/MyStrings.o ./src/MyFunction.o -o hello
$ ./hello
elloH elloH

And if we want to be a little cleaner about putting our artifacts in a single location, we can use the -hidir and -odir arguments for storing everything in a bin directory.

$ mkdir bin
$ ghc src/MyStrings.hs src/MyFunction.hs -hidir ./bin -odir ./bin
$ ghc -c app/Main.hs -i./bin -hidir ./bin -odir ./bin
$ ghc bin/Main.o ./bin/MyStrings.o ./bin/MyFunction.o -o ./bin/hello
$ ./bin/hello
elloH elloH

And we're done! Our program is compiling as we wanted it to, without our "Main" compilation command directly using the library source files.

Conclusion

So with that fun little adventure concluded, what can we learn from this? Well first of all, prompts matter a great deal when you're using a Chatbot. The more detailed your prompt, and the more you spell out your assumptions, the more likely you'll get the answer you're looking for. My second prompt was waaay more detailed than my first prompt, and the solution was much better as a result.

But a more pertinent lesson for Haskellers might be that using GHC by itself can be a big pain. So if you're a beginner, you might be asking:

What's the normal way to build Haskell Code?

You can learn all about building and running your Haskell code in our new free course, Setup.hs. This course will teach you the easy steps to set up your Haskell toolchain, and show you how to build and run your code using Stack, Haskell's most popular build system. You'll even learn how to get Haskell integrations in several popular code editors so you can learn from your mistakes much more quickly. Learn more about it on the course page.

And if you subscribe to our monthly newsletter, you'll get a code for 20% off any of our paid courses until May 1st! So don't miss out on that offer!

How to Stump ChatGPT in 1 Easy Step (Hint: Ask a Windows Question)

While doing some research for Setup.hs, my new free course, I confronted a fact of Haskell development with Stack.

The commands for running individual test suites are cumbersome.

For example, if you have a parse-tests suite in your Project.cabal file, you can run it (without running other suites) like so:

$ stack build Project:test:parse-tests

The structure isn't hard once you're used to it, but remembering the exact names of your test suites isn't always easy on larger projects. This leads to the question:

How can I create aliases for 'stack build' commands?

On Unix-based systems, it's quite easy. Just slap something like this in your .bashrc (or equivalent) and you're good to go:

alias ppt="stack build Project:test:parse-tests"

...

$ source ~/.bashrc
$ ppt
(Tests run)

Windows though is, as always, a special beast. And much as I generally prefer to use Windows Subsystem for Linux, I wanted to help any users who might be working with a pure Windows setup. This article will focus on:

How do I enable 'stack' aliases in Windows PowerShell?

Now, when I lay out the code I put in my PowerShell profile, it looks a little dense but fairly innocuous:

Function ProjectParseTest { Invoke-Expression "stack build Project:test:parse-tests" }
New-Alias -Name qpt -Value ProjectParseTest

But discovering the exact formula for getting all this to work was quite difficult. And as we'll see, in fact, even ChatGPT, the new kid on the block, struggles with this question. After a brief while struggling through Windows documentation, I turned to the chat bot. Let's review the first question:

ChatGPT Question 1

Me

How can I make command aliases in Windows 11 Powershell?

Chat GPT

You can create command aliases in Windows 11 PowerShell by using the New-Alias cmdlet. Here's an example that creates an alias ls for the Get-ChildItem cmdlet:

New-Alias -Name ls -Value Get-ChildItem

This will create an alias named ls that can be used just like the Get-ChildItem cmdlet. For example:

ls .

This will list the contents of the current directory, just as if you had typed Get-ChildItem ..

To persist your alias across sessions, you need to add it to your PowerShell profile. To do that, first determine the path to your profile by running the following command:

$ $profile

If the profile does not exist yet, you can create it by running the following command:

New-Item -Type File -Path $profile -Force

Then, open the profile in your favorite text editor and add the New-Alias command. Save the file and restart your PowerShell session. Your alias will be available every time you start PowerShell.

Analysis

Now, everything Chat has told me is technically true. The command it gives successfully creates an alias. Notably though, if you want to try this yourself, you should use a different alias, as Powershell already includes the alias ls in most recent Windows version (previously you would need to use the alias dir to get that effect).

Additionally, I found that the PROFILE variable was indeed set, but the file did not exist. In fact, the directory it was supposed to be in (Documents/WindowsPowerShell) did not exist either! So you'll have to make that before creating the file. But placing an alias command in there will persist the alias across sessions.

However, notice that this doesn't quite answer our question, because the subject of the alias was an established Windows expression (Get-ChildItem, also know as a cmdlet), rather than a terminal command we would write.

Let's suppose we just want to alias sb=stack build. It does not work to simply use a string for the command:

There are some clues to our eventual solution in the error message though. It tells us we need a cmdlet, a function, a script file, or an operable program.

Now stack by itself is an operable program. So we could actually do the following:

PS > New-Alias -Name st -Value "stack"
PS > st
(stack usage output)

But we want to provide arguments as part of the alias. So let's see if ChatGPT knows how. We'll see that its answers are not quite right.

ChatGPT Question 2

Me

How do I make the alias run a command, like "stack build"

Chat GPT

You can create an alias that runs a command by using the Invoke-Expression cmdlet to evaluate a string as a PowerShell expression. Here's an example that creates an alias sb for the stack build command:

New-Alias -Name sb -Value { Invoke-Expression "stack build" }

This will create an alias named sb that can be used to run the stack build command. For example:

sb

This will run the stack build command, just as if you had typed it into the PowerShell prompt.

As with the previous example, you can add this alias to your PowerShell profile to persist it across sessions.

Me

I get an error "Cannot evaluate parameter 'Value' because its argument is specified as a script block and there is no input"

How do I specify input in this alias?

Chat GPT

To specify input in a script block alias, you can use the Param keyword to declare input parameters and then pass the input to the underlying command. Here's an example that creates an alias sb for the stack build command, allowing you to specify the target to build as input:

New-Alias -Name sb -Value {
    Param ($target)
    Invoke-Expression "stack build $target"
}

This will create an alias named sb that can be used to run the stack build command with a specified target. For example:

sb my-target

This will run the stack build my-target command, just as if you had typed it into the PowerShell prompt.

As with the previous examples, you can add this alias to your PowerShell profile to persist it across sessions.

Analysis

Both of the proposed approaches fail! I commented on the issue explicitly after the first response, and then got the same error running the second approach ("Cannot evaluate parameter 'Value' because its argument is specified as a script block..."). So I gave up on my interrogation and started searching for documentation again.

Solving the Problem

I don't remember exactly where I found the solution. I think it involved cobbling together points from multiple Stack Overflow posts. Windows documentation was not very helpful. Let's recall the error message from earlier:

The term 'stack build' is not recognized as the name of a cmdlet, function, script file, or operable program.

We can't make an official Windows PS cmdlet out of our program, nor can we make an operable program with the arguments we want. We could make a script file, but running scripts from PowerShell is surprisingly difficult (there are some extra steps with permissions). The answer is that we can make a function that our alias will refer to.

This function will, incidentally, use the Invoke-Expression idea ChatGPT recommended as well, just not directly as the alias value!

The following code will go in our $PROFILE. First, we make a function that invokes our expression. We can give this function an arbitrary name, but I used a capitalized name (ProjectParseTest) to distinguish from any potential aliases.

Function ProjectParseTest { Invoke-Expression "stack build Project:test:parse-tests" }

Now we can use this function as the object of a New-Alias call! So we use the command ChatGPT suggested, just substituting our function for the -Value instead of providing a raw Invoke-Expression command.

Function ProjectParseTest { Invoke-Expression "stack build Project:test:parse-tests" }
New-Alias -Name ppt -Value ProjectParseTest

This alias succeeds now, and by putting this in my PowerShell profile, I can persist the alias across sessions!

$ ppt
(Test runs)

Haskell on Windows

Now aliases are just one small piece of the Haskell puzzle. So if you're trying to get started with Haskell, but don't have a Mac, and aren't familiar with Linux, you might want to know:

How do I set up my Haskell toolchain on Windows?

My new free course Setup.hs goes over all the basics of setting up your Haskell toolchain, including how to get started with code hints in three of the most popular editors out there. Plus, every lecture includes a walkthrough video for Windows* so you can learn what kinds of odd quirks you might come across! You can read more about the course in this article.

Plus, if you subscribe to our monthly newsletter, you'll also get a 20% discount code for all our paid courses that is good until May 1! So don't miss out on your chance to learn about Haskell!

*Separate walkthrough videos for MacOS and Linux are also included.

New Free Course: Setup.hs!

You can read all the Haskell articles you want, but unless you write the code for yourself, you'll never get anywhere! But there are so many different tools and ideas floating around out there, so how are you supposed to know what to do? How do you even get started writing a Haskell project? And how can you make your development process as efficient as possible?

The first course I ever made, Your First Haskell Project, was designed to help beginners answer these questions. But over the years, it's become a bit dated, and I thought it would be good to sunset that course and replace it with a new alternative, Setup.hs. Like its predecessor, Setup.hs is totally free!

Setup.hs is a short course designed for many levels of Haskellers! Newcomers will learn all the basics of building and running your code. More experienced Haskellers will get some new tools for managing all your Haskell-related programs, as well as some tips for integrating Haskell features into your code editor!

