From 6dcec4f473460847a7c35a34edb7af7d7ce985bb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Cl=C3=A9ment=20Renault?= <renault.cle@gmail.com>
Date: Mon, 24 Sep 2018 14:50:33 +0200
Subject: [PATCH] feat: Correct the sum of typos criterion

Thanks to @tpayet!
---
 raptor/src/rank/sum_of_typos.rs | 111 ++++++++++++++++++++++++++++++--
 1 file changed, 106 insertions(+), 5 deletions(-)

diff --git a/raptor/src/rank/sum_of_typos.rs b/raptor/src/rank/sum_of_typos.rs
index 4d1c80eea..ab90595c4 100644
--- a/raptor/src/rank/sum_of_typos.rs
+++ b/raptor/src/rank/sum_of_typos.rs
@@ -4,13 +4,18 @@ use crate::Match;
 use crate::rank::{match_query_index, Document};
 
 #[inline]
-fn sum_matches_typos(matches: &[Match]) -> u8 {
+fn sum_matches_typos(matches: &[Match]) -> i8 {
+    let mut sum_typos = 0;
+    let mut number_words = 0;
+
     // note that GroupBy will never return an empty group
     // so we can do this assumption safely
-    // matches must and will never be empty
-    GroupBy::new(matches, match_query_index).map(|group| unsafe {
-        group.get_unchecked(0).distance
-    }).min().unwrap()
+    for group in GroupBy::new(matches, match_query_index) {
+        sum_typos += unsafe { group.get_unchecked(0).distance } as i8;
+        number_words += 1;
+    }
+
+    sum_typos - number_words
 }
 
 #[inline]
@@ -20,3 +25,99 @@ pub fn sum_of_typos(lhs: &Document, rhs: &Document) -> Ordering {
 
     lhs.cmp(&rhs)
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    // typing: "Geox CEO"
+    //
+    // doc0: "Geox SpA: CEO and Executive"
+    // doc1: "Mt. Gox CEO Resigns From Bitcoin Foundation"
+    #[test]
+    fn one_typo_reference() {
+        let doc0 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
+                Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 2, is_exact: false },
+            ];
+            Document {
+                document_id: 0,
+                matches: matches,
+            }
+        };
+
+        let doc1 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 1, attribute: 0, attribute_index: 0, is_exact: false },
+                Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 2, is_exact: false },
+            ];
+            Document {
+                document_id: 1,
+                matches: matches,
+            }
+        };
+
+        assert_eq!(sum_of_typos(&doc0, &doc1), Ordering::Less);
+    }
+
+    // typing: "bouton manchette"
+    //
+    // doc0: "bouton manchette"
+    // doc1: "bouton"
+    #[test]
+    fn no_typo() {
+        let doc0 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
+                Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 1, is_exact: false },
+            ];
+            Document {
+                document_id: 0,
+                matches: matches,
+            }
+        };
+
+        let doc1 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
+            ];
+            Document {
+                document_id: 1,
+                matches: matches,
+            }
+        };
+
+        assert_eq!(sum_of_typos(&doc0, &doc1), Ordering::Less);
+    }
+
+    // typing: "bouton manchztte"
+    //
+    // doc0: "bouton manchette"
+    // doc1: "bouton"
+    #[test]
+    fn one_typo() {
+        let doc0 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
+                Match { query_index: 1, distance: 1, attribute: 0, attribute_index: 1, is_exact: false },
+            ];
+            Document {
+                document_id: 0,
+                matches: matches,
+            }
+        };
+
+        let doc1 = {
+            let matches = vec![
+                Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
+            ];
+            Document {
+                document_id: 1,
+                matches: matches,
+            }
+        };
+
+        assert_eq!(sum_of_typos(&doc0, &doc1), Ordering::Equal);
+    }
+}