GH-355: Handle duplicates in --people-dict and extend docs

Signed-off-by: Oliver Bristow <oliver.bristow@project-tracr.com>

Author: Code0x58

README.md

@@ -223,7 +223,14 @@ and add the unknown email to the list of that developer's emails.
 
 If `--people-dict` is specified, it should point to a text file with the custom identities. The
 format is: every line is a single developer, it contains all the matching emails and names separated
-by `|`. The case is ignored.
+by `|`. The case is ignored. Example file contents:
+```
+Linus Torvalds|torvalds@linux-foundation.org
+Vadim Markovtsev|vadim@sourced.tech|another@one.com
+```
+
+If `--people-dict` is not specified a [`.mailmap`](https://git-scm.com/docs/git-check-mailmap) file
+will be used if it exists in the latest commit.
 
 #### Overwrites matrix
 

internal/plumbing/identity/identity.go

@@ -186,11 +186,18 @@ func (detector *Detector) LoadPeopleDict(path string) error {
 	size := 0
 	for scanner.Scan() {
 		ids := strings.Split(scanner.Text(), "|")
+		canon := ids[0]
+		var exists bool
+		var canon_i int
+		// lookup or create a new canonical value
+		if canon_i, exists = dict[canon]; !exists {
+			reverseDict = append(reverseDict, canon)
+			size++
+			canon_i = size
+		}
 		for _, id := range ids {
-			dict[strings.ToLower(id)] = size
+			dict[strings.ToLower(id)] = canon_i
 		}
-		reverseDict = append(reverseDict, ids[0])
-		size++
 	}
 	reverseDict = append(reverseDict, AuthorMissingName)
 	detector.PeopleDict = dict

internal/plumbing/identity/identity_test.go

@@ -177,19 +177,32 @@ func TestIdentityDetectorLoadPeopleDict(t *testing.T) {
 	id := fixtureIdentityDetector()
 	err := id.LoadPeopleDict(path.Join("..", "..", "test_data", "identities"))
 	assert.Nil(t, err)
-	assert.Equal(t, len(id.PeopleDict), 7)
+	assert.Equal(t, len(id.PeopleDict), 10)
 	assert.Contains(t, id.PeopleDict, "linus torvalds")
 	assert.Contains(t, id.PeopleDict, "torvalds@linux-foundation.org")
 	assert.Contains(t, id.PeopleDict, "vadim markovtsev")
 	assert.Contains(t, id.PeopleDict, "vadim@sourced.tech")
 	assert.Contains(t, id.PeopleDict, "another@one.com")
 	assert.Contains(t, id.PeopleDict, "máximo cuadros")
 	assert.Contains(t, id.PeopleDict, "maximo@sourced.tech")
-	assert.Equal(t, len(id.ReversedPeopleDict), 4)
+	assert.Contains(t, id.PeopleDict, "duplicate")
+	assert.Contains(t, id.PeopleDict, "first@example.com")
+	assert.Contains(t, id.PeopleDict, "second@example.com")
+
+	assert.Equal(t, len(id.ReversedPeopleDict), 5)
 	assert.Equal(t, id.ReversedPeopleDict[0], "Linus Torvalds")
 	assert.Equal(t, id.ReversedPeopleDict[1], "Vadim Markovtsev")
 	assert.Equal(t, id.ReversedPeopleDict[2], "Máximo Cuadros")
-	assert.Equal(t, id.ReversedPeopleDict[3], AuthorMissingName)
+	assert.Equal(t, id.ReversedPeopleDict[3], "Duplicate")
+	assert.Equal(t, id.ReversedPeopleDict[4], AuthorMissingName)
+
+	assert.Equal(t, id.PeopleDict["duplicate"], id.PeopleDict["first@example.com"])
+	assert.Equal(t, id.PeopleDict["duplicate"], id.PeopleDict["second@example.com"])
+
+	assert.Equal(t, id.PeopleDict["vadim markovtsev"], id.PeopleDict["vadim@sourced.tech"])
+	assert.Equal(t, id.PeopleDict["vadim markovtsev"], id.PeopleDict["another@one.com"])
+
+	assert.NotEqual(t, id.PeopleDict["duplicate"], id.PeopleDict["vadim markovtsev"])
 }
 
 func TestIdentityDetectorLoadPeopleDictWrongPath(t *testing.T) {

internal/test_data/identities

@@ -1,3 +1,5 @@
 Linus Torvalds|torvalds@linux-foundation.org
 Vadim Markovtsev|vadim@sourced.tech|another@one.com
-Máximo Cuadros|maximo@sourced.tech
+Máximo Cuadros|maximo@sourced.tech
+Duplicate|first@example.com
+Duplicate|second@example.com