Here's what you'll learn in the course:

1. How to install and manage all of the core Haskell tools (GHC, Cabal, Stack)
2. What components you need in your Haskell project and how you can build and run them all
3. How to get your editor to use advanced features, like flagging compilation errors and providing autocomplete suggestions.

We'll do all of this in a hands-on way with detailed, step-by-step exercises!

Improvements

Setup.hs makes a few notable updates and improvements compared to Your First Haskell Project.

First, it uses GHCup to install all the necessary tools instead of the now-deprecated Haskell Platform. GHCup allows for seamless switching between the different versions of all our tools, which can be very useful when you have many projects on your system!

Second, it goes into more details about pretty much every topic, whether that's project organization, Stack snapshots, and extra dependencies.

Third and probably most importantly, Setup.hs will teach you how to get Haskell code hints in three of the most common code editors (VS Code, Vim & Emacs) using Haskell Language Server. Even if these lectures don't cover the particular editor you use, they'll give you a great idea of what you need to search for to learn how. I can't overstate how useful these kinds of integrations are. They'll massively speed up your development and, if you're a beginner, they'll rapidly accelerate your learning.

If all this sounds super interesting to you, head over to the course page and sign up!

Series Spotlight: Monads and Functional Structures!

Every so often I like to spotlight some of the permanent series you can find on the skills page, which contains over a dozen tutorial series for you to follow! This week I’m highlighting my series on Monads and Functional Structures!

Monads are widely understood to be one of the trickier concepts for newcomers to Haskell, since they are very important, very abstract and conceptual, and do not really appear in most mainstream languages. There are a lot of monad tutorials out there on the internet, most of which are either too shallow, or too deep.

This series will help you understand the concept from the group up, starting with simpler abstract structures like functors and applicative functors.

Here’s an outline of the series:

1. Functors

2. Applicative Functors

3. Monad Basics

4. Reader and Writer Monads

5. State Monad

6. Monad Transformers

7. Monad Laws

For a more in depth look at monads and the effects they help us implement in our code, you can check out our two courses, Making Sense of Monads and Effectful Haskell!

GHC 9.6.1 Includes Javascript Backend

Some exciting news this week, as the release of GHC 9.6.1 includes the merger of a couple of different web-based backends - one for Web Assembly and one for Javascript. These features move Haskell in the direction of being a first-class web development language!

From the release notes:

The WebAssembly backend has been merged. This allows GHC to be built as a cross-compiler that targets wasm32-wasi and compiles Haskell code to self-contained WebAssembly modules that can be executed on a variety of different runtimes.

The JavaScript backend has been merged. GHC is now able to be built as a cross-compiler targeting the JavaScript platform.

This is a particularly exciting direction for me, since I’ve been exploring ways to use Haskell in web development for many years, but found a lot of the current approaches require a lot of onboarding work to really get going in a meaningful way. In my Practical Haskell course, I show how to do a basic integration of a Haskell Web Server and an Elm frontend. But I look forward to the day when I can re-do that section of the course entirely with Haskell!

Of course, merging the backends is just a first step - there’s a long way to go. A few caveats mentioned in the release notes as well:

There are a few caveats to be aware of [with the WebAssembly backend]:

To use the WebAssembly backend, one would need to follow the instructions on ghc-wasm-meta. The WebAssembly backend is not included in the GHC release bindists for the time being, nor is it supported by ghcup or stack yet.

The WebAssembly backend is still under active development. It’s presented in this GHC version as a technology preview, bugs and missing features are expected.

The [Javascript] backend should be considered a technology preview. As such it is not ready for use in production, is not distributed in the GHC release bindists and requires the user to manually build GHC as a cross-compiler. See the JavaScript backend wiki page on the GHC wiki for the current status, project roadmap, build instructions and demos.

Both of these backends are technology previews, meaning they’re only ready for the most adventurous Haskellers to start experimenting - a lot of setup work is still required. But it’s certainly an important step in the right direction! Since these are included with GHC 9.6.1, improvements are possible in further minor releases to GHC 9.6, rather than needing to wait for the next major release of GHC 9.8.

Adding a Database to our AWS Server

In the last few articles on the blog, we've been exploring how to launch a Haskell web server using AWS. Here are the steps we've done so far:

1. Create a local Docker Image
2. Upload the Docker Image to ECR
3. Deploy your Server using Elastic Beanstalk

In this final part of the series, we're going to learn to attach a database to our application.

There are a few gotchas to this. Setting up the database for first time use is a bit tricky, because we have to do some initial migrations. Then we need to use environment variables to ensure it works both locally and on the remote server. Let's get started.

A Basic Schema

Let's first assume we have a super basic schema using the Persistent library. (If you want some details on how this works, see our Real World Haskell series). We'll just have one type in our database, and users will use server endpoints to create or fetch these "text entries".

import           Database.Persist.Sql
import qualified Database.Persist.TH as PTH
import           Data.Text (Text)

PTH.share [PTH.mkPersist PTH.sqlSettings, PTH.mkMigrate "migrateAll"] [PTH.persistLowerCase|

  TextEntry sql=text_entries
    text Text

|]

An important product of this template Haskell sequence is the migrateAll function, which will run the proper commands to migrate a Postgres database to fit our schema by creating tables.

Whenever we first create a database, we have to make sure it's migrated. But before we even do that we have to make sure we've created a database for Postgres to use! Let's see the commands we need for this, and how to run them in Haskell.

Running Setup Commands

When you install Postgres on your machine, you'll have separate "databases" on your system to help you keep your data separate. Separating data allows each database to have its own "users" table without having any name conflicts, for one example. By default, Postgresql comes installed with a database called postgres.

But we don't want to use this to store our data. We want to create a separate database. We want to create this database if it's the first time we're running the server with a database. But otherwise, we just want to make sure its migrations are up to date.

Now, the command we would run to create this database is simple:

CREATE DATABASE quiz;

But we can first run this command to see if this database already exists:

SELECT datname FROM pg_database WHERE datname = 'quiz';

Both these commands assume we are connected to the postgres database.

Since these first two instructions are raw commands, we can run them using the Postgresql Simple library. Here's some code to do this.

createDBIfMissing :: String -> IO ()
createDBIfMissing connString = do
  connection <- connectPostgreSQL (pack connString)
  putStrLn "Checking/Creating 'quiz' Database"
  let checkQuery = "SELECT datname FROM pg_database WHERE datname = 'quiz';"
  (checkResult :: [Only String]) <- query_ connection checkQuery
  when (null checkResult) $ do
    putStrLn "Not found! Creating 'quiz' database!"
    let createQuery = "CREATE DATABASE quiz;"
    void $ execute_ connection createQuery

When we run checkQuery, it sees if the quiz database exists. If its result is null, then we'll run the additional command to create our database.

Once we have this function, we can write a wrapper that will create the database and then migrate it for our schema. Here's what this wrapper looks like:

migrateDb :: String -> String -> IO ()
migrateDb baseConnString quizConnString = do
  createDBIfMissing baseConnString
  putStrLn "Migrating Database"
  runPG quizConnString (runMigration migrateAll)

runPG :: String -> SqlPersistT (LoggingT IO) a -> IO a
runPG connString action = runStdoutLoggingT $
  withPostgresqlConn (pack connString) $ \backend ->
    runReaderT action backend

Notice migrateDb takes two different connection strings. One is for the base (postgres) database. The other is for our new quiz database. The creation queries run on the first, the migration runs on the second.

But how do we use these functions within our server?

Loading the URI

When we kick off our server, we have to load the database URI for our Postgres database. We'll use the format of {host}:{port}. If you're running it locally, this would just be localhost:5432. But when we deploy the server, we'll use a different URI. So let's write a function to load the host and port (separated by a colon) from an environment variable named DATABASE_URI.

loadDatabaseEnv :: IO (String, String)
loadDatabaseEnv = do
  dbEnv <- lookupEnv "DATABASE_URI"
  if isNothing dbEnv || ':' `notElem` fromJust dbEnv
    then return ("localhost", "5432")
    else return (span (/= ':') (fromJust dbEnv))

Now we need to construct the full Postgres connection string. This has the following general format:

host={host} port={port} dbname={dbname} user={user} password={password}

As a default value, you can often just have the username and password both be postgres (though of course this isn't recommended for a serious database). But let's make a function to substitute in the other values:

mkPostgresUri :: String -> String -> String -> String
mkPostgresUri host port dbname =
  "host='" <> host <> "' port=" <> tail port <> " dbname='" <> dbname <> "' user='postgres' password='postgres'"

Finally, we'll pull our different pieces together, get both URIs, and launch our server. In my example, I'm using a Servant server (more details on that in this article), and this will often require passing the database string as an argument.

server :: String -> Server QuizAPI
server connString = ...

runServer :: IO ()
runServer = do
  (host, port) <- loadDatabaseEnv
  let baseConnString = mkPostgresUri host port "postgres"
  let quizConnString = mkPostgresUri host port "quiz"
  migrateDb baseConnString quizConnString
  putStrLn "Running Server!"
  run 8080 (serve api (server quizConnString))

Having made all these modifications to our server, of course we have to rebuild and redeploy our docker image for that! We can create the new local image with:

docker build -t quiz-server .

Then for more detailed instructions on deploying it, refer to part 2 and part 3 of this series!

When you deploy the server, you'll find it's crashing of course, because we haven't configured the database! So let's get to the real meat of this article…setting up the database on AWS!

Create a Database with RDS

This process is not actually too challenging. The first thing we're going to do is use RDS (Relational Database Service) to set up our database. This is easily done from the AWS console.

1. Select the RDS service
2. Hit the orange "Create Database" button
3. Go through the creation wizard, making sure to select "Postgres" and the "Free Tier" option (assuming you're just making a test app).

Most of the default options are fine, but as I mentioned above I specified postgres for the username and password of the database. I also unchecked the box for "Performance Insights" since this could lead to additional billing charges if you forget to turn it off.

Once you've created your database, you can then click the "databases" link on the sidebar, and select your new database. On that screen, you'll be able to see the "endpoint" and "port" of your database. These are the values you'll need for your environment!

Add Environment Variable

To connect your environment to the database, now you just have to add an environment variable! To do this, you have to access the configuration from the web portal:

1. Go to the Elastic Beanstalk service
2. Select "Environments" from the sidebar and then click the environment you have running your server.
3. Click on the "Configuration" link on the side, and then select the "Edit" button in the "Software" section.
4. At the very bottom, you'll find the "Environment Properties" section. Fill in DATABASE_URI as the key, and the {host}:{port} combination you got from your database in RDS.
5. Click "Apply" to make the change!

By adding an environment variable, you are changing the configuration of your server, so it will reboot. Once it relaunches, you should find that it works, and you can persist information from your database!

Conclusion

Hopefully this series has helped you learn how to deploy your Haskell code to AWS! If you'd like to see all this article in video form, you can check out our YouTube video covering these steps!

For more tips on creating a "Real World" application, you can read our series on web skills! You can also download our Haskell Production checklist for some ideas of other libraries and tools you can use to improve your Haskell!

Deploying a Haskell Server to AWS

In the last few articles, we've been talking about how to deploy a Haskell application using AWS. This is part 3 of the series. So if you haven't done parts 1 & 2, you should start there so you can follow along!

In Part 1, we wrote a Dockerfile and created a local Docker image containing a simple program for a Haskell web server.

In the Part 2, we pushed our container image to the AWS container registry (ECR). Notably, this involved creating an AWS account, downloading AWS command line tools and authenticating on the command line. We'll run a couple more of these commands today, so hopefully you're still authenticated!

But now that our container is uploaded, deploying that container is fairly straightforward. But it requires us to use a couple new concepts, as we'll see.

Adding ECR Permission

Before we get started, there's one step we have to take on the web portal. You need to give Elastic Beanstalk permission to download your ECR containers. You can do this using the IAM service from the AWS portal. Then follow these steps:

1. Select "roles" on the left hand menu.
2. Select "aws-elasticbeanstalk-ec2-role" in the list in the middle of the screen.
3. Click "Add Permissions"
4. Search for and select "AmazonEC2ContainerRegistryReadOnly"

Now let's get into the steps on our local machine.

Configuration File

There are multiple approaches to deploying a docker container, but the one that worked most easily for me was to create a file called Dockerrun.aws.json. (Full other methods, refer to the documentation). This approach involves a counter-intuitive idea. We're going to create a separate directory outside of our main project directory. We'll call it remote.

~/Quiz $ cd ..
~/ $ mkdir remote && cd remote

In this directory, we'll make a single file, called Dockerrun.aws.json. This will, of course, be a JSON file. It will be a very simple configuration file telling our application to use the docker image we pushed last time to ECR. We have to start it by specifying the version of the program (which is 1 because we're only using a single container).

{
  "AWSEBDockerrunVersion": "1",
  ...
}

Now we'll use tell it to use the Docker image we pushed last time by giving the URI under the Image object:

{
  "AWSEBDockerrunVersion": "1",
  "Image": {
    "Name": "165102442442.dkr.ecr.us-west-2.amazonaws.com/quiz-server"
  },
  ...
}

Finally, we'll specify the port, similar to a Dockerfile. We'll use 8080 both for the "Container" port and the "Host" port.

{
  "AWSEBDockerrunVersion": "1",
  "Image": {
    "Name": "165102442442.dkr.ecr.us-west-2.amazonaws.com/quiz-server"
  },
  "Ports": [{
    "ContainerPort": 8080,
    "HostPort": 8080
  }]
}

This is the only file we need in this directory! So now let's see what commands we need to run.

Creating the Application

Now we have two more steps that can largely be accomplished on the command line. First, we have to create an application. Then we have to create an environment to use for that application.

Before we can create an application though, we have to create a Git repository, just to store our single file! That's how AWS figures out what to push for configuration.

~/remote $ git init
~/remote $ git add .
~/remote $ git commit -m "First Commit"

Now we can create the application using the eb init command. We'll give our application the name quiz-server.

~/remote $ eb init -p docker quiz-server

You can then see your application on the web portal by accessing the "Elastic Beanstalk" service and clicking the "Applications" tab on the left menu.

Creating the Environment

Now we have to deploy an environment to deploy for our application. When first creating this environment, we use the eb create command. We'll give this environment the name quiz-server-env.

~/remote $ eb create quiz-server-env

This will take a while to deploy. But once it's done, you should be able to see it by clicking the "Environments" tab from the previous screen in the web portal. This will also show you the URL you can use to access your server. It's now successfully deployed!

Debugging

Sometimes, your deployment might fail. For example, you might misspell the name of your container. If you click on your environment (from the "Environments" tab), then you'll be able to access the "Logs" on the left hand menu. This can help you debug. If you need to change your configuration file, you'll need to commit it, though you don't need to push it to any remote repository. You instead use eb deploy to push your changes.

~/remote $ git add Dockerrun.aws.json
~/remote $ git commit -m "New Commit"
~/remote $ eb deploy

Now the deployment process should start again!

Video

You can also watch our YouTube video to see all these steps in action!

Conclusion

You now have enough information to deploy a Haskell web application to Heroku! We'll have one more installment in this series around adding a database to our application, so stay tuned for that! In the meantime, subscribe to our monthly newsletter so you can stay up to date with all the latest news!

Pushing our Container to AWS ECR

In the first part of this blog series we saw how to create a local docker image containing a simple web server program. In order to run this server remotely, we have to upload this image somewhere to deploy it.

One service that lets us deploy docker images is Amazon Web Services (AWS). In this article, we're going to take the first step, and walk through the process of publishing our container image to the AWS Elastic Container Registry (ECR). Next time around, we'll see how to actually deploy our application using this image.

In principle, publishing the image is a simple task. But in my experience with AWS, the processes and documentation just aren't quite as clear as one would like them to be. There tend to be a lot of branches in their tutorials, and it's often not clear which path is the right path. The sheer amount of AWS-specific terminology can get extremely confusing, and this can make it hard to know if you've satisfied the prerequisites for the tutorial.

So in this article I'm going to be as explicit as possible, and include a video at the end so you can follow along. Here's the high level overview:

1. Create an AWS account
2. Create an ECR Repository
3. Install the AWS Command Line Interface
4. Login using the CLI
5. Push the container using Docker

Create an AWS Account

First of course, you need to create an account with Amazon Web Services. This is a separate account from a normal Amazon account. But a massive gotcha is that you should not use the exact email address from your Amazon account. This can cause a weird loop preventing you from logging in successfully (see this Stack Overflow issue).

If you have Gmail though, it should work to use the '+' trick with email aliases. So you can have `name@gmail.comfor your Amazon account andname+aws@gmail.com` for your AWS account.

Create an ECR Repository

Next you'll need to login to your account on the web portal and create an ECR repository. To do this, you'll simply click the services tab and search for "Elastic Container Registry". Assuming you have no existing repositories, you'll be prompted with a description page of the service, and you'll want to find the "Get Started" button under the "Create a Repository" header off in the top right corner.

The only thing you need to do on the next page is to assign a name to the repository. The prefix of the repository will always have the format of {account-id}.dkr.ecr.{region}.amazonaws.com, where the account ID is a 12-digit number.

If you want, you can also set the repository as public, but my instructions will assume that you'd made a private repository. To finish up, you'll just click the "Create Repository" button at the bottom of the page. This part is also covered in the video at the bottom if you want to see it in action!

Install the AWS CLI

Our next few actions will happen on our local command line prompt. To interact with our AWS account, we'll need to install the AWS Command Line Interface. To install these tools, you can follow this user guide. It is fairly straightforward to follow once you select your operating system. You know it's succeeded when the command aws --version succeeds on your command line.

Login Using the CLI

Now assuming you created a private repository, you'll need to authenticate on the command line. The first step in this process is to create an access key. You can do this from the web portal by clicking your account name in the top right corner to open up a menu and then going to the "Security Credentials" page. There's a section for "Access Keys" about midpage, and you'll want to use "Create Access Key".

If you do this as a "root" user, AWS will warn you that this is not the advised practice and you should instead create such keys as an "IAM User". But it is possible to do use root for demonstration purposes.

You'll want to copy the "Access Key ID" and the key itself. The latter must be copied or downloaded before you leave the page (you can't come back to it later).

You can then login using the aws configure command in your command line terminal. This will ask you to enter your access key ID and then the key itself, as well as the region.

Now that you're authenticated with AWS, we have to allow AWS to login to Docker for us. The following command would give us the Docker password for AWS in the us-west-2 region:

>> aws ecr get-login-password --region us-west-2

We can pipe this password into the docker login command and connect to the repository we created with this command, where you should substitute your region and your account ID.

>> aws ecr get-login-password --region {region} | \
  docker login --username AWS --password-stdin {account-id}.dkr.ecr.{region}.amazonaws.com

Note how you actually do not need the repository name for this command! Just the prefix formed by your account and the region ID.

Pushing the Image

Now that we're authenticated, we just need to push the container image. We'll start by reminding ourselves of what our image ID is:

>> docker images
REPOSITORY TAG    IMAGE ID ...
quiz-server latest b9eab6a22b12 ...

The first thing we need to do is provide a "tag" for this image corresponding to the remote ECR repository we created. This requires the image ID and the full repository URI. We'll also attach :latest to indicate that this is the most recent push. Here's the specific command I used for my IDs:

>> docker tag b9eab6a22b12 165102442442.dkr.ecr.us-west-2.amazonaws.com/quiz-server:latest

Here's a more generic command template:

>> docker tag {image-id} {account-id}.dkr.ecr.{region}.amazonaws.com/{repo-name}:latest

Finally, we just need to push it using this new repository/tag combo! Here's what it looks like for me:

>> docker push 165102442442.dkr.ecr.us-west-2.amazonaws.com/quiz-server:latest

And more generically:

>> docker push {account-id}.dkr.ecr.{region}.amazonaws.com/{repo-name}:latest

You should then be able to see your image if you head to your ECR dashboard!

Video Walkthrough

If you want to see all this in action, you can head to YouTube and take a look at the video walkthrough! If you are enjoying this series, make sure to subscribe to our monthly newsletter!

AoC 2022: The End!

I've completed the final two video walkthroughs of problems from Advent of Code 2022! You can view them on my YouTube Channel, or find links to all solutions on our summary page.

Day 24 is the final graph problem, one that involves a couple clever optimizations. See the video here, or you can read a written summary as well.

And last of all, Day 25 introduces us to the concept of balanced quinary. Quinary is a number system like binary or ternary, but in the balanced form, characters range for (-2) to 2 instead of 0 to 4. This final video is on YouTube here, or you can read the write-up.

I promise there will be more problem-solving related content later this year! If these videos and walkthroughs have helped you improve your skills, make sure to subscribe to our monthly newsletter so you stay up to date!

Creating a Local Docker Image

Running a web server locally is easy. Deploying it so other people can use your web application can be challenging. This is especially true with Haskell, since a lot of deployment platforms don't support Haskell natively (unlike say, Python or Javascript). In the past, I've used Heroku for deploying Haskell applications. In fact, in my Practical Haskell and Effectful Haskell courses I walk through how to launch a basic Haskell application on Heroku.

Unfortunately, Heroku recently took away its free tier, so I've been looking for other platforms that could potentially fill this gap for small projects. The starting point for a lot of alternatives though, is to use Docker containers. Generally speaking, Docker makes it easy to package your code into a container image that you can deploy in many different places.

So today, we're going to explore the basics of packing a simple Haskell application into such a container. As a note, this is different from building our project with stack using Docker. That's a subject for a different time. My next few articles will focus on eventually publishing and deploying our work.

Starting the Dockerfile

So for this article, we're going to assume we've already got a basic web server application that builds and runs locally on port 8080. The key step in enabling us to package this application for deployment with Docker is a Dockerfile.

The Dockerfile specifies how to set up the environment in which our code will operate. It can include instructions for downloading any dependencies (e.g. Stack or GHC), building our code from source, and running the necessary executable. Dockerfiles have a procedural format, where most of the functions have analogues to commands we would run on a terminal.

Doing all the setup work from scratch would be a little exhausting and error-prone. So the first step is often that we want to "inherit" from a container that someone else has published using the FROM command. In our case, we want to base our container off of one of the containers in the Official Haskell repository. We'll use one for GHC 9.2.5. So here is the first line we'll put in our Dockerfile:

FROM haskell:9.2.5

Building the Code

Now we have to actually copy our code into the container and build it. We use the COPY command to copy everything from our project root (.) into the absolute path /app of the Docker container. Then we set this /app directory as our working directory with the WORKDIR command.

FROM haskell:9.2.5

COPY . /app
WORKDIR /app

Now we'll build our code. To run setup commands, we simply use the RUN descriptor followed by the command we want. We'll use 2-3 commands to build our Haskell code. First we use stack setup to download GHC onto the container, and then we build the dependencies for our code. Finally, we use the normal stack build command to build the source code for the application.

FROM haskell:9.2.5

...

RUN stack setup && stack build --only-dependencies
RUN stack build

Running the Application

We're almost done with the Dockerfile! We just need a couple more commands. First, since we are running a web server, we want to expose the port the server runs on. We do this with the EXPOSE command.

FROM haskell:9.2.5
...

EXPOSE 8080

Finally, we want to specify the command to run the server itself. Supposing our project's cabal file specifies the executable quiz-server, our normal command would be stack exec quiz-server. You might expect we would accomplish this with RUN stack exec quiz-server. However, we actually want to use CMD instead of RUN:

FROM haskell:9.2.5
...

CMD stack exec quiz-server

If we were to use RUN, then the command would be run while building the docker container. Since the command is a web server that listens indefinitely, this means the build step will never complete, and we'll never get our image! However, by using CMD, this command will happen when we run the container, not when we build the container.

Here's our final Dockerfile (which we have to save as "Dockerfile" in our project root directory):

FROM haskell:9.2.5

COPY . /app

WORKDIR /app

RUN stack setup && stack build --only-dependencies
RUN stack build

EXPOSE 8080
CMD stack exec quiz-server

Creating the Image

Once we have finished our Dockerfile, we still need to build it to create the image we can deploy elsewhere. To do this, you need to make sure you have Docker installed on your local system. Then you can use the docker build command to create a local image.

>> docker build -t quiz-server .

You can then see the image you created with the docker images command!

>> docker images
REPOSITORY TAG    IMAGE ID ...
quiz-server latest abcdef123456 ...

If you want, you can then run your application locally with the docker run command! The key thing with a web server is that you have to use the -p argument to make sure that the exposed ports on the docker container are then re-exposed on your local machine. It's possible to use a different port locally, but for our purposes, we'll just use 8080 for both like so:

>> docker run -it -p 8080:8080 --rm quiz-server

Conclusion

This creates a local docker image for us. But this isn't enough to run the program anywhere on the web! Next time we'll upload this image to a service to deploy our application to the internet!

If you want to keep up with this series, make sure to subscribe to our monthly newsletter!

3 More Advent of Code Videos!

I'm almost done with the Advent of Code recap videos. This week you can now find the reviews for Days 21, 22, and 23 by going to my YouTube Channel. As a reminder, you can find links to all Advent of Code solutions on this page.

Day 21 featured a recursive variable tree problem. See the video here or read the write-up.

Day 22 started out as an innocuous 2D maze traversal problem, but ended up being waaay more complicated in part 2, which puts your geometry skills to the test and forces some very precise programming. Watch the recap or read about it.

Finally, Day 23 was the final state evolution problem. Our elf friends want to plant a grove of trees, so they would like to spread out in an optimal fashion. Here's the video, and here's the blog post.

If these recaps have helped you improve your Haskell problem solving skills, make sure the subscribe to our monthly newsletter! We'll have a lot more problem-solving content throughout the year, so stay tuned for that!

What's Next in 2023?

To start the year I wrote a few different articles about Chat GPT looking at what it thinks about Haskell. After some extra reflection, my emotional reaction probably matched that of a lot of other programmers. I felt plenty of excitement about the tasks that could become a lot easier using AI tools, but also a twinge of existential panic over the prospect of being replaced as a programmer. This latter feeling was especially acute when I thought about the future of this blog.

After all, what's the point of writing a blog post about a topic when someone can type their question into Chat GPT, get a response, and then even have a coherent conversation about the details?

However, there are still plenty of things that Chat GPT isn't able to help with yet. And as I'm planning out what I'm going to write about over the next year, I want to focus on material that isn't quite so replaceable.

The most pithy distinction that will be guiding me is show, don't tell. AI can tell you what you really want to know, but it's still going to be a long time before it can really show it to you in the context you want to use it. Even if you can have a coherent conversation with a chatbot, this doesn't mean it can adequately respond to all the needs you can present for a larger project. You can't just feed in your whole codebase to Chat GPT and have it tell you what you need to write next.

So on the blog this year, you can expect a lot of content will be more project-based. I'll be working on some larger, overarching ideas. I'll often be presenting these more through video than written style, since a video presents more of the context and struggles you can expect while actually creating something.

Another distinction with AI is that it can tell you what to learn, but not necessarily how to learn it. It doesn't necessarily know how humans acquire knowledge, especially in something as complex as programming. So you can expect more course-related content in this area, but I'll be aiming for multiple shorter courses (including one or two free courses) this year to complement the existing, longer course offerings.

If you want to keep up to date with all this, make sure to subscribe to our monthly newsletter!

Advent of Code: Fetching Puzzle Input using the API

When solving Advent of Code problems, my first step is always to access the full puzzle input and copy it into a file on my local system. This doesn't actually take very long, but it's still fun to see how we can automate it! In today's article, we'll write some simple Haskell code to make a network request to find this data.

We'll write a function that can take a particular year and day (like 2022 Day 5), and save the puzzle input for that day to a file that the rest of our code can use.

As a note, there's a complete Advent of Code API that allows you to do much more than access the puzzle input. You can submit your input, view leaderboards, and all kinds of other things. There's an existing Haskell library for all this, written in 2019. But we'll just be writing a small amount of code from scratch in this article, rather than using this library.

Authentication

In order to get the puzzle input for a certain day, you must be authenticated with Advent of Code. This typically means logging in with GitHub or another service. This saves a session cookie in your browser that is sent with every request you make to the site.

Our code needs to access this cookie somehow. It's theoretically possible to do this in an automated way by accessing your browser's data. For this example though, I found it easier to just copy the session token manually and save it as an environment variable. The token doesn't change as long as you don't log out, so you can keep reusing it.

This GitHub issue gives a good explanation with images for how to access this token using a browser like Google Chrome. At a high level, these are the steps:

1. Log in to Advent of Code and access and puzzle input page (e.g. http://adventofcode.com/2022/day/1/input)
2. Right click the page and click "inspect"
3. Navigate to the "Network" tab
4. Click on any request, and go to the "Headers" tab
5. Search through the "Request Headers" for a header named cookie.
6. You should find one value that starts with session=, followed by a long string of hexadecimal characters. Copy the whole value, starting with session= and including all the hex characters until you hit a semicolon.
7. Save this value as an environment variable on your system using the name AOC_TOKEN.

The rest of the code will assume you have this session token (starting with the string session=) saved as the variable AOC_TOKEN in your environment. So for example, on my Windows Linux subsystem, I have a line like this in my .bashrc:

export AOC_TOKEN="session=12345abc..."

We're now ready to start writing some code!

Caching

Now before we jump into any shenanigans with network code, let's first write a caching function. All this will do is see if a specified file already exists and has data. We don't want to send unnecessary network requests (the puzzle input never changes), so once we have our data locally, we can short circuit our process.

So this function will take our FilePath and just return a boolean value. We first ensure the file exists.

checkFileExistsWithData :: FilePath -> IO Bool
checkFileExistsWithData fp = do
  exists <- doesFileExist fp
  if not exists
    then return False
    ...

As long as the file exists, we'll also ensure that it isn't empty.

checkFileExistsWithData :: FilePath -> IO Bool
checkFileExistsWithData fp = do
  exists <- doesFileExist fp
  if not exists
    then return False
    else do
      size <- getFileSize fp
      return $ size > 0

If there's any data there, we return True. Otherwise, we need to fetch the data from the API!

Setting Up the Function

Before we dive into the specifics of sending a network request, let's specify what our function will do. We'll take 3 inputs for this function:

1. The problem year (e.g. 2022)
2. The problem day (1-25)
3. The file path to store the data

Here's what that type signature looks like:

fetchInputToFile :: (MonadLogger m, MonadThrow m, MonadIO m)
  => Int -- Year
  -> Int -- Day
  -> FilePath -- Destination File
  -> m ()

We'll need MonadIO for reading and writing to files, as well as reading environment variables. Using a MonadLogger allows us to tell the user some helpful information about whether the process worked, and MonadThrow is needed by our network library when parsing the route.

Now let's kick this function off with some setup tasks. We'll first run our caching check, and we'll also look for the session token as an environment variable.

fetchInputToFile :: (MonadLogger m, MonadThrow m, MonadIO m) => Int -> Int -> FilePath -> m ()
fetchInputToFile year day filepath = do
  isCached <- liftIO $ checkFileExistsWithData filepath
  token' <- liftIO $ lookupEnv "AOC_TOKEN"
  case (isCached, token') of
    (True, _) -> logDebugN "Input is cached!"
    (False, Nothing) -> logErrorN "Not cached but didn't find session token!"
    (False, Just token) -> ...

If it's cached, we can just return immediately. The file should already contain our data. If it isn't cached and we don't have a token, we're still forced to "do nothing" but we'll log an error message for the user.

Now we can move on to the network specific tasks.

Making the Network Request

Now let's prepare to actually send our request. We'll do this using the Network.HTTP.Simple library. We'll use four of its functions to create, send, and parse our request.

parseRequest :: MonadThrow m => String -> m Request

addRequestHeader :: HeaderName -> ByteString -> Request -> Request

httpBS :: MonadIO m => Request -> m (Response ByteString)

getResponseBody :: Response a -> a

Here's what these do:

1. parseRequest generates a base request using the given route string (e.g. http://www.adventofcode.com)
2. addRequestHeader adds a header to the request. We'll use this for our session token.
3. httpBS sends the request and gives us a response containing a ByteString.
4. getResponseBody just pulls the main content out of the Response object.

When using this library for other tasks, you'd probably use httpJSON to translate the response to any object you can parse from JSON. However, the puzzle input pages are luckily just raw data we can write to a file, without having any HTML wrapping or anything like that.

So let's pick our fetchInput function back up where we left off, and start by creating our "base" request. We determine the proper "route" for the request using the year and the day. Then we use parseRequest to make this base request.

fetchInputToFile :: (MonadLogger m, MonadThrow m, MonadIO m) => Int -> Int -> FilePath -> m ()
fetchInputToFile year day filepath = do
  isCached <- liftIO $ checkFileExistsWithData filepath
  token' <- liftIO $ lookupEnv "AOC_TOKEN"
  case (isCached, token') of
    ...
    (False, Just token) -> do
      let route = "https://adventofcode.com/" <> show year <> "/day/" <> show day <> "/input"
      baseRequest <- parseRequest route
      ...

Now we need to modify the request to incorporate the token we fetched from the environment. We add it using the addRequestHeader function with the cookie field. Note we have to pack our token into a ByteString.

import Data.ByteString.Char8 (pack)

fetchInputToFile :: (MonadLogger m, MonadThrow m, MonadIO m) => Int -> Int -> FilePath -> m ()
fetchInputToFile year day filepath = do
  isCached <- liftIO $ checkFileExistsWithData filepath
  token' <- liftIO $ lookupEnv "AOC_TOKEN"
  case (isCached, token') of
    ...
    (False, Just token) -> do
      let route = "https://adventofcode.com/" <> show year <> "/day/" <> show day <> "/input"
      baseRequest <- parseRequest route
      {- Add Request Header -}
      let finalRequest = addRequestHeader "cookie" (pack token) baseRequest 
      ...

Finally, we send the request with httpBS to get its response. We unwrap the response with getResponseBody, and then write that output to a file.

fetchInputToFile :: (MonadLogger m, MonadThrow m, MonadIO m) => Int -> Int -> FilePath -> m ()
fetchInputToFile year day filepath = do
  isCached <- liftIO $ checkFileExistsWithData filepath
  token' <- liftIO $ lookupEnv "AOC_TOKEN"
  case (isCached, token') of
    (True, _) -> logDebugN "Input is cached!"
    (False, Nothing) -> logErrorN "Not cached but didn't find session token!"
    (False, Just token) -> do
      let route = "https://adventofcode.com/" <> show year <> "/day/" <> show day <> "/input"
      baseRequest <- parseRequest route
      let finalRequest = addRequestHeader "cookie" (pack token) baseRequest 
      {- Send request, retrieve body from response -}
      response <- getResponseBody <$> httpBS finalRequest
      {- Write body to the file -}
      liftIO $ Data.ByteString.writeFile filepath response

And now we're done! We can bring this function up in a GHCI session and run it a couple times!

>> import Control.Monad.Logger
>> runStdoutLoggingT (fetchInputToFile 2022 1 "day_1_test.txt")

This results in the puzzle input (for Day 1 of this past year) appearing in the `day_1_test.txt" file in our current directory! We can run the function again and we'll find that it is cached, so no network request is necessary:

>> runStdoutLoggingT (fetchInputToFile 2022 1 "day_1_test.txt")
[Debug] Retrieving input from cache!

Now we've got a neat little function we can use each day to get the input!

Conclusion

To see all this code online, you can read the file on GitHub. This will be the last Advent of Code article for a little while, though I'll be continuing with video walkthrough catch-up on Thursdays. I'm sure I'll come back to it before too long, since there's a lot of depth to explore, especially with the harder problems.

If you're enjoying this content, make sure to subscribe to Monday Morning Haskell! If you sign up for our mailing list, you'll get our monthly newsletter, as well as access to our Subscriber Resources!

Advent of Code: Days 19 & 20 Videos

You can now find two more Advent of Code videos on our YouTube channel! You can see a summary of all my work on the Advent of Code summary page!

Here is the video for Day 19, another difficult graph problem involving efficient use of resources! (The write-up can be found here).

And here's the Day 20 video, a tricky problem involving iterated sequences of numbers. You can also read about the problem in our writeup.

Reflections on Advent of Code 2022

Now that I've had a couple weeks off from Advent of Code, I wanted to reflect a bit on some of the lessons I learned after my second year of doing all the puzzles. In this article I'll list some of the things that worked really well for me in my preparation so that I could solve a lot of the problems quickly! Hopefully you can learn from these ideas if you're still new to using Haskell for problem solving.

Things that Worked

File Template and Tests

In 2021, my code got very disorganized because I kept reusing the same module, but eventually needed to pick an arbitrary point to start writing a new module. So this year I prepared my project by having a Template File with a lot of boilerplate already in place. Then I prepared separate modules for each day, as well as a test-harness to run the code. This simplified the process so that I could start each problem by just copying and pasting the inputs into pre-existing files and then run the code by running the test command. This helped me get started a lot faster each day.

