# INTRO

[[exercises]]
name = "intro1"
path = "exercises/00_intro/intro1.rs"
mode = "compile"
hint = """
Remove the `I AM NOT DONE` comment in the `exercises/intro00/intro1.rs` file
to move on to the next exercise."""

[[exercises]]
name = "intro2"
path = "exercises/00_intro/intro2.rs"
mode = "compile"
hint = """
The compiler is informing us that we've got the name of the print macro wrong, and has suggested an alternative."""

# VARIABLES

[[exercises]]
name = "variables1"
path = "exercises/01_variables/variables1.rs"
mode = "compile"
hint = """
The declaration in the first line in the main function is missing a keyword
that is needed in Rust to create a new variable binding."""

[[exercises]]
name = "variables2"
path = "exercises/01_variables/variables2.rs"
mode = "compile"
hint = """
The compiler message is saying that Rust cannot infer the type that the
variable binding `x` has with what is given here.

What happens if you annotate the first line in the main function with a type
annotation?

What if you give `x` a value?

What if you do both?

What type should `x` be, anyway?

What if `x` is the same type as `10`? What if it's a different type?"""

[[exercises]]
name = "variables3"
path = "exercises/01_variables/variables3.rs"
mode = "compile"
hint = """
Oops! In this exercise, we have a variable binding that we've created on in the
first line in the `main` function, and we're trying to use it in the next line,
but we haven't given it a value.

We can't print out something that isn't there; try giving `x` a value!

This is an error that can cause bugs that's very easy to make in any
programming language -- thankfully the Rust compiler has caught this for us!"""

[[exercises]]
name = "variables4"
path = "exercises/01_variables/variables4.rs"
mode = "compile"
hint = """
In Rust, variable bindings are immutable by default. But here we're trying
to reassign a different value to `x`! There's a keyword we can use to make
a variable binding mutable instead."""

[[exercises]]
name = "variables5"
path = "exercises/01_variables/variables5.rs"
mode = "compile"
hint = """
In `variables4` we already learned how to make an immutable variable mutable
using a special keyword. Unfortunately this doesn't help us much in this
exercise because we want to assign a different typed value to an existing
variable. Sometimes you may also like to reuse existing variable names because
you are just converting values to different types like in this exercise.

Fortunately Rust has a powerful solution to this problem: 'Shadowing'!
You can read more about 'Shadowing' in the book's section 'Variables and
Mutability':
https://doc.rust-lang.org/book/ch03-01-variables-and-mutability.html#shadowing

Try to solve this exercise afterwards using this technique."""

[[exercises]]
name = "variables6"
path = "exercises/01_variables/variables6.rs"
mode = "compile"
hint = """
We know about variables and mutability, but there is another important type of
variable available: constants.

Constants are always immutable and they are declared with keyword `const` rather
than keyword `let`.

Constants types must also always be annotated.

Read more about constants and the differences between variables and constants
under 'Constants' in the book's section 'Variables and Mutability':
https://doc.rust-lang.org/book/ch03-01-variables-and-mutability.html#constants
"""

# FUNCTIONS

[[exercises]]
name = "functions1"
path = "exercises/02_functions/functions1.rs"
mode = "compile"
hint = """
This main function is calling a function that it expects to exist, but the
function doesn't exist. It expects this function to have the name `call_me`.
It expects this function to not take any arguments and not return a value.
Sounds a lot like `main`, doesn't it?"""

[[exercises]]
name = "functions2"
path = "exercises/02_functions/functions2.rs"
mode = "compile"
hint = """
Rust requires that all parts of a function's signature have type annotations,
but `call_me` is missing the type annotation of `num`."""

[[exercises]]
name = "functions3"
path = "exercises/02_functions/functions3.rs"
mode = "compile"
hint = """
This time, the function *declaration* is okay, but there's something wrong
with the place where we're calling the function.

As a reminder, you can freely play around with different solutions in Rustlings!
Watch mode will only jump to the next exercise if you remove the `I AM NOT
DONE` comment."""

[[exercises]]
name = "functions4"
path = "exercises/02_functions/functions4.rs"
mode = "compile"
hint = """
The error message points to the function `sale_price` and says it expects a type
after the `->`. This is where the function's return type should be -- take a
look at the `is_even` function for an example!"""

[[exercises]]
name = "functions5"
path = "exercises/02_functions/functions5.rs"
mode = "compile"
hint = """
This is a really common error that can be fixed by removing one character.
It happens because Rust distinguishes between expressions and statements:
expressions return a value based on their operand(s), and statements simply
return a `()` type which behaves just like `void` in C/C++ language.

We want to return a value of `i32` type from the `square` function, but it is
returning a `()` type...

They are not the same. There are two solutions:
1. Add a `return` ahead of `num * num;`
2. remove `;`, make it to be `num * num`"""

# IF

[[exercises]]
name = "if1"
path = "exercises/03_if/if1.rs"
mode = "test"
hint = """
It's possible to do this in one line if you would like!

