Example case:

let person_1 = {
  name = "John";
  ageGroup = #minor";
};

let minor_permissions = { ... };
let adult_permissions = { ... };

let person_1_plus_permissions = person_1 and (
  switch(person1.ageGroup) {
    case (#minor) { minorPermissions };
    case (#adult) { adultPermissions };
  },  // prettier puts a `;` here instead of a `,`, which throws a syntax error
);

Great work with the plugin 👍

I gave it a try in Papyrs backend / providers (PR 21) and landed on following issue at build time after having formatted the code:

/Users/daviddalbusco/.../storage.store.mo:96.5-96.6: syntax error [M0001], unexpected token '>', expected one of token or sequence:

The root cause of the issue is the following type of function:

private func update(key: Text, record: Data, putData: PutData): Result.Result<Data, Text> {

Gets formatted to:

private func checkTimestamp(record : Data, id : Text, updated_at : Time.Time) : Result.Result<
  Text,
  Text
> {

The Motoko compiler does not expect the last > to finds place on a new line. Adapting code manually as following fix the build issue:

private func checkTimestamp(record : Data, id : Text, updated_at : Time.Time) : Result.Result<
  Text,
  Text> {

dfx 0.11.1

Before formatting

After formatting

"Full" snippet:

assertAllTrue([
  fundedPlan.size() == 1,
  failures.size() == 1,
  fundedPlan[0].2 == 500_000_000_000,
  fundedPlan[0].3 == { e8s = 50_000_000 },
  failures[0].2 == 1_000_000_000_000,
  failures[0].3 == 500_000_000_000,
]);

According to the style guide, code should be indented with 2 spaces, currently 4 spaces are used.

https://internetcomputer.org/docs/current/developer-docs/build/cdks/motoko-dfinity/style/#indentation

Each level of indentation should be 2 spaces.

Issue

While adding following comment to one of my canister, I noticed that the auto formatting was breaking - was generating an incorrect format for my code.

/**
 * Does a bucket id exists? - i.e. not a bucket for a user but is a canister id linked to any user's bucket? 
 */

After debugging I noticed that the root cause was using a quote ' in the comment. i.e. if I remove the quote, format is generated correctly.

Current behavior:

test(
  "example test",
  func() {
    assert 1 == 1;
  },
);

Expected behavior(?):

test("example test", func() {
  assert 1 == 1;
});

canisters : HashMap.HashMap<
  // The principal of the canister.
  Principal, 
  // The configuration of the canister.
  CanisterConfig
>;

becomes

canisters : HashMap.HashMap</* The principal of the canister. */ Principal, /* The configuration of the canister. */ CanisterConfig>;`

using the experimental LS with dfx 0.11.2 the following happens

IMO it would be nice if it would be formatted into multiple lines if the string exceeds a certain width to maintain readability

https://github.com/flowerpowerdao/punks-nft-canister/pull/32/commits/521759d663068a907fdd95756e9ce823c4b7fe46#diff-303c0b742b59bc66b0e9f6e381809d383386ab3b0ba9cda25a74f1405fc411edL195-L196

This was motoko code generated by Go. I try and make the generator as close to the prettified output as possible.

so then I put a newline after update(, and Ctrl-S

and then finally one after do { on line 81, and Ctrl-S

I've found myself just trying to enter random newlines to make it format correctly, and I wasn't sure this was the desired behaviour.

Enforce the [style guide recommendation](https://internetcomputer.org/docs/current/developer-docs/build/cdks/motoko-dfinity/style/#miscellaneous] for numbers:

Group by 3 digits in decimal numbers and by 4 in hexadecimal notation.

In addition, add leading zeros in hex notation to be 2 digits or a multiple of 4 digits.

Examples:

• 1000000 would be formatted as 1_000_000
• 0x2 would be formatted as 0x02
• 0x123456 would be formatted as 0x0012_3456
• 12_34_56 would be formatted as 123_456
• 0.123456 would be formatted as 0.123_456

After a comment marker (//) at least one space should be inserted to improve readability. Additional spaces should be preserved.

Examples:

• //some comment would be formatted as // some comment

// module {
//   asdf;
//}

would be formatted as

// module {          <- keep one space
//   asdf;           <- keep multiple spaces
// }                 <- add one space