Megaparsec

It took me a while to get fluent with the Megaparsec library. But once I got used to it, it turned out to be a substantial improvement over the previous year when I did most of the parsing by hand. Some of my favorite perks of using this library included easy handling of alternatives and recursive parsing. I highly recommend it.

Utilities

Throughout this year's contest, I relied a lot on a Utilities file that I wrote based on my experience from 2021. This included refactored code from that year for a lot of common use cases like 2D grid algorithms. It especially included some parsing functions for common cases like numbers and grids. Having these kinds of functions at my disposal meant I could focus a lot more on core algorithms instead of details.

List Library Functions

This year solidified my belief that the Data.List library is one of the most important tools for basic problem solving in Haskell. For many problems, the simplest solution for a certain part of the problem often involved chaining together a bunch of list functions, whether from that library or just list-related functions in Prelude. For just a couple examples, see my code for Day 1 and Day 11

Earlier in 2022 I did several articles on List functions and this was very useful practice for me. Still, there were times when I needed to consult the documentation, so I need more practice to get more fluent! It's best if you can recall these functions from memory so you can apply them quickly!

Data Structures

My series on Data Structures was also great practice. While I rarely had trouble selecting the right data structure back in 2021, I felt a lot more fluent this year applying data structure functions without needing to consult documentation. So I highly recommend reading my series and getting used to Haskell's data structure APIs! In particular, learning how to fold with your data structures will make your code a lot smoother.

Folds and For Loops

Speaking of folds, it's extremely important to become fluent with different ways of using folds in your solutions. In most languages, for-loops will be involved in a lot of complex tasks, and in Haskell folds are the most common replacement for for-loops. So get used to using them, including in a nested fashion!

In fact, the pattern of "loop through each line of input" is so common in Advent of Code that I had some lines in my file template for this solution pattern! (Incidentally, I also had a second solution pattern for "state evolution" which was also useful).

Graph Algorithms

Advent of Code always seems to have graph problems. The Algorithm.Search library helps make these a lot easier to work through. Learn how to use it! I first stumbled it on it while writing about Dijkstra's Algorithm last year.

The Logger Monad

I wrote last year about wanting to use the Logger monad to make it easier to debug my code. This led to almost all of my AoC code this year using it, even when it wasn't necessary. Overall, I think this was worth it! I was able to debug problem much faster this year when I ran into issues. I didn't get stuck needing to go back and add monads into pure code or trying to use the trace library.

Things I Missed

There were still some bumps in the road though! Here are a couple areas where I needed improvement.

Blank Lines in Input

Certain problems, like Day 1, Day 11, Day 13 and Day 19, incorporated blank lines in the input to delineate between different sections.

For whatever reason, it seems like 2021 didn't have this pattern, so I didn't have a utility for dealing with this pattern and spent an inordinate amount of time dealing with it, even though it's not actually too challenging.

Pruning Graph Algorithms

Probably the more important miss in my preparation was how I handled the graph problems. In last year's edition, there were only a couple graph problems, and I was able to get answers reasonably quickly simply by applying Dijkstra's algorithm.

This year, there were three problems (Day 16, Day 19 & Day 24) that relied heavily on graph algorithms. The first two of these were generally considered among the hardest problems this year.

For both of these, it was easy enough to frame them as graph problems and apply an algorithm like Dijkstra's or BFS. However, the scale of the problem proved to be too large to finish in a reasonable amount of time with these algorithms.

The key in all three problems was to apply pruning to the search. That is, you needed to be proactive in limiting the search space to avoid spending a lot of time on search states that can't produce the optimal answer.

Haskell's Algorithm.Search library I mentioned earlier actually provides a great tool for this! The pruning and pruningM functions allow you to modify your normal search function with an additional predicate to avoid unnecessary states.

I also learned more about the A-Star algorithm than I had known before. Previously, I'd vaguely thought of A-Star as "Dijkstra with a heuristic", which is kind of correct but not helpful in describing what the heuristic actually does.

I eventually came to the understanding that Dijkstra is comparable to Breadth-First-Search for a weighted graph. A-Star is closer to a Depth-First-Search on a weighted graph. While a normal DFS blindly goes down a path until it finds an end-state, A-Star uses the heuristic to attempt to make sure the first parts of the graph that we search are the most likely to lead us to the goal.

It was difficult to find a heuristic for Days 16 & 19, but I was able to apply A-Star on Day 24 to get a noticeable speed-up.

Conclusion

For the next few Thursdays I'm still going to be catching up on the video walkthroughs from Advent of Code. And then next Monday I'll have one more AoC-related article exploring how we can build an API to access our question inputs without even using the web browser!

Make sure to subscribe to our mailing list if you haven't already to get access to our Subscriber Resources!

Advent of Code: Days 17 & 18 Videos

You can now take a look at my latest videos recapping Advent of Code from last year! If you missed any of the write-ups or videos, you can head to the summary page!

Here’s the Day 17 video, another challenging problem where we had to essentially write a tetris simulator!

And here’s the Day 18 video. This one was much easier than days 16 & 17. The key component here was performing a breadth-first-search in 3D. Here’s the writeup if you missed it.

Writing Haskell with Chat GPT

In our last couple of articles, we've asked Chat GPT a bunch of questions about Haskell and functional programming. Today we're going to do one more exploration to see what kinds of Haskell code this chatbot can write.

Hello World

We'll start with a simple "Hello World" program.

How do I write a "Hello World" program in Haskell?

Chat GPT:

To write a "Hello World" program in Haskell, you can use the putStrLn function from the Prelude, which is Haskell's default library. This function takes a string as an argument and prints it to the console. Here's an example of a simple "Hello World" program in Haskell:

main :: IO ()
main = putStrLn "Hello, World!"

This program defines a function main that has the type IO (). The IO type is used for actions that perform input/output, and the () type represents the empty tuple, which has no contents. The main function is special in Haskell because it is the entry point of every executable Haskell program. When the program is run, the main function will be called and the string "Hello, World!" will be printed to the console.

You can compile and run this program using the ghc compiler, which is included with the Haskell Platform. First, save the program to a file named HelloWorld.hs. Then, open a terminal and navigate to the directory where you saved the file. To compile the program, type the following command:

ghc HelloWorld.hs

This will produce an executable file named HelloWorld. To run the program, type the following command:

./HelloWorld

This will print "Hello, World!" to the console.

Commentary

This not only gives the code to run the program, it also gives detailed instructions for running it using GHC. Recommending Stack and GHCup would be better advice for beginners though. I'm sure it would know how to describe using these tools if asked about them. But it would be best if it knew to recommend these tools to someone who didn't know about them. Still, I'm nit-picking a bit here.

Sorting Algorithms

So a "Hello World" program is one of the simplest things you can do though. How does ChatGPT do with writing some algorithmic code? Let's start with a couple simple sorting algorithms, quicksort and mergesort.

How can I write quicksort in Haskell?

Chat GPT:

Here is an implementation of quicksort in Haskell:

quicksort :: Ord a => [a] -> [a]
quicksort [] = []
quicksort (x:xs) = quicksort [y | y <- xs, y <= x] ++ [x] ++ quicksort [y | y <- xs, y > x]

This function defines a quicksort that sorts a list of elements that are part of the Ord type class (which includes types like Int, Char, and Double). The function uses a recursive approach to sort the list.

The base case of the recursion is the empty list, which is already sorted. For the recursive case, the function divides the list into three parts:

1. A sublist of elements that are less than or equal to the pivot element (x)
2. The pivot element itself
3. A sublist of elements that are greater than the pivot element

The function then recursively sorts the first and third sublists, and concatenates the sorted sublists with the pivot element to produce the final sorted list.

You can test this implementation by calling quicksort with a list of elements:

quicksort [4, 6, 9, 8, 3, 5, 1, 7, 2]

This should return the sorted list [1, 2, 3, 4, 5, 6, 7, 8, 9].

How can I write mergesort in Haskell?

Chat GPT:

Here is an implementation of mergesort in Haskell:

merge :: Ord a => [a] -> [a] -> [a]
merge xs [] = xs
merge [] ys = ys
merge (x:xs) (y:ys)
  | x <= y    = x : merge xs (y:ys)
  | otherwise = y : merge (x:xs) ys

mergesort :: Ord a => [a] -> [a]
mergesort []  = []
mergesort [x] = [x]
mergesort xs  = merge (mergesort left) (mergesort right)
  where
    mid   = length xs `div` 2
    left  = take mid xs
    right = drop mid xs

This implementation defines a merge function that merges two sorted lists into a single sorted list, and a mergesort function that sorts a list using the merge sort algorithm.

The merge function works by comparing the head elements of the two input lists, and adding the smaller of the two to the result list. It repeats this process until one of the input lists is empty, at which point it appends the remaining elements of the other list to the result.

The mergesort function works by dividing the input list into two halves, sorting each half using a recursive call to mergesort, and then merging the two sorted halves using the merge function.

You can test this implementation by calling mergesort with a list of elements:

mergesort [4, 6, 9, 8, 3, 5, 1, 7, 2]

This should return the sorted list [1, 2, 3, 4, 5, 6, 7, 8, 9].

Commentary

For quicksort, it comes up with a simple 3-line solution, and the mergesort solution is also quite succinct. However, one change I would make to the solution is to tail recursion. As with the "Hello, World" example though, the descriptions are very helpful for someone just learning the language.

Heap Sort

Quicksort and mergesort are one thing. These algorithms have fairly straightforward recursive implementations that work well in Haskell. What about heapsort?

How can I write heapsort in Haskell?

Chat GPT:

Here is an implementation of heapsort in Haskell:

heapify :: Ord a => [a] -> [a]
heapify xs = heapify' (length xs `div` 2 - 1) xs
  where
    heapify' (-1) xs = xs
    heapify' i xs = heapify' (i - 1) (heapifyOne i xs)

heapifyOne :: Ord a => Int -> [a] -> [a]
heapifyOne i xs = case cmp of
                    GT -> xs
                    _  -> heapifyOne' i cmp xs
  where
    cmp = compare (xs !! i) (xs !! child)
    child = if r < length xs && xs !! r > xs !! l then r else l
    l = 2 * i + 1
    r = 2 * i + 2

heapifyOne' i GT xs = xs
heapifyOne' i _  xs = heapifyOne (if r < length xs && xs !! r > xs !! l then r else l) $ swap i (if r < length xs && xs !! r > xs !! l then r else l) xs
  where
    l = 2 * i + 1
    r = 2 * i + 2

swap i j xs = [if k == i then xs !! j else if k == j then xs !! i else xs !! k | k <- [0..length xs - 1]]

sort :: Ord a => [a] -> [a]
sort [] = []
sort xs = y : sort ys
  where
    (y:ys) = heapify xs

This implementation defines a heapify function that turns a list into a heap, and a sort function that sorts a list using the heap sort algorithm.

The heapify function works by recursively transforming the list into a heap, starting from the middle of the list and working towards the beginning. It does this by repeatedly calling the heapifyOne function, which restores the heap property to a single element and its children in the heap.

The sort function works by repeatedly extracting the maximum element (the root of the heap) and placing it at the end of the result list, until the heap is empty.

You can test this implementation by calling sort with a list of elements:

sort [4, 6, 9, 8, 3, 5, 1, 7, 2]

This should return the sorted list [1, 2, 3, 4, 5, 6, 7, 8, 9].

Commentary

Very impressive, especially with the detailed description! This time, we do see tail recursion applied in the heapify function.

List Functions

How does it do with a more abstract prompt? I came up with this question based on Day 1 of the recently concluded 2022 Advent of Code.

Write a Haskell function to solve the following problem. The input is a list of lists of integers. We want to find the three lists with the largest sums, and add their sums together. So the output is just an integer.

Chat GPT:

Here is a function that solves the problem you described:

largestThreeSums :: [[Int]] -> Int
largestThreeSums xs = sum $ take 3 $ sortBy (flip compare) $ map sum xs

This function first applies the sum function to each sublist in xs, producing a list of sums. It then sorts this list in descending order using the sortBy function, and takes the first three elements using the take function. Finally, it sums these three elements using the sum function and returns the result.

To use this function, you can call it with a list of lists of integers as an argument:

largestThreeSums [[1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12]]

This should return the result 33, which is the sum of the three largest sublists [7, 8, 9], [10, 11, 12], and [4, 5, 6].

Commentary

This is quite similar to my solution! So Chat GPT did an effective job parsing my description, and determining the right list functions to use to make an efficient, one-line solution.

Making a Server

For one more example, let's try doing something more practical, like a web server. This will require using libraries beyond Haskell's base packages. Once again, this works quite well!

How can I write a basic HTTP server in Haskell?

Chat GPT:

Here is an example of a basic HTTP server in Haskell using the wai and wai-extra libraries:

{-# LANGUAGE OverloadedStrings #-}

import Network.Wai
import Network.Wai.Handler.Warp
import Network.HTTP.Types (status200)
import Blaze.ByteString.Builder (copyByteString)
import qualified Data.ByteString.UTF8 as BU

app :: Application
app request respond = respond $
    case rawPathInfo request of
        "/hello" -> hello
        _        -> notFound
  where
    hello = responseBuilder
        status200
        [("Content-Type", "text/plain")]
        (copyByteString "Hello, World!")
    notFound = responseBuilder
        status404
        [("Content-Type", "text/plain")]
        (copyByteString "Not Found")

main :: IO ()
main = do
    putStrLn "Listening on port 3000"
    run 3000 app

This server listens for incoming HTTP requests on port 3000, and responds with a "Hello, World!" message for requests to the /hello path, and a "Not Found" message for all other paths.

To run this server, save the code to a file and compile it using the ghc compiler:

ghc Server.hs

This will produce an executable file named Server. To start the server, run the executable:

./Server

You should see the message "Listening on port 3000" printed to the console. You can then send HTTP requests to the server using a tool like curl or by opening a web browser and going to http://localhost:3000/hello. You should receive a "Hello, World!" response for the /hello path, and a "Not Found" response for all other paths.

Conclusion

So overall, Chat GPT does quite well with these basic challenges! It would be interesting to take this further still and see if we could make our server program more and more complex, like adding custom functionality for different routes. But Chat GPT definitely seems useful enough to help with basic tasks, even in a less well-known language like Haskell!

As a final note, I did check (out of curiosity) to see if Chat GPT seemed to know much about this blog. The answer appeared to be a resounding "no", leading to a good laugh for me! I asked, "do you know about Monday Morning Haskell?" This was the response I got from Chat GPT (emphasis is mine):

Yes, I am familiar with the term "Monday Morning Haskell." It refers to a programming style or approach that involves using the Haskell programming language to solve problems or write software on a Monday morning, with the implication being that Haskell is a particularly well-suited language for starting the work week off with a clear and focused mind. Haskell is a functional programming language known for its strong static type system, expressive type inference, and purity (lack of side effects). Many developers find that programming in Haskell can help them write clearer, more concise, and more maintainable code, and that it is especially well-suited for tasks that require reasoning about abstract concepts or dealing with complex data.

I might just have to add that description to the home page!

Advent of Code Video Catchup Begins!

If you were following my Advent of Code series from December, you may have noticed that I stopped publishing videos after Day 14 or so. Unfortunately, the problems got challenging, so I didn't have time for videos in addition to writeups. I also went on vacation for Christmas, so I was away from my recording setup for a while.

But now I'm hoping to catch up with those videos! I'll be releasing two each week on Wednesday and Thursday. So today, you can now go to my YouTube channel and see the videos for Day 15 and Day 16!

If you enjoy the Advent of Code series and all the writeups I did, make sure to subscribe to our monthly newsletter!

Day 15

Day 15 Writeup

Day 15 Video link

Day 16

Day 16 Writeup

Day 16 Video link

Note, the Day 16 writeup is combined with Day 17 and gives only a high-level overview.

Chat GPT and Functional Programming

Last week I explored what Chat GPT has to say about monads. But it's interesting to go beyond this single area and see what its thoughts are about functional programming (FP) in general, and how different languages fit within this paradigm.

So I asked the bot a few questions in these areas to see its response. In each section, the question is either the header, or if it is a longer question, it is given as a quote below the header. This will then be followed by Chat GPT's answer. I'll then follow most sections with some commentary.

We'll see that despite providing a lot of useful knowledge, the bot still sometimes struggles with consistency and the handling of ambiguous and contentious issues.

What is Functional Programming?

Chat GPT:

Functional programming is a programming paradigm that treats computation as the evaluation of mathematical functions. It is based on the idea of immutability and the avoidance of changing state and mutable data. In functional programming, functions are first-class citizens, meaning that they can be passed as arguments to other functions and returned as values from functions. This allows for the creation of higher-order functions that take other functions as input or return them as output.

Functional programming languages typically provide features such as higher-order functions, anonymous functions, and lazy evaluation to support the functional programming paradigm. Some examples of functional programming languages include Lisp, ML, Haskell, and Erlang.

Functional programming has a number of advantages, including the ability to write code that is easier to reason about, easier to test, and less prone to errors. It can also be more efficient in some cases, since it avoids the overhead of mutable state and allows for the use of techniques such as memoization and parallelization. However, it can also be more difficult to understand and learn for programmers who are used to imperative programming styles.

Commentary

So we start with an abstract definition: FP is a paradigm that views programming as the evaluation of mathematical functions. Then we get a slew of features (immutable state, functions as first class citizens, etc.). The features it lists include lazy evaluation, which is noteworthy.

It's interesting though to compare the advantages and disadvantages the bot mentions. It says that functional code is "easier" to reason about and test. It also lists efficiency as an important feature.

It then claims a disadvantage is that it is more difficult to understand...but mainly for people coming from imperative styles. So overall it should be easier to understand, but programmers who have been trained in other systems have a difficult time migrating to FP.

We'll come back to a number of these points later. But first, let's ask about a couple other languages and how they relate to functional programming.

Is C++ a Functional Language?

Chat GPT:

C++ is a multi-paradigm programming language, meaning that it supports several programming paradigms, including functional programming. However, C++ is not a purely functional language, as it also supports other paradigms such as imperative programming and object-oriented programming.

While C++ does not have some of the features that are commonly found in functional programming languages, such as first-class functions and lazy evaluation, it does support the creation of functions and the use of function pointers, which allows for the use of functional programming techniques to some extent.

For example, it is possible to write code in C++ that uses higher-order functions, anonymous functions, and recursive functions, which are all common features of functional programming languages. However, C++ does not have some of the other features that are commonly found in functional programming languages, such as pattern matching and algebraic data types.

