Fixes #1181: Improved "Normalize to BibTeX name format" (#1470)

CHANGELOG.md

@@ -51,6 +51,7 @@ We refer to [GitHub issues](https://github.com/JabRef/jabref/issues) by using `#
 - Fixed [#1882](https://github.com/JabRef/jabref/issues/1882): Crash after saving illegal bibtexkey in entry editor
 - Fixed field `location` containing only city is not exported correctly to MS-Office 2007 xml format
 - Fixed field `key` field is not exported to MS-Office 2008 xml format
+- Fixed [#1181](https://github.com/JabRef/jabref/issues/1181) and [#1504](https://github.com/JabRef/jabref/issues/1504): Improved "Normalize to BibTeX name format": Support separated names with commas and colons. Considered name affixes such as "Jr".
 - Fixed download files failed silently when an invalid directory is selected
 - Fixed [#1949](https://github.com/JabRef/jabref/issues/1949): Error message directs to the wrong preference tab
 - Fixed InvalidBackgroundColor flickering with Ctrl-s and File > Save database

src/main/java/net/sf/jabref/logic/formatter/bibtexfields/NormalizeNamesFormatter.java

@@ -1,5 +1,7 @@
 package net.sf.jabref.logic.formatter.bibtexfields;
 
+import java.util.Objects;
+
 import net.sf.jabref.logic.formatter.Formatter;
 import net.sf.jabref.logic.l10n.Localization;
 import net.sf.jabref.model.entry.AuthorList;
@@ -20,8 +22,9 @@ public String getKey() {
     }
 
     @Override
-    public String format(String value) {
-        AuthorList authorList = AuthorList.parse(value);
+    public String format(String nameList) {
+        Objects.requireNonNull(nameList);
+        AuthorList authorList = AuthorList.parse(nameList);
         return authorList.getAsLastFirstNamesWithAnd(false);
     }
 

src/main/java/net/sf/jabref/model/entry/AuthorList.java

@@ -1,7 +1,11 @@
 package net.sf.jabref.model.entry;
 
+import java.util.Arrays;
+import java.util.Collection;
 import java.util.Collections;
+import java.util.HashSet;
 import java.util.List;
+import java.util.Locale;
 import java.util.Objects;
 import java.util.WeakHashMap;
 import java.util.stream.Collectors;
@@ -130,6 +134,9 @@ public class AuthorList {
 
     private static final WeakHashMap<String, AuthorList> AUTHOR_CACHE = new WeakHashMap<>();
 
+    // Avoid partition where these values are contained
+    private final static Collection<String> avoidTermsInLowerCase = Arrays.asList("jr", "sr", "jnr", "snr", "von", "zu", "van", "der");
+
     /**
      * Creates a new list of authors.
      * <p>
@@ -157,6 +164,46 @@ protected AuthorList(Author author) {
     public static AuthorList parse(String authors) {
         Objects.requireNonNull(authors);
 
+        // Handle case names in order lastname, firstname and separated by ","
+        // E.g., Ali Babar, M., Dingsøyr, T., Lago, P., van der Vliet, H.
+        if (!authors.toUpperCase(Locale.ENGLISH).contains(" AND ") && !authors.contains("{") && !authors.contains(";")) {
+            List<String> arrayNameList = Arrays.asList(authors.split(","));
+
+            // Delete spaces for correct case identification
+            arrayNameList.replaceAll(String::trim);
+
+            // Looking for space between pre- and lastname
+            boolean spaceInAllParts = arrayNameList.stream().filter(name -> name.contains(" ")).collect(Collectors
+                    .toList()).size() == arrayNameList.size();
+
+            // We hit the comma name separator case
+            // Usually the getAsLastFirstNamesWithAnd method would separate them if pre- and lastname are separated with "and"
+            // If not, we check if spaces separate pre- and lastname
+            if (spaceInAllParts) {
+                authors = authors.replaceAll(",", " and");
+            } else {
+                // Looking for name affixes to avoid
+                // arrayNameList needs to reduce by the count off avoiding terms
+                // valuePartsCount holds the count of name parts without the avoided terms
+
+                int valuePartsCount = arrayNameList.size();
+                // Holds the index of each term which needs to be avoided
+                Collection<Integer> avoidIndex = new HashSet<>();
+
+                for (int i = 0; i < arrayNameList.size(); i++) {
+                    if (avoidTermsInLowerCase.contains(arrayNameList.get(i).toLowerCase())) {
+                        avoidIndex.add(i);
+                        valuePartsCount--;
+                    }
+                }
+
+                if ((valuePartsCount % 2) == 0) {
+                    // We hit the described special case with name affix like Jr
+                    authors = buildWithAffix(avoidIndex, arrayNameList).toString();
+                }
+            }
+        }
+
         AuthorList authorList = AUTHOR_CACHE.get(authors);
         if (authorList == null) {
             AuthorListParser parser = new AuthorListParser();
@@ -577,4 +624,34 @@ public String getForAlphabetization() {
         return authorsAlph;
     }
 
+    /**
+     * Builds a new array of strings with stringbuilder.
+     * Regarding to the name affixes.
+     * @return New string with correct seperation
+     */
+    private static StringBuilder buildWithAffix(Collection<Integer> indexArray, List nameList) {
+        StringBuilder stringBuilder = new StringBuilder();
+        // avoidedTimes needs to be increased by the count of avoided terms for correct odd/even calculation
+        int avoidedTimes = 0;
+        for (int i = 0; i < nameList.size(); i++) {
+            if (indexArray.contains(i)) {
+                // We hit a name affix
+                stringBuilder.append(nameList.get(i));
+                stringBuilder.append(',');
+                avoidedTimes++;
+            } else {
+                stringBuilder.append(nameList.get(i));
+                if (((i + avoidedTimes) % 2) == 0) {
+                    // Hit separation between last name and firstname --> comma has to be kept
+                    stringBuilder.append(',');
+                } else {
+                    // Hit separation between full names (e.g., Ali Babar, M. and Dingsøyr, T.) --> semicolon has to be used
+                    // Will be treated correctly by AuthorList.parse(authors);
+                    stringBuilder.append(';');
+                }
+            }
+        }
+        return stringBuilder;
+    }
+
 }

src/test/java/net/sf/jabref/logic/formatter/bibtexfields/NormalizeNamesFormatterTest.java

@@ -1,6 +1,5 @@
 package net.sf.jabref.logic.formatter.bibtexfields;
 
-import org.junit.Assert;
 import org.junit.Before;
 import org.junit.Test;
 
@@ -20,91 +19,160 @@ public void setUp() {
 
     @Test
     public void testNormalizeAuthorList() {
-        expectCorrect("Staci D Bilbo", "Bilbo, Staci D.");
-        expectCorrect("Staci D. Bilbo", "Bilbo, Staci D.");
+        assertEquals("Bilbo, Staci D.", formatter.format("Staci D Bilbo"));
+        assertEquals("Bilbo, Staci D.", formatter.format("Staci D. Bilbo"));
 
-        expectCorrect("Staci D Bilbo and Smith SH and Jaclyn M Schwarz", "Bilbo, Staci D. and Smith, S. H. and Schwarz, Jaclyn M.");
+        assertEquals("Bilbo, Staci D. and Smith, S. H. and Schwarz, Jaclyn M.", formatter.format("Staci D Bilbo and Smith SH and Jaclyn M Schwarz"));
 
-        expectCorrect("Ølver MA", "Ølver, M. A.");
+        assertEquals("Ølver, M. A.", formatter.format("Ølver MA"));
 
-        expectCorrect("Ølver MA; GG Øie; Øie GG; Alfredsen JÅÅ; Jo Alfredsen; Olsen Y.Y. and Olsen YY.",
-                "Ølver, M. A. and Øie, G. G. and Øie, G. G. and Alfredsen, J. Å. Å. and Alfredsen, Jo and Olsen, Y. Y. and Olsen, Y. Y.");
+        assertEquals("Ølver, M. A. and Øie, G. G. and Øie, G. G. and Alfredsen, J. Å. Å. and Alfredsen, Jo and Olsen, Y. Y. and Olsen, Y. Y.",
+                formatter.format("Ølver MA; GG Øie; Øie GG; Alfredsen JÅÅ; Jo Alfredsen; Olsen Y.Y. and Olsen YY."));
 
-        expectCorrect("Ølver MA; GG Øie; Øie GG; Alfredsen JÅÅ; Jo Alfredsen; Olsen Y.Y.; Olsen YY.",
-                "Ølver, M. A. and Øie, G. G. and Øie, G. G. and Alfredsen, J. Å. Å. and Alfredsen, Jo and Olsen, Y. Y. and Olsen, Y. Y.");
+        assertEquals("Ølver, M. A. and Øie, G. G. and Øie, G. G. and Alfredsen, J. Å. Å. and Alfredsen, Jo and Olsen, Y. Y. and Olsen, Y. Y.",
+                formatter.format("Ølver MA; GG Øie; Øie GG; Alfredsen JÅÅ; Jo Alfredsen; Olsen Y.Y.; Olsen YY."));
 
-        expectCorrect("Alver, Morten and Alver, Morten O and Alfredsen, JA and Olsen, Y.Y.", "Alver, Morten and Alver, Morten O. and Alfredsen, J. A. and Olsen, Y. Y.");
+        assertEquals("Alver, Morten and Alver, Morten O. and Alfredsen, J. A. and Olsen, Y. Y.", formatter.format("Alver, Morten and Alver, Morten O and Alfredsen, JA and Olsen, Y.Y."));
 
-        expectCorrect("Alver, MA; Alfredsen, JA; Olsen Y.Y.", "Alver, M. A. and Alfredsen, J. A. and Olsen, Y. Y.");
+        assertEquals("Alver, M. A. and Alfredsen, J. A. and Olsen, Y. Y.", formatter.format("Alver, MA; Alfredsen, JA; Olsen Y.Y."));
 
-        expectCorrect("Kolb, Stefan and J{\\\"o}rg Lenhard and Wirtz, Guido", "Kolb, Stefan and Lenhard, J{\\\"o}rg and Wirtz, Guido");
+        assertEquals("Kolb, Stefan and Lenhard, J{\\\"o}rg and Wirtz, Guido", formatter.format("Kolb, Stefan and J{\\\"o}rg Lenhard and Wirtz, Guido"));
     }
 
     @Test
     public void twoAuthorsSeperatedByColon() {
-        expectCorrect("Staci Bilbo; Morten Alver", "Bilbo, Staci and Alver, Morten");
+        assertEquals("Bilbo, Staci and Alver, Morten", formatter.format("Staci Bilbo; Morten Alver"));
     }
 
     @Test
     public void threeAuthorsSeperatedByColon() {
-        expectCorrect("Staci Bilbo; Morten Alver; Test Name", "Bilbo, Staci and Alver, Morten and Name, Test");
+        assertEquals("Bilbo, Staci and Alver, Morten and Name, Test", formatter.format("Staci Bilbo; Morten Alver; Test Name"));
     }
 
     // Test for https://github.com/JabRef/jabref/issues/318
     @Test
     public void threeAuthorsSeperatedByAnd() {
-        expectCorrect("Stefan Kolb and J{\\\"o}rg Lenhard and Guido Wirtz", "Kolb, Stefan and Lenhard, J{\\\"o}rg and Wirtz, Guido");
+        assertEquals("Kolb, Stefan and Lenhard, J{\\\"o}rg and Wirtz, Guido", formatter.format("Stefan Kolb and J{\\\"o}rg Lenhard and Guido Wirtz"));
     }
 
     // Test for https://github.com/JabRef/jabref/issues/318
     @Test
     public void threeAuthorsSeperatedByAndWithDash() {
-        expectCorrect("Heng-Yu Jian and Xu, Z. and Chang, M.-C.F.", "Jian, Heng-Yu and Xu, Z. and Chang, M.-C. F.");
+        assertEquals("Jian, Heng-Yu and Xu, Z. and Chang, M.-C. F.", formatter.format("Heng-Yu Jian and Xu, Z. and Chang, M.-C.F."));
     }
 
     // Test for https://github.com/JabRef/jabref/issues/318
     @Test
     public void threeAuthorsSeperatedByAndWithLatex() {
-        expectCorrect("Oscar Gustafsson and Linda S. DeBrunner and Victor DeBrunner and H{\\aa}kan Johansson", "Gustafsson, Oscar and DeBrunner, Linda S. and DeBrunner, Victor and Johansson, H{\\aa}kan");
+        assertEquals("Gustafsson, Oscar and DeBrunner, Linda S. and DeBrunner, Victor and Johansson, H{\\aa}kan",
+                formatter.format("Oscar Gustafsson and Linda S. DeBrunner and Victor DeBrunner and H{\\aa}kan Johansson"));
     }
 
     @Test
     public void lastThenInitial() {
-        expectCorrect("Smith S", "Smith, S.");
+        assertEquals("Smith, S.", formatter.format("Smith S"));
     }
 
     @Test
     public void lastThenInitials() {
-        expectCorrect("Smith SH", "Smith, S. H.");
+        assertEquals("Smith, S. H.", formatter.format("Smith SH"));
     }
 
     @Test
     public void initialThenLast() {
-        expectCorrect("S Smