Developing the Library’s Functionality with Test-Driven Development

Now that we’ve extracted the logic into src/lib.rs and left the argument collecting and error handling in src/main.rs, it’s much easier to write tests for the core functionality of our code. We can call functions directly with various arguments and check return values without having to call our binary from the command line.

In this section, we’ll add the searching logic to the minigrep program using the test-driven development (TDD) process with the following steps:

  1. Write a test that fails and run it to make sure it fails for the reason you expect.
  2. Write or modify just enough code to make the new test pass.
  3. Refactor the code you just added or changed and make sure the tests continue to pass.
  4. Repeat from step 1!

Though it’s just one of many ways to write software, TDD can help drive code design. Writing the test before you write the code that makes the test pass helps to maintain high test coverage throughout the process.

We’ll test-drive the implementation of the functionality that will actually do the searching for the query string in the file contents and produce a list of lines that match the query. We’ll add this functionality in a function called search.

Writing a Failing Test

Because we don’t need them anymore, let’s remove the println! statements from src/lib.rs and src/main.rs that we used to check the program’s behavior. Then, in src/lib.rs, we’ll add a tests module with a test function, as we did in Chapter 11. The test function specifies the behavior we want the search function to have: it will take a query and the text to search, and it will return only the lines from the text that contain the query. Listing 12-15 shows this test, which won’t compile yet.

Filename: src/lib.rs
  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &’static str> {
  11.         if args.len() < 3 {
  12.             return Err(“not enough arguments”);
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     Ok(())
  26. }
  28. #[cfg(test)]
  29. mod tests {
  30.     use super::*;
  32.     #[test]
  33.     fn one_result() {
  34.         let query = “duct”;
  35.         let contents = “\
  36. Rust:
  37. safe, fast, productive.
  38. Pick three.”;
  40.         assert_eq!(vec![“safe, fast, productive.”], search(query, contents));
  41.     }
  42. }
Listing 12-15: Creating a failing test for the search function we wish we had

This test searches for the string "duct". The text we’re searching is three lines, only one of which contains "duct" (note that the backslash after the opening double quote tells Rust not to put a newline character at the beginning of the contents of this string literal). We assert that the value returned from the search function contains only the line we expect.

We aren’t yet able to run this test and watch it fail because the test doesn’t even compile: the search function doesn’t exist yet! In accordance with TDD principles, we’ll add just enough code to get the test to compile and run by adding a definition of the search function that always returns an empty vector, as shown in Listing 12-16. Then the test should compile and fail because an empty vector doesn’t match a vector containing the line "safe, fast, productive."

Filename: src/lib.rs
  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &’static str> {
  11.         if args.len() < 3 {
  12.             return Err(“not enough arguments”);
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     Ok(())
  26. }
  28. pub fn search<’a>(query: &str, contents: &’a str) -> Vec<&’a str> {
  29.     vec![]
  30. }
  32. #[cfg(test)]
  33. mod tests {
  34.     use super::*;
  36.     #[test]
  37.     fn one_result() {
  38.         let query = “duct”;
  39.         let contents = “\
  40. Rust:
  41. safe, fast, productive.
  42. Pick three.”;
  44.         assert_eq!(vec![“safe, fast, productive.”], search(query, contents));
  45.     }
  46. }
Listing 12-16: Defining just enough of the search function so our test will compile

Notice that we need to define an explicit lifetime 'a in the signature of search and use that lifetime with the contents argument and the return value. Recall in Chapter 10 that the lifetime parameters specify which argument lifetime is connected to the lifetime of the return value. In this case, we indicate that the returned vector should contain string slices that reference slices of the argument contents (rather than the argument query).

In other words, we tell Rust that the data returned by the search function will live as long as the data passed into the search function in the contents argument. This is important! The data referenced by a slice needs to be valid for the reference to be valid; if the compiler assumes we’re making string slices of query rather than contents, it will do its safety checking incorrectly.

If we forget the lifetime annotations and try to compile this function, we’ll get this error:

  1. $ cargo build
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3. error[E0106]: missing lifetime specifier
  4.   --> src/lib.rs:28:51
  5.    |
  6. 28 | pub fn search(query: &str, contents: &str) -> Vec<&str> {
  7.    |                      ----            ----         ^ expected named lifetime parameter
  8.    |
  9.    = help: this function's return type contains a borrowed value, but the signature does not say whether it is borrowed from `query` or `contents`
  10. help: consider introducing a named lifetime parameter
  11.    |
  12. 28 | pub fn search<'a>(query: &'a str, contents: &'a str) -> Vec<&'a str> {
  13.    |              ++++         ++                 ++              ++
  15. For more information about this error, try `rustc --explain E0106`.
  16. error: could not compile `minigrep` (lib) due to 1 previous error

Rust can’t possibly know which of the two arguments we need, so we need to tell it explicitly. Because contents is the argument that contains all of our text and we want to return the parts of that text that match, we know contents is the argument that should be connected to the return value using the lifetime syntax.