Is Rust a Functional Language?

Chat GPT:

Like C++, Rust is a multi-paradigm programming language that supports functional programming, among other paradigms. Rust provides a number of features that make it suitable for functional programming, such as first-class functions, pattern matching, and algebraic data types.

However, Rust is not a purely functional language, as it also supports imperative programming and object-oriented programming. Rust's support for functional programming is more limited than that of languages that were specifically designed for functional programming, such as Haskell or ML.

Despite this, Rust's support for functional programming is sufficient to allow developers to write code in a functional style, and many Rust developers use functional programming techniques in their Rust code.

Commentary

What we see here is the difference between "what is functional programming" and "what is a functional language". C++ and Rust both allow you to use functional programming techniques even if FP isn't their primary paradigm. They're both intended to be flexible, multi-paradigm languages.

However, Rust facilitates FP much more through specific features, specifically abstract data types and pattern matching. Features like these are definitely an important part of why, culturally speaking, Rust programmers are more likely to employ functional programming ideas in their code.

What about the reverse? Does Chat GPT think we can use Haskell for Object Oriented Programming (OOP)?

Can Haskell be used in an object-oriented way?

Chat GPT:

Haskell is a purely functional programming language, which means that it does not support object-oriented programming as a primary paradigm. However, that does not necessarily mean that it is impossible to write object-oriented code in Haskell.

One way to write object-oriented code in Haskell is to use a typeclass-based approach, where types are defined as data types and behaviors are defined as functions on those data types. This can provide a way to define classes and objects, and to define methods on those objects.

Another way to write object-oriented code in Haskell is to use the "Object I/O" library, which provides a set of libraries and tools for building graphical user interface (GUI) applications in Haskell using an object-oriented style. This library allows developers to define classes, objects, and methods in a way that is similar to object-oriented languages such as Java or C++.

Overall, while Haskell is not a language that was designed for object-oriented programming, it is possible to use it in an object-oriented way with some effort. However, the functional programming paradigm is more idiomatic in Haskell, and most Haskell code is written in a functional style.

Commentary

This answer arrives at a conclusion I've had for a while. Typeclasses provide most of the abstraction you really need to get a lot of the benefits of OOP. Of course, they don't inherently provide stateful mutations, which are implicit in OOP, but this is possible to include in Haskell with the State monad.

It is interesting to see the trend that FP is a "more restrictive" paradigm. FP languages don't naturally address OOP as easily or as often as OOP languages allow functional ideas. But I would spin this as a positive for FP. Java and C++ have added more functional features (especially with lambda syntax) in the last few years because functional ideas are inherently useful. Haskell doesn't feel the need to incorporate OOP ideas as much because it doesn't need them.

Lazy Evaluation

In response to the first question, the bot included "lazy evaluation" as one of the features of functional programming, even though this is more specific to a few languages than to the whole paradigm. So I asked the following question:

Is lazy evaluation a defining characteristic of functional programming? Are there functional languages that use eager evaluation?

Chat GPT:

Lazy evaluation is a common feature of functional programming languages, but it is not a defining characteristic of functional programming. Lazy evaluation is a evaluation strategy that delays the evaluation of an expression until it is needed, rather than evaluating it as soon as it is bound to a variable.

Lazy evaluation can be useful in functional programming because it allows for the creation of infinite data structures, and it can also lead to more efficient code in some cases by avoiding the computation of unnecessary expressions. However, lazy evaluation is not a requirement for a language to be considered functional.

There are indeed functional programming languages that use eager evaluation, which is the opposite of lazy evaluation. Eager evaluation evaluates expressions as soon as they are bound to a variable, rather than delaying their evaluation until they are needed.

Some examples of functional programming languages that use eager evaluation include Miranda and Mercury. These languages provide many of the same features as lazy functional languages, such as higher-order functions and immutable data structures, but they use eager evaluation instead of lazy evaluation.

Commentary

So it seems like I've caught the bot in a bit of inconsistency. Lazy evaluation is not actually a necessary part of FP (like it implied in its first answer). But it's certainly more common in FP-oriented languages.

Disadvantages of FP

You said functional programming is more difficult to understand for programmers used to an imperative style. Are there any other disadvantages to functional programming?

Chat GPT:

Functional programming can be more difficult to understand for programmers who are used to imperative programming styles because it involves a different way of thinking about computation. In functional programming, the emphasis is on expressing the desired result of a computation rather than specifying how the computation should be performed, which can be a shift for programmers who are used to an imperative approach.

There are a few other potential disadvantages to functional programming to consider:

1. Performance: In some cases, functional code can be less efficient than imperative code, particularly if it makes heavy use of higher-order functions or lazy evaluation.
2. Debugging: Debugging functional code can be more difficult in some cases because it can be harder to understand the flow of control in a functional program, especially if the program makes heavy use of higher-order functions or anonymous functions.
3. Lack of familiar libraries: Depending on the language, the functional programming ecosystem may not have as many libraries and frameworks available as other paradigms. This can make it more difficult to build certain types of applications, and may require more work to integrate functional code with existing systems.
4. Limited adoption: Functional programming is not as widely used as imperative programming, which means that there may be a smaller pool of developers with functional programming experience to draw from. This can make it more difficult to find developers with functional programming skills, and can also make it more difficult to find resources and documentation when learning functional programming.

It's worth noting that these disadvantages are not unique to functional programming, and many of them can be mitigated with the right tools and techniques. Ultimately, the choice of programming paradigm should be based on the needs of the project and the preferences of the development team.

Commentary

So two of the 4 disadvantages still mainly come down to "Haskell/FP are hard because fewer people use them". If there was wider adoption (point #4), then there would be more familiar libraries (point #3).

It's curious though that performance is now listed as a disadvantage when "efficiency" was given as an advantage before! But both points are salient. Immutability does make parallelization easier, but a variety of issues can make FP less efficient, especially in Haskell, where a simple action like changing a value in a contiguous array might require re-copying the whole array.

Debugging is definitely an important issue to list as a disadvantage. Besides GHCI, I haven't used a specific debugging tool for Haskell with much success, so this is a good topic to explore.

In this discussion of the advantages and disadvantages of FP, I thought it would be a good idea to check on formal research in the area. Here's what Chat GPT had to say:

Research

You said functional programming can be easier to reason about and easier to test. Are there experiments or research comparing functional programming and other paradigms in these areas?

Chat GPT:

There have been a number of studies and surveys conducted over the years that have compared functional programming with other paradigms in terms of readability, maintainability, and other factors. These studies have generally found that functional programming can be easier to reason about and easier to test in some cases, although the results are not always clear-cut and can depend on a variety of factors.

For example, a study published in the Journal of Systems and Software in 2002 compared the maintainability of functional, imperative, and object-oriented programs and found that functional programs had fewer maintenance-related defects and were easier to understand than imperative programs.

Another study published in the Journal of Functional Programming in 2005 surveyed developers who had used both functional and imperative programming languages and found that the functional programmers reported fewer debugging and testing problems, and felt that their programs were easier to understand and modify.

Overall, it seems that functional programming can offer some benefits in terms of readability and testability, but more research is needed to determine the extent to which these benefits hold up in practice. It's also important to note that the choice of programming paradigm is just one factor that can impact the readability and maintainability of code, and other factors such as code style, architecture, and the use of good design practices can also be important.

Are There More Recent Studies?

Chat GPT:

There have been a number of more recent studies and surveys that have compared functional programming with other paradigms in terms of readability, maintainability, and other factors.

For example, a study published in the Journal of Systems and Software in 2015 examined the effect of functional programming on the reliability of software systems and found that functional programs had fewer defects and were easier to understand than imperative programs.

Another survey published in the ACM SIGPLAN Haskell Symposium in 2018 surveyed developers who had used both functional and imperative programming languages and found that the functional programmers reported fewer debugging problems and felt that their programs were easier to understand and modify.

Overall, it seems that the results of these more recent studies are consistent with the findings of earlier studies, which is that functional programming can offer some benefits in terms of readability, maintainability, and reliability, but more research is needed to determine the extent to which these benefits hold up in practice.

Linking to Studies

Can you link me to these studies?

Chat GPT:

Sure! Here are the links to the studies I mentioned:

1. "A comparison of the maintainability of object-oriented, imperative and functional programs" (2002): https://www.sciencedirect.com/science/article/pii/S0164121201001866
2. "Functional programming experience: A survey of functional and imperative programmers" (2005): https://www.researchgate.net/publication/220654465_Functional_programming_experience_A_survey_of_functional_and_imperative_programmers
3. "Functional programming and the reliability of software systems" (2015): https://www.sciencedirect.com/science/article/pii/S0164121214001442
4. "The State of Haskell in Industry" (2018): https://dl.acm.org/doi/10.1145/3236779.3236782

Commentary

Similar to my experience with asking Chat GPT for monad tutorials, none of these links work unfortunately. So Google is still definitely a better option for finding research and tutorials. But the summary we get from the bot is consistent with what I had thought before. A few groups have tried to study FP vs. OOP, and cautiously concluded some small advantages with FP, but it's difficult to resolve this issue conclusively.

Conclusion

Chat GPT provides a very good summary of the broad points about programming issues. However, when you really dig into the details, things aren't always completely consistent, and it is definitely not the best tool for finding research results just yet.

Next week, we'll take one final run with Chat GPT and see if it can write some Haskell code for us!