Some similar examples from other languages:
- In C(++) this would be: `a > b ? a : b`
- In Python this would be:  `a if a > b else b`

Remember in Rust that:
- the `if` condition does not need to be surrounded by parentheses
- `if`/`else` conditionals are expressions
- Each condition is followed by a `{}` block."""

[[exercises]]
name = "if2"
path = "exercises/03_if/if2.rs"
mode = "test"
hint = """
For that first compiler error, it's important in Rust that each conditional
block returns the same type! To get the tests passing, you will need a couple
conditions checking different input values."""

[[exercises]]
name = "if3"
path = "exercises/03_if/if3.rs"
mode = "test"
hint = """
In Rust, every arm of an `if` expression has to return the same type of value.
Make sure the type is consistent across all arms."""

# QUIZ 1

[[exercises]]
name = "quiz1"
path = "exercises/quiz1.rs"
mode = "test"
hint = "No hints this time ;)"

# PRIMITIVE TYPES

[[exercises]]
name = "primitive_types1"
path = "exercises/04_primitive_types/primitive_types1.rs"
mode = "compile"
hint = "No hints this time ;)"

[[exercises]]
name = "primitive_types2"
path = "exercises/04_primitive_types/primitive_types2.rs"
mode = "compile"
hint = "No hints this time ;)"

[[exercises]]
name = "primitive_types3"
path = "exercises/04_primitive_types/primitive_types3.rs"
mode = "compile"
hint = """
There's a shorthand to initialize Arrays with a certain size that does not
require you to type in 100 items (but you certainly can if you want!).

For example, you can do:
```
let array = ["Are we there yet?"; 10];
```

Bonus: what are some other things you could have that would return `true`
for `a.len() >= 100`?"""

[[exercises]]
name = "primitive_types4"
path = "exercises/04_primitive_types/primitive_types4.rs"
mode = "test"
hint = """
Take a look at the 'Understanding Ownership -> Slices -> Other Slices' section
of the book: https://doc.rust-lang.org/book/ch04-03-slices.html and use the
starting and ending (plus one) indices of the items in the `Array` that you
want to end up in the slice.

If you're curious why the first argument of `assert_eq!` does not have an
ampersand for a reference since the second argument is a reference, take a look
at the coercion chapter of the nomicon:
https://doc.rust-lang.org/nomicon/coercions.html"""

[[exercises]]
name = "primitive_types5"
path = "exercises/04_primitive_types/primitive_types5.rs"
mode = "compile"
hint = """
Take a look at the 'Data Types -> The Tuple Type' section of the book:
https://doc.rust-lang.org/book/ch03-02-data-types.html#the-tuple-type
Particularly the part about destructuring (second to last example in the
section).

You'll need to make a pattern to bind `name` and `age` to the appropriate parts
of the tuple. You can do it!!"""

[[exercises]]
name = "primitive_types6"
path = "exercises/04_primitive_types/primitive_types6.rs"
mode = "test"
hint = """
While you could use a destructuring `let` for the tuple here, try
indexing into it instead, as explained in the last example of the
'Data Types -> The Tuple Type' section of the book:
https://doc.rust-lang.org/book/ch03-02-data-types.html#the-tuple-type
Now you have another tool in your toolbox!"""

# VECS

[[exercises]]
name = "vecs1"
path = "exercises/05_vecs/vecs1.rs"
mode = "test"
hint = """
In Rust, there are two ways to define a Vector.
1. One way is to use the `Vec::new()` function to create a new vector
   and fill it with the `push()` method.
2. The second way, which is simpler is to use the `vec![]` macro and
   define your elements inside the square brackets.

Check this chapter: https://doc.rust-lang.org/stable/book/ch08-01-vectors.html
of the Rust book to learn more.
"""

[[exercises]]
name = "vecs2"
path = "exercises/05_vecs/vecs2.rs"
mode = "test"
hint = """
In the first function we are looping over the Vector and getting a reference to
one `element` at a time.

To modify the value of that `element` we need to use the `*` dereference
operator. You can learn more in this chapter of the Rust book:
https://doc.rust-lang.org/stable/book/ch08-01-vectors.html#iterating-over-the-values-in-a-vector

In the second function this dereferencing is not necessary, because the `map`
function expects the new value to be returned.

After you've completed both functions, decide for yourself which approach you
like better.

What do you think is the more commonly used pattern under Rust developers?
"""

# MOVE SEMANTICS

[[exercises]]
name = "move_semantics1"
path = "exercises/06_move_semantics/move_semantics1.rs"
mode = "test"
hint = """
So you've got the "cannot borrow immutable local variable `vec` as mutable"
error on the line where we push an element to the vector, right?

The fix for this is going to be adding one keyword, and the addition is NOT on
the line where we push to the vector (where the error is).

Also: Try accessing `vec0` after having called `fill_vec()`. See what
happens!"""

[[exercises]]
name = "move_semantics2"
path = "exercises/06_move_semantics/move_semantics2.rs"
mode = "test"
hint = """
When running this exercise for the first time, you'll notice an error about
"borrow of moved value". In Rust, when an argument is passed to a function and
it's not explicitly returned, you can't use the original variable anymore.
We call this "moving" a variable. When we pass `vec0` into `fill_vec`, it's
being "moved" into `vec1`, meaning we can't access `vec0` anymore after the
fact.

Rust provides a couple of different ways to mitigate this issue, feel free to
try them all:
1. You could make another, separate version of the data that's in `vec0` and
   pass that to `fill_vec` instead.
2. Make `fill_vec` borrow its argument instead of taking ownership of it,
   and then copy the data within the function (`vec.clone()`) in order to
   return an owned `Vec<i32>`.
"""

[[exercises]]
name = "move_semantics3"
path = "exercises/06_move_semantics/move_semantics3.rs"
mode = "test"
hint = """
The difference between this one and the previous ones is that the first line
of `fn fill_vec` that had `let mut vec = vec;` is no longer there. You can,
instead of adding that line back, add `mut` in one place that will change
an existing binding to be a mutable binding instead of an immutable one :)"""

[[exercises]]
name = "move_semantics4"
path = "exercises/06_move_semantics/move_semantics4.rs"
mode = "test"
hint = """
Stop reading whenever you feel like you have enough direction :) Or try
doing one step and then fixing the compiler errors that result!
So the end goal is to:
   - get rid of the first line in main that creates the new vector
   - so then `vec0` doesn't exist, so we can't pass it to `fill_vec`
   - `fill_vec` has had its signature changed, which our call should reflect
   - since we're not creating a new vec in `main` anymore, we need to create
     a new vec in `fill_vec`, and fill it with the expected values"""

[[exercises]]
name = "move_semantics5"
path = "exercises/06_move_semantics/move_semantics5.rs"
mode = "test"
hint = """
Carefully reason about the range in which each mutable reference is in
scope. Does it help to update the value of `x` immediately after
the mutable reference is taken? Read more about 'Mutable References'
in the book's section 'References and Borrowing':
https://doc.rust-lang.org/book/ch04-02-references-and-borrowing.html#mutable-references.
"""

[[exercises]]
name = "move_semantics6"
path = "exercises/06_move_semantics/move_semantics6.rs"
mode = "compile"
hint = """
To find the answer, you can consult the book section "References and Borrowing":
https://doc.rust-lang.org/stable/book/ch04-02-references-and-borrowing.html

The first problem is that `get_char` is taking ownership of the string. So
`data` is moved and can't be used for `string_uppercase`. `data` is moved to
`get_char` first, meaning that `string_uppercase` cannot manipulate the data.

Once you've fixed that, `string_uppercase`'s function signature will also need
to be adjusted.

Can you figure out how?

Another hint: it has to do with the `&` character."""

# STRUCTS

[[exercises]]
name = "structs1"
path = "exercises/07_structs/structs1.rs"
mode = "test"
hint = """
Rust has more than one type of struct. Three actually, all variants are used to
package related data together.

There are normal (or classic) structs. These are named collections of related
data stored in fields.

Tuple structs are basically just named tuples.

Finally, Unit-like structs. These don't have any fields and are useful for
generics.

In this exercise you need to complete and implement one of each kind.
Read more about structs in The Book:
https://doc.rust-lang.org/book/ch05-01-defining-structs.html"""

[[exercises]]
name = "structs2"
path = "exercises/07_structs/structs2.rs"
mode = "test"
hint = """
Creating instances of structs is easy, all you need to do is assign some values
to its fields.

There are however some shortcuts that can be taken when instantiating structs.
Have a look in The Book, to find out more:
https://doc.rust-lang.org/stable/book/ch05-01-defining-structs.html#creating-instances-from-other-instances-with-struct-update-syntax"""

[[exercises]]
name = "structs3"
path = "exercises/07_structs/structs3.rs"
mode = "test"
hint = """
For `is_international`: What makes a package international? Seems related to
the places it goes through right?

For `get_fees`: This method takes an additional argument, is there a field in
the `Package` struct that this relates to?

Have a look in The Book, to find out more about method implementations:
https://doc.rust-lang.org/book/ch05-03-method-syntax.html"""

# ENUMS

[[exercises]]
name = "enums1"
path = "exercises/08_enums/enums1.rs"
mode = "compile"
hint = "No hints this time ;)"

[[exercises]]
name = "enums2"
path = "exercises/08_enums/enums2.rs"
mode = "compile"
hint = """
You can create enumerations that have different variants with different types
such as no data, anonymous structs, a single string, tuples, ...etc"""

[[exercises]]
name = "enums3"
path = "exercises/08_enums/enums3.rs"
mode = "test"
hint = """
As a first step, you can define enums to compile this code without errors.

And then create a match expression in `process()`.

Note that you need to deconstruct some message variants in the match expression
to get value in the variant."""

# STRINGS

[[exercises]]
name = "strings1"
path = "exercises/09_strings/strings1.rs"
mode = "compile"
hint = """
The `current_favorite_color` function is currently returning a string slice
with the `'static` lifetime. We know this because the data of the string lives
in our code itself -- it doesn't come from a file or user input or another
program -- so it will live as long as our program lives.

But it is still a string slice. There's one way to create a `String` by
converting a string slice covered in the Strings chapter of the book, and
another way that uses the `From` trait."""