Other programming languages don’t require you to connect arguments to return values in the signature, but this practice will get easier over time. You might want to compare this example with the examples in the “Validating References with Lifetimes” section in Chapter 10.

Now let’s run the test:

  1. $ cargo test
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3.     Finished `test` profile [unoptimized + debuginfo] target(s) in 0.97s
  4.      Running unittests src/lib.rs (target/debug/deps/minigrep-9cd200e5fac0fc94)
  6. running 1 test
  7. test tests::one_result ... FAILED
  9. failures:
  11. ---- tests::one_result stdout ----
  12. thread 'tests::one_result' panicked at src/lib.rs:44:9:
  13. assertion `left == right` failed
  14.   left: ["safe, fast, productive."]
  15.  right: []
  16. note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
  19. failures:
  20.     tests::one_result
  22. test result: FAILED. 0 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
  24. error: test failed, to rerun pass `--lib`

Great, the test fails, exactly as we expected. Let’s get the test to pass!

Writing Code to Pass the Test

Currently, our test is failing because we always return an empty vector. To fix that and implement search, our program needs to follow these steps:

  1. Iterate through each line of the contents.
  2. Check whether the line contains our query string.
  3. If it does, add it to the list of values we’re returning.
  4. If it doesn’t, do nothing.
  5. Return the list of results that match.

Let’s work through each step, starting with iterating through lines.

Iterating Through Lines with the lines Method

Rust has a helpful method to handle line-by-line iteration of strings, conveniently named lines, that works as shown in Listing 12-17. Note that this won’t compile yet.

Filename: src/lib.rs
  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &’static str> {
  11.         if args.len() < 3 {
  12.             return Err(“not enough arguments”);
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     Ok(())
  26. }
  28. pub fn search<’a>(query: &str, contents: &’a str) -> Vec<&’a str> {
  29.     for line in contents.lines() {
  30.         // do something with line
  31.     }
  32. }
  34. #[cfg(test)]
  35. mod tests {
  36.     use super::*;
  38.     #[test]
  39.     fn one_result() {
  40.         let query = “duct”;
  41.         let contents = “\
  42. Rust:
  43. safe, fast, productive.
  44. Pick three.”;
  46.         assert_eq!(vec![“safe, fast, productive.”], search(query, contents));
  47.     }
  48. }
Listing 12-17: Iterating through each line in contents

The lines method returns an iterator. We’ll talk about iterators in depth in Chapter 13, but recall that you saw this way of using an iterator in Listing 3-5, where we used a for loop with an iterator to run some code on each item in a collection.

Searching Each Line for the Query

Next, we’ll check whether the current line contains our query string. Fortunately, strings have a helpful method named contains that does this for us! Add a call to the contains method in the search function, as shown in Listing 12-18. Note that this still won’t compile yet.

Filename: src/lib.rs
  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &’static str> {
  11.         if args.len() < 3 {
  12.             return Err(“not enough arguments”);
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     Ok(())
  26. }
  28. pub fn search<’a>(query: &str, contents: &’a str) -> Vec<&’a str> {
  29.     for line in contents.lines() {
  30.         if line.contains(query) {
  31.             // do something with line
  32.         }
  33.     }
  34. }
  36. #[cfg(test)]
  37. mod tests {
  38.     use super::*;
  40.     #[test]
  41.     fn one_result() {
  42.         let query = “duct”;
  43.         let contents = “\
  44. Rust:
  45. safe, fast, productive.
  46. Pick three.”;
  48.         assert_eq!(vec![“safe, fast, productive.”], search(query, contents));
  49.     }
  50. }
Listing 12-18: Adding functionality to see whether the line contains the string in query

At the moment, we’re building up functionality. To get the code to compile, we need to return a value from the body as we indicated we would in the function signature.

Storing Matching Lines

To finish this function, we need a way to store the matching lines that we want to return. For that, we can make a mutable vector before the for loop and call the push method to store a line in the vector. After the for loop, we return the vector, as shown in Listing 12-19.

Filename: src/lib.rs
  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &’static str> {
  11.         if args.len() < 3 {
  12.             return Err(“not enough arguments”);
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     Ok(())
  26. }
  28. pub fn search<’a>(query: &str, contents: &’a str) -> Vec<&’a str> {
  29.     let mut results = Vec::new();
  31.     for line in contents.lines() {
  32.         if line.contains(query) {
  33.             results.push(line);
  34.         }
  35.     }
  37.     results
  38. }
  40. #[cfg(test)]
  41. mod tests {
  42.     use super::*;
  44.     #[test]
  45.     fn one_result() {
  46.         let query = “duct”;
  47.         let contents = “\
  48. Rust:
  49. safe, fast, productive.
  50. Pick three.”;
  52.         assert_eq!(vec![“safe, fast, productive.”], search(query, contents));
  53.     }
  54. }
Listing 12-19: Storing the lines that match so we can return them

Now the search function should return only the lines that contain query, and our test should pass. Let’s run the test:

  1. $ cargo test
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3.     Finished `test` profile [unoptimized + debuginfo] target(s) in 1.22s
  4.      Running unittests src/lib.rs (target/debug/deps/minigrep-9cd200e5fac0fc94)
  6. running 1 test
  7. test tests::one_result ... ok
  9. test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
  11.      Running unittests src/main.rs (target/debug/deps/minigrep-9cd200e5fac0fc94)
  13. running 0 tests
  15. test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
  17.    Doc-tests minigrep
  19. running 0 tests
  21. test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s

Our test passed, so we know it works!

At this point, we could consider opportunities for refactoring the implementation of the search function while keeping the tests passing to maintain the same functionality. The code in the search function isn’t too bad, but it doesn’t take advantage of some useful features of iterators. We’ll return to this example in Chapter 13, where we’ll explore iterators in detail, and look at how to improve it.

Using the search Function in the run Function

Now that the search function is working and tested, we need to call search from our run function. We need to pass the config.query value and the contents that run reads from the file to the search function. Then run will print each line returned from search:

Filename: src/lib.rs

  1. use std::error::Error;
  2. use std::fs;
  4. pub struct Config {
  5.     pub query: String,
  6.     pub file_path: String,
  7. }
  9. impl Config {
  10.     pub fn build(args: &[String]) -> Result<Config, &'static str> {
  11.         if args.len() < 3 {
  12.             return Err("not enough arguments");
  13.         }
  15.         let query = args[1].clone();
  16.         let file_path = args[2].clone();
  18.         Ok(Config { query, file_path })
  19.     }
  20. }
  22. pub fn run(config: Config) -> Result<(), Box<dyn Error>> {
  23.     let contents = fs::read_to_string(config.file_path)?;
  25.     for line in search(&config.query, &contents) {
  26.         println!("{line}");
  27.     }
  29.     Ok(())
  30. }
  32. pub fn search<'a>(query: &str, contents: &'a str) -> Vec<&'a str> {
  33.     let mut results = Vec::new();
  35.     for line in contents.lines() {
  36.         if line.contains(query) {
  37.             results.push(line);
  38.         }
  39.     }
  41.     results
  42. }
  44. #[cfg(test)]
  45. mod tests {
  46.     use super::*;
  48.     #[test]
  49.     fn one_result() {
  50.         let query = "duct";
  51.         let contents = "\
  52. Rust:
  53. safe, fast, productive.
  54. Pick three.";
  56.         assert_eq!(vec!["safe, fast, productive."], search(query, contents));
  57.     }
  58. }

We’re still using a for loop to return each line from search and print it.

Now the entire program should work! Let’s try it out, first with a word that should return exactly one line from the Emily Dickinson poem: frog.

  1. $ cargo run -- frog poem.txt
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3.     Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.38s
  4.      Running `target/debug/minigrep frog poem.txt`
  5. How public, like a frog

Cool! Now let’s try a word that will match multiple lines, like body:

  1. $ cargo run -- body poem.txt
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3.     Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.0s
  4.      Running `target/debug/minigrep body poem.txt`
  5. I'm nobody! Who are you?
  6. Are you nobody, too?
  7. How dreary to be somebody!

And finally, let’s make sure that we don’t get any lines when we search for a word that isn’t anywhere in the poem, such as monomorphization:

  1. $ cargo run -- monomorphization poem.txt
  2.    Compiling minigrep v0.1.0 (file:///projects/minigrep)
  3.     Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.0s
  4.      Running `target/debug/minigrep monomorphization poem.txt`

Excellent! We’ve built our own mini version of a classic tool and learned a lot about how to structure applications. We’ve also learned a bit about file input and output, lifetimes, testing, and command line parsing.

To round out this project, we’ll briefly demonstrate how to work with environment variables and how to print to standard error, both of which are useful when you’re writing command line programs.