Remove the trivial but annoying avg field

havenwood · Oct 10, 2024 · 0218370 · 0218370
1 parent 2876633
commit 0218370
Show file tree

Hide file tree

Showing 4 changed files with 5 additions and 49 deletions.
diff --git a/src/lib.rs b/src/lib.rs
@@ -15,7 +15,7 @@
 //! `Original` (case sensitive) and `Lower` or `Upper` case normalization. `Sort`
 //! order can be `Unsorted` or sorted `Desc` (descending) or `Asc` (ascending).
 //! A `tally` can be sorted upon contruction or sorted later with the `sort` method.
-//! Sorting doesn't impact the `count`, `uniq_count` or `avg` fields. `Filter`s can
+//! Sorting doesn't impact the `count` or `uniq_count` fields. `Filter`s can
 //! be used to provide list of words that should or shouldn't be tallied.
 //!
 //! # Examples
@@ -29,7 +29,6 @@
 //!
 //! assert_eq!(words.tally(), expected_tally);
 //! ```
-use core::hash::{Hash, Hasher};
 use indexmap::IndexMap;
 #[cfg(feature = "serde")]
 use serde::{Deserialize, Serialize};
@@ -43,7 +42,7 @@ pub use filters::{Filters, MinChars, MinCount, WordsExclude, WordsOnly};
 pub use options::{Case, Options, Sort};
 
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Clone, Debug, Default, PartialEq)]
+#[derive(Clone, Debug, Default, PartialEq, Eq, Hash)]
 #[non_exhaustive]
 pub struct WordTally {
     /// Ordered pairs of words and the count of times they appear.
@@ -54,18 +53,6 @@ pub struct WordTally {
 
     /// The sum of uniq words tallied.
     uniq_count: usize,
-
-    /// The mean average count per word, if there are words.
-    avg: Option<f64>,
-}
-
-impl Eq for WordTally {}
-
-/// Since the other fields are derived from it, hash by just the `tally`.
-impl Hash for WordTally {
-    fn hash<H: Hasher>(&self, state: &mut H) {
-        self.tally.hash(state);
-    }
 }
 
 /// A `tally` supports `iter` and can also be represented as a `Vec`.
@@ -85,12 +72,10 @@ impl WordTally {
         let count = tally_map.values().sum();
         let tally: Box<[_]> = tally_map.into_iter().collect();
         let uniq_count = tally.len();
-        let avg = Self::calculate_avg(count, uniq_count);
         let mut word_tally = Self {
             tally,
             count,
             uniq_count,
-            avg,
         };
         word_tally.sort(options.sort);
 
@@ -117,17 +102,6 @@ impl WordTally {
         self.count
     }
 
-    /// Gets the `avg` field.
-    pub const fn avg(&self) -> Option<f64> {
-        self.avg
-    }
-
-    /// Calculates an approximate mean average word count if there are words.
-    /// Note: Casting `u64` to `f64` and floating point arithmatic cause a loss of precision.
-    fn calculate_avg(count: u64, uniq_count: usize) -> Option<f64> {
-        (count > 0).then(|| count as f64 / uniq_count as f64)
-    }
-
     /// Creates a tally of normalized words from an input that implements `Read`.
     fn tally_map<T: Read>(input: T, case: Case) -> IndexMap<Box<str>, u64> {
         let mut tally = IndexMap::new();

diff --git a/src/main.rs b/src/main.rs
@@ -140,10 +140,6 @@ fn log_details(
         log_detail(&mut w, "source", delimiter, source)?;
         log_detail(&mut w, "total-words", delimiter, word_tally.count())?;
         log_detail(&mut w, "unique-words", delimiter, word_tally.uniq_count())?;
-
-        if let Some(avg) = word_tally.avg() {
-            log_detail(&mut w, "average-word-count", delimiter, format!("{avg:.3}"))?;
-        }
     }
 
     if log_config.debug {

diff --git a/tests/lib.rs b/tests/lib.rs
@@ -9,7 +9,6 @@ const TEST_WORDS_PATH: &str = "tests/files/words.txt";
 struct ExpectedFields<'a> {
     count: u64,
     uniq_count: usize,
-    avg: Option<f64>,
     tally: Vec<(&'a str, u64)>,
 }
 
@@ -24,7 +23,6 @@ fn word_tally_test(case: Case, sort: Sort, filters: Filters, fields: &ExpectedFi
     let word_tally = word_tally(Options { case, sort }, filters);
     assert_eq!(word_tally.count(), fields.count);
     assert_eq!(word_tally.uniq_count(), fields.uniq_count);
-    assert_eq!(word_tally.avg(), fields.avg);
 
     let expected_tally = fields
         .tally
@@ -44,7 +42,6 @@ fn lower_case_desc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("c", 15), ("d", 11), ("123", 9), ("b", 7), ("a", 3)],
         },
     );
@@ -62,7 +59,6 @@ fn min_char_count_at_max() {
         &ExpectedFields {
             count: 9,
             uniq_count: 1,
-            avg: Some(9.0),
             tally: vec![("123", 9)],
         },
     );
@@ -80,7 +76,6 @@ fn min_char_count_above_max() {
         &ExpectedFields {
             count: 0,
             uniq_count: 0,
-            avg: None,
             tally: vec![],
         },
     );
@@ -95,7 +90,6 @@ fn min_char_count_at_min() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("c", 15), ("d", 11), ("123", 9), ("b", 7), ("a", 3)],
         },
     );
@@ -113,7 +107,6 @@ fn min_word_count_at_max() {
         &ExpectedFields {
             count: 15,
             uniq_count: 1,
-            avg: Some(15.0),
             tally: vec![("c", 15)],
         },
     );
@@ -128,7 +121,6 @@ fn default_case_unsorted_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("d", 11), ("123", 9), ("a", 3), ("c", 15), ("b", 7)],
         },
     );
@@ -143,7 +135,6 @@ fn upper_case_desc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("C", 15), ("D", 11), ("123", 9), ("B", 7), ("A", 3)],
         },
     );
@@ -158,7 +149,6 @@ fn lower_case_asc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("a", 3), ("b", 7), ("123", 9), ("d", 11), ("c", 15)],
         },
     );
@@ -173,7 +163,6 @@ fn upper_case_asc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 5,
-            avg: Some(9.0),
             tally: vec![("A", 3), ("B", 7), ("123", 9), ("D", 11), ("C", 15)],
         },
     );
@@ -188,7 +177,6 @@ fn original_case_desc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 9,
-            avg: Some(5.0),
             tally: vec![
                 ("123", 9),
                 ("C", 8),
@@ -213,7 +201,6 @@ fn original_case_asc_order() {
         &ExpectedFields {
             count: 45,
             uniq_count: 9,
-            avg: Some(5.0),
             tally: vec![
                 ("a", 1),
                 ("A", 2),
@@ -391,7 +378,7 @@ fn test_to_json() {
     );
     let serialized = serde_json::to_string(&expected).unwrap();
 
-    let expected_json = r#"{"tally":[["wombat",2],["bat",1]],"count":3,"uniq_count":2,"avg":1.5}"#;
+    let expected_json = r#"{"tally":[["wombat",2],["bat",1]],"count":3,"uniq_count":2}"#;
     assert_eq!(serialized, expected_json);
 }
 
@@ -407,8 +394,7 @@ fn test_from_json() {
     {
         "tally": [["wombat", 2], ["bat", 1]],
         "count": 3,
-        "uniq_count": 2,
-        "avg": 1.5
+        "uniq_count": 2
     }
     "#;
 

diff --git a/tests/main.rs b/tests/main.rs
@@ -75,7 +75,7 @@ fn verbose_with_input() {
     let assert = word_tally().write_stdin("wombat").arg("-v").assert();
     assert
         .success()
-        .stderr("source -\ntotal-words 1\nunique-words 1\naverage-word-count 1.000\n\n")
+        .stderr("source -\ntotal-words 1\nunique-words 1\n\n")
         .stdout("wombat 1\n");
 }