[[exercises]]
name = "strings2"
path = "exercises/09_strings/strings2.rs"
mode = "compile"
hint = """
Yes, it would be really easy to fix this by just changing the value bound to
`word` to be a string slice instead of a `String`, wouldn't it?? There is a way
to add one character to the `if` statement, though, that will coerce the
`String` into a string slice.

Side note: If you're interested in learning about how this kind of reference
conversion works, you can jump ahead in the book and read this part in the
smart pointers chapter:
https://doc.rust-lang.org/stable/book/ch15-02-deref.html#implicit-deref-coercions-with-functions-and-methods"""

[[exercises]]
name = "strings3"
path = "exercises/09_strings/strings3.rs"
mode = "test"
hint = """
There's tons of useful standard library functions for strings. Let's try and use some of them:
https://doc.rust-lang.org/std/string/struct.String.html#method.trim

For the `compose_me` method: You can either use the `format!` macro, or convert
the string slice into an owned string, which you can then freely extend."""

[[exercises]]
name = "strings4"
path = "exercises/09_strings/strings4.rs"
mode = "compile"
hint = "No hints this time ;)"

# MODULES

[[exercises]]
name = "modules1"
path = "exercises/10_modules/modules1.rs"
mode = "compile"
hint = """
Everything is private in Rust by default-- but there's a keyword we can use
to make something public! The compiler error should point to the thing that
needs to be public."""

[[exercises]]
name = "modules2"
path = "exercises/10_modules/modules2.rs"
mode = "compile"
hint = """
The delicious_snacks module is trying to present an external interface that is
different than its internal structure (the `fruits` and `veggies` modules and
associated constants). Complete the `use` statements to fit the uses in main and
find the one keyword missing for both constants.

Learn more at https://doc.rust-lang.org/book/ch07-04-bringing-paths-into-scope-with-the-use-keyword.html#re-exporting-names-with-pub-use"""

[[exercises]]
name = "modules3"
path = "exercises/10_modules/modules3.rs"
mode = "compile"
hint = """
`UNIX_EPOCH` and `SystemTime` are declared in the `std::time` module. Add a
`use` statement for these two to bring them into scope. You can use nested
paths or the glob operator to bring these two in using only one line."""

# HASHMAPS

[[exercises]]
name = "hashmaps1"
path = "exercises/11_hashmaps/hashmaps1.rs"
mode = "test"
hint = """
Hint 1: Take a look at the return type of the function to figure out
        the type for the `basket`.

Hint 2: Number of fruits should be at least 5. And you have to put
        at least three different types of fruits.
"""

[[exercises]]
name = "hashmaps2"
path = "exercises/11_hashmaps/hashmaps2.rs"
mode = "test"
hint = """
Use the `entry()` and `or_insert()` methods of `HashMap` to achieve this.
Learn more at https://doc.rust-lang.org/stable/book/ch08-03-hash-maps.html#only-inserting-a-value-if-the-key-has-no-value
"""

[[exercises]]
name = "hashmaps3"
path = "exercises/11_hashmaps/hashmaps3.rs"
mode = "test"
hint = """
Hint 1: Use the `entry()` and `or_insert()` methods of `HashMap` to insert
        entries corresponding to each team in the scores table.

Learn more at https://doc.rust-lang.org/stable/book/ch08-03-hash-maps.html#only-inserting-a-value-if-the-key-has-no-value

Hint 2: If there is already an entry for a given key, the value returned by
        `entry()` can be updated based on the existing value.

Learn more at https://doc.rust-lang.org/book/ch08-03-hash-maps.html#updating-a-value-based-on-the-old-value
"""

# QUIZ 2

[[exercises]]
name = "quiz2"
path = "exercises/quiz2.rs"
mode = "test"
hint = "No hints this time ;)"

# OPTIONS

[[exercises]]
name = "options1"
path = "exercises/12_options/options1.rs"
mode = "test"
hint = """
Options can have a `Some` value, with an inner value, or a `None` value,
without an inner value.

There's multiple ways to get at the inner value, you can use `unwrap`, or
pattern match. Unwrapping is the easiest, but how do you do it safely so that
it doesn't panic in your face later?"""

[[exercises]]
name = "options2"
path = "exercises/12_options/options2.rs"
mode = "test"
hint = """
Check out:

- https://doc.rust-lang.org/rust-by-example/flow_control/if_let.html
- https://doc.rust-lang.org/rust-by-example/flow_control/while_let.html

Remember that `Option`s can be stacked in `if let` and `while let`.

For example: `Some(Some(variable)) = variable2`

Also see `Option::flatten`
"""

[[exercises]]
name = "options3"
path = "exercises/12_options/options3.rs"
mode = "compile"
hint = """
The compiler says a partial move happened in the `match` statement. How can
this be avoided? The compiler shows the correction needed.

After making the correction as suggested by the compiler, do read:
https://doc.rust-lang.org/std/keyword.ref.html"""

# ERROR HANDLING

[[exercises]]
name = "errors1"
path = "exercises/13_error_handling/errors1.rs"
mode = "test"
hint = """
`Ok` and `Err` are the two variants of `Result`, so what the tests are saying
is that `generate_nametag_text` should return a `Result` instead of an `Option`.

To make this change, you'll need to:
  - update the return type in the function signature to be a `Result<String,
    String>` that could be the variants `Ok(String)` and `Err(String)`
  - change the body of the function to return `Ok(stuff)` where it currently
    returns `Some(stuff)`
  - change the body of the function to return `Err(error message)` where it
    currently returns `None`"""

[[exercises]]
name = "errors2"
path = "exercises/13_error_handling/errors2.rs"
mode = "test"
hint = """
One way to handle this is using a `match` statement on
`item_quantity.parse::<i32>()` where the cases are `Ok(something)` and
`Err(something)`.

This pattern is very common in Rust, though, so there's a `?` operator that
does pretty much what you would make that match statement do for you!

Take a look at this section of the 'Error Handling' chapter:
https://doc.rust-lang.org/book/ch09-02-recoverable-errors-with-result.html#a-shortcut-for-propagating-errors-the--operator
and give it a try!"""

[[exercises]]
name = "errors3"
path = "exercises/13_error_handling/errors3.rs"
mode = "compile"
hint = """
If other functions can return a `Result`, why shouldn't `main`? It's a fairly
common convention to return something like `Result<(), ErrorType>` from your
main function.

The unit (`()`) type is there because nothing is really needed in terms of
positive results."""

[[exercises]]
name = "errors4"
path = "exercises/13_error_handling/errors4.rs"
mode = "test"
hint = """
`PositiveNonzeroInteger::new` is always creating a new instance and returning
an `Ok` result.

It should be doing some checking, returning an `Err` result if those checks
fail, and only returning an `Ok` result if those checks determine that
everything is... okay :)"""

[[exercises]]
name = "errors5"
path = "exercises/13_error_handling/errors5.rs"
mode = "compile"
hint = """
There are two different possible `Result` types produced within `main()`, which
are propagated using `?` operators. How do we declare a return type from
`main()` that allows both?

Under the hood, the `?` operator calls `From::from` on the error value to
convert it to a boxed trait object, a `Box<dyn error::Error>`. This boxed trait
object is polymorphic, and since all errors implement the `error::Error` trait,
we can capture lots of different errors in one "Box" object.

Check out this section of the book:
https://doc.rust-lang.org/book/ch09-02-recoverable-errors-with-result.html#a-shortcut-for-propagating-errors-the--operator

Read more about boxing errors:
https://doc.rust-lang.org/stable/rust-by-example/error/multiple_error_types/boxing_errors.html

Read more about using the `?` operator with boxed errors:
https://doc.rust-lang.org/stable/rust-by-example/error/multiple_error_types/reenter_question_mark.html
"""

[[exercises]]
name = "errors6"
path = "exercises/13_error_handling/errors6.rs"
mode = "test"
hint = """
This exercise uses a completed version of `PositiveNonzeroInteger` from
errors4.

Below the line that `TODO` asks you to change, there is an example of using
the `map_err()` method on a `Result` to transform one type of error into
another. Try using something similar on the `Result` from `parse()`. You
might use the `?` operator to return early from the function, or you might
use a `match` expression, or maybe there's another way!

You can create another function inside `impl ParsePosNonzeroError` to use
with `map_err()`.

Read more about `map_err()` in the `std::result` documentation:
https://doc.rust-lang.org/std/result/enum.Result.html#method.map_err"""

# Generics

[[exercises]]
name = "generics1"
path = "exercises/14_generics/generics1.rs"
mode = "compile"
hint = """
Vectors in Rust make use of generics to create dynamically sized arrays of any
type.

You need to tell the compiler what type we are pushing onto this vector."""

[[exercises]]
name = "generics2"
path = "exercises/14_generics/generics2.rs"
mode = "test"
hint = """
Currently we are wrapping only values of type `u32`.

Maybe we could update the explicit references to this data type somehow?

If you are still stuck https://doc.rust-lang.org/stable/book/ch10-01-syntax.html#in-method-definitions
"""

# TRAITS

[[exercises]]
name = "traits1"
path = "exercises/15_traits/traits1.rs"
mode = "test"
hint = """
A discussion about Traits in Rust can be found at:
https://doc.rust-lang.org/book/ch10-02-traits.html
"""

[[exercises]]
name = "traits2"
path = "exercises/15_traits/traits2.rs"
mode = "test"
hint = """
Notice how the trait takes ownership of `self`, and returns `Self`.

Try mutating the incoming string vector. Have a look at the tests to see
what the result should look like!

Vectors provide suitable methods for adding an element at the end. See
the documentation at: https://doc.rust-lang.org/std/vec/struct.Vec.html"""

[[exercises]]
name = "traits3"
path = "exercises/15_traits/traits3.rs"
mode = "test"
hint = """
Traits can have a default implementation for functions. Structs that implement
the trait can then use the default version of these functions if they choose not
to implement the function themselves.

See the documentation at: https://doc.rust-lang.org/book/ch10-02-traits.html#default-implementations
"""

[[exercises]]
name = "traits4"
path = "exercises/15_traits/traits4.rs"
mode = "test"
hint = """
Instead of using concrete types as parameters you can use traits. Try replacing
the '??' with 'impl [what goes here?]'

See the documentation at: https://doc.rust-lang.org/book/ch10-02-traits.html#traits-as-parameters
"""

[[exercises]]
name = "traits5"
path = "exercises/15_traits/traits5.rs"
mode = "compile"
hint = """
To ensure a parameter implements multiple traits use the '+ syntax'. Try
replacing the '??' with 'impl [what goes here?] + [what goes here?]'.

See the documentation at: https://doc.rust-lang.org/book/ch10-02-traits.html#specifying-multiple-trait-bounds-with-the--syntax
"""

# QUIZ 3

[[exercises]]
name = "quiz3"
path = "exercises/quiz3.rs"
mode = "test"
hint = """
To find the best solution to this challenge you're going to need to think back
to your knowledge of traits, specifically 'Trait Bound Syntax'

You may also need this: `use std::fmt::Display;`."""

# LIFETIMES

[[exercises]]
name = "lifetimes1"
path = "exercises/16_lifetimes/lifetimes1.rs"
mode = "compile"
hint = """
Let the compiler guide you. Also take a look at the book if you need help:
https://doc.rust-lang.org/book/ch10-03-lifetime-syntax.html"""

[[exercises]]
name = "lifetimes2"
path = "exercises/16_lifetimes/lifetimes2.rs"
mode = "compile"
hint = """
Remember that the generic lifetime `'a` will get the concrete lifetime that is
equal to the smaller of the lifetimes of `x` and `y`.

You can take at least two paths to achieve the desired result while keeping the
inner block:
1. Move the `string2` declaration to make it live as long as `string1` (how is
   `result` declared?)
2. Move `println!` into the inner block"""

[[exercises]]
name = "lifetimes3"
path = "exercises/16_lifetimes/lifetimes3.rs"
mode = "compile"
hint = """
If you use a lifetime annotation in a struct's fields, where else does it need
to be added?"""

# TESTS

[[exercises]]
name = "tests1"
path = "exercises/17_tests/tests1.rs"
mode = "test"
hint = """
You don't even need to write any code to test -- you can just test values and
run that, even though you wouldn't do that in real life. :)

`assert!` is a macro that needs an argument. Depending on the value of the
argument, `assert!` will do nothing (in which case the test will pass) or
`assert!` will panic (in which case the test will fail).

So try giving different values to `assert!` and see which ones compile, which
ones pass, and which ones fail :)"""

[[exercises]]
name = "tests2"
path = "exercises/17_tests/tests2.rs"
mode = "test"
hint = """
Like the previous exercise, you don't need to write any code to get this test
to compile and run.

`assert_eq!` is a macro that takes two arguments and compares them. Try giving
it two values that are equal! Try giving it two arguments that are different!
Try giving it two values that are of different types! Try switching which
argument comes first and which comes second!"""

[[exercises]]
name = "tests3"
path = "exercises/17_tests/tests3.rs"
mode = "test"
hint = """
You can call a function right where you're passing arguments to `assert!`. So
you could do something like `assert!(having_fun())`.

If you want to check that you indeed get `false`, you can negate the result of
what you're doing using `!`, like `assert!(!having_fun())`."""

[[exercises]]
name = "tests4"
path = "exercises/17_tests/tests4.rs"
mode = "test"
hint = """
We expect method `Rectangle::new()` to panic for negative values.

To handle that you need to add a special attribute to the test function.

You can refer to the docs:
https://doc.rust-lang.org/stable/book/ch11-01-writing-tests.html#checking-for-panics-with-should_panic"""

# STANDARD LIBRARY TYPES

[[exercises]]
name = "iterators1"
path = "exercises/18_iterators/iterators1.rs"
mode = "test"
hint = """
Step 1:

We need to apply something to the collection `my_fav_fruits` before we start to
go through it. What could that be? Take a look at the struct definition for a
vector for inspiration:
https://doc.rust-lang.org/std/vec/struct.Vec.html

Step 2 & step 3:

Very similar to the lines above and below. You've got this!

Step 4:

An iterator goes through all elements in a collection, but what if we've run
out of elements? What should we expect here? If you're stuck, take a look at
https://doc.rust-lang.org/std/iter/trait.Iterator.html for some ideas.
"""

[[exercises]]
name = "iterators2"
path = "exercises/18_iterators/iterators2.rs"
mode = "test"
hint = """
Step 1:

The variable `first` is a `char`. It needs to be capitalized and added to the
remaining characters in `c` in order to return the correct `String`.

The remaining characters in `c` can be viewed as a string slice using the
`as_str` method.

The documentation for `char` contains many useful methods.
https://doc.rust-lang.org/std/primitive.char.html

Step 2:

Create an iterator from the slice. Transform the iterated values by applying
the `capitalize_first` function. Remember to `collect` the iterator.

Step 3:

This is surprisingly similar to the previous solution. `collect` is very
powerful and very general. Rust just needs to know the desired type."""

[[exercises]]
name = "iterators3"
path = "exercises/18_iterators/iterators3.rs"
mode = "test"
hint = """
The `divide` function needs to return the correct error when even division is
not possible.

The `division_results` variable needs to be collected into a collection type.

The `result_with_list` function needs to return a single `Result` where the
success case is a vector of integers and the failure case is a `DivisionError`.

The `list_of_results` function needs to return a vector of results.

See https://doc.rust-lang.org/std/iter/trait.Iterator.html#method.collect for
how the `FromIterator` trait is used in `collect()`. This trait is REALLY
powerful! It can make the solution to this exercise infinitely easier."""

[[exercises]]
name = "iterators4"
path = "exercises/18_iterators/iterators4.rs"
mode = "test"
hint = """
In an imperative language, you might write a `for` loop that updates a mutable
variable. Or, you might write code utilizing recursion and a match clause. In
Rust you can take another functional approach, computing the factorial
elegantly with ranges and iterators.

Hint 2: Check out the `fold` and `rfold` methods!"""

[[exercises]]
name = "iterators5"
path = "exercises/18_iterators/iterators5.rs"
mode = "test"
hint = """
The documentation for the `std::iter::Iterator` trait contains numerous methods
that would be helpful here.

The `collection` variable in `count_collection_iterator` is a slice of
`HashMap`s. It needs to be converted into an iterator in order to use the
iterator methods.

The `fold` method can be useful in the `count_collection_iterator` function.

For a further challenge, consult the documentation for `Iterator` to find
a different method that could make your code more compact than using `fold`."""

# SMART POINTERS

[[exercises]]
name = "box1"
path = "exercises/19_smart_pointers/box1.rs"
mode = "test"
hint = """
Step 1:

The compiler's message should help: since we cannot store the value of the
actual type when working with recursive types, we need to store a reference
(pointer) to its value.

We should, therefore, place our `List` inside a `Box`. More details in the book
here: https://doc.rust-lang.org/book/ch15-01-box.html#enabling-recursive-types-with-boxes

Step 2:

Creating an empty list should be fairly straightforward (hint: peek at the
assertions).

For a non-empty list keep in mind that we want to use our `Cons` "list builder".
Although the current list is one of integers (`i32`), feel free to change the
definition and try other types!
"""

[[exercises]]
name = "rc1"
path = "exercises/19_smart_pointers/rc1.rs"
mode = "test"
hint = """
This is a straightforward exercise to use the `Rc<T>` type. Each `Planet` has
ownership of the `Sun`, and uses `Rc::clone()` to increment the reference count
of the `Sun`.

After using `drop()` to move the `Planet`s out of scope individually, the
reference count goes down.

In the end the `Sun` only has one reference again, to itself.

See more at: https://doc.rust-lang.org/book/ch15-04-rc.html

* Unfortunately Pluto is no longer considered a planet :(
"""

[[exercises]]
name = "arc1"
path = "exercises/19_smart_pointers/arc1.rs"
mode = "compile"
hint = """
Make `shared_numbers` be an `Arc` from the numbers vector. Then, in order
to avoid creating a copy of `numbers`, you'll need to create `child_numbers`
inside the loop but still in the main thread.

`child_numbers` should be a clone of the `Arc` of the numbers instead of a
thread-local copy of the numbers.

This is a simple exercise if you understand the underlying concepts, but if this
is too much of a struggle, consider reading through all of Chapter 16 in the
book:
https://doc.rust-lang.org/stable/book/ch16-00-concurrency.html
"""

[[exercises]]
name = "cow1"
path = "exercises/19_smart_pointers/cow1.rs"
mode = "test"
hint = """
If `Cow` already owns the data it doesn't need to clone it when `to_mut()` is
called.

Check out https://doc.rust-lang.org/std/borrow/enum.Cow.html for documentation
on the `Cow` type.
"""

# THREADS

[[exercises]]
name = "threads1"
path = "exercises/20_threads/threads1.rs"
mode = "compile"
hint = """
`JoinHandle` is a struct that is returned from a spawned thread:
https://doc.rust-lang.org/std/thread/fn.spawn.html

A challenge with multi-threaded applications is that the main thread can
finish before the spawned threads are completed.
https://doc.rust-lang.org/book/ch16-01-threads.html#waiting-for-all-threads-to-finish-using-join-handles

Use the `JoinHandle`s to wait for each thread to finish and collect their
results.

https://doc.rust-lang.org/std/thread/struct.JoinHandle.html
"""

[[exercises]]
name = "threads2"
path = "exercises/20_threads/threads2.rs"
mode = "compile"
hint = """
`Arc` is an Atomic Reference Counted pointer that allows safe, shared access
to **immutable** data. But we want to *change* the number of `jobs_completed`
so we'll need to also use another type that will only allow one thread to
mutate the data at a time. Take a look at this section of the book:
https://doc.rust-lang.org/book/ch16-03-shared-state.html#atomic-reference-counting-with-arct

Keep reading if you'd like more hints :)

Do you now have an `Arc<Mutex<JobStatus>>` at the beginning of `main`? Like:
```
let status = Arc::new(Mutex::new(JobStatus { jobs_completed: 0 }));
```

Similar to the code in the following example in the book:
https://doc.rust-lang.org/book/ch16-03-shared-state.html#sharing-a-mutext-between-multiple-threads
"""

[[exercises]]
name = "threads3"
path = "exercises/20_threads/threads3.rs"
mode = "test"
hint = """
An alternate way to handle concurrency between threads is to use an `mpsc`
(multiple producer, single consumer) channel to communicate.

With both a sending end and a receiving end, it's possible to send values in
one thread and receive them in another.

Multiple producers are possible by using clone() to create a duplicate of the
original sending end.

See https://doc.rust-lang.org/book/ch16-02-message-passing.html for more info.
"""

# MACROS

[[exercises]]
name = "macros1"
path = "exercises/21_macros/macros1.rs"
mode = "compile"
hint = """
When you call a macro, you need to add something special compared to a
regular function call. If you're stuck, take a look at what's inside
`my_macro`."""

[[exercises]]
name = "macros2"
path = "exercises/21_macros/macros2.rs"
mode = "compile"
hint = """
Macros don't quite play by the same rules as the rest of Rust, in terms of
what's available where.

Unlike other things in Rust, the order of "where you define a macro" versus
"where you use it" actually matters."""

[[exercises]]
name = "macros3"
path = "exercises/21_macros/macros3.rs"
mode = "compile"
hint = """
In order to use a macro outside of its module, you need to do something
special to the module to lift the macro out into its parent.

The same trick also works on "extern crate" statements for crates that have
exported macros, if you've seen any of those around."""

[[exercises]]
name = "macros4"
path = "exercises/21_macros/macros4.rs"
mode = "compile"
hint = """
You only need to add a single character to make this compile.

The way macros are written, it wants to see something between each "macro arm",
so it can separate them.

That's all the macro exercises we have in here, but it's barely even scratching
the surface of what you can do with Rust's macros. For a more thorough
introduction, you can have a read through 'The Little Book of Rust Macros':
https://veykril.github.io/tlborm/"""

#  CLIPPY

[[exercises]]
name = "clippy1"
path = "exercises/22_clippy/clippy1.rs"
mode = "clippy"
hint = """
Rust stores the highest precision version of any long or infinite precision
mathematical constants in the Rust standard library:
https://doc.rust-lang.org/stable/std/f32/consts/index.html

We may be tempted to use our own approximations for certain mathematical
constants, but clippy recognizes those imprecise mathematical constants as a
source of potential error.

See the suggestions of the clippy warning in compile output and use the
appropriate replacement constant from `std::f32::consts`..."""

[[exercises]]
name = "clippy2"
path = "exercises/22_clippy/clippy2.rs"
mode = "clippy"
hint = """
`for` loops over `Option` values are more clearly expressed as an `if let`"""

[[exercises]]
name = "clippy3"
path = "exercises/22_clippy/clippy3.rs"
mode = "clippy"
hint = "No hints this time!"

# TYPE CONVERSIONS

[[exercises]]
name = "using_as"
path = "exercises/23_conversions/using_as.rs"
mode = "test"
hint = """
Use the `as` operator to cast one of the operands in the last line of the
`average` function into the expected return type."""

[[exercises]]
name = "from_into"
path = "exercises/23_conversions/from_into.rs"
mode = "test"
hint = """
Follow the steps provided right before the `From` implementation"""

[[exercises]]
name = "from_str"
path = "exercises/23_conversions/from_str.rs"
mode = "test"
hint = """
The implementation of `FromStr` should return an `Ok` with a `Person` object,
or an `Err` with an error if the string is not valid.

This is almost like the `from_into` exercise, but returning errors instead
of falling back to a default value.

Look at the test cases to see which error variants to return.

Another hint: You can use the `map_err` method of `Result` with a function
or a closure to wrap the error from `parse::<usize>`.

Yet another hint: If you would like to propagate errors by using the `?`
operator in your solution, you might want to look at
https://doc.rust-lang.org/stable/rust-by-example/error/multiple_error_types/reenter_question_mark.html
"""

[[exercises]]
name = "try_from_into"
path = "exercises/23_conversions/try_from_into.rs"
mode = "test"
hint = """
Follow the steps provided right before the `TryFrom` implementation.
You can also use the example at
https://doc.rust-lang.org/std/convert/trait.TryFrom.html

Is there an implementation of `TryFrom` in the standard library that
can both do the required integer conversion and check the range of the input?

Another hint: Look at the test cases to see which error variants to return.

Yet another hint: You can use the `map_err` or `or` methods of `Result` to
convert errors.

Yet another hint: If you would like to propagate errors by using the `?`
operator in your solution, you might want to look at
https://doc.rust-lang.org/stable/rust-by-example/error/multiple_error_types/reenter_question_mark.html

Challenge: Can you make the `TryFrom` implementations generic over many integer types?"""

[[exercises]]
name = "as_ref_mut"
path = "exercises/23_conversions/as_ref_mut.rs"
mode = "test"
hint = """
Add `AsRef<str>` or `AsMut<u32>` as a trait bound to the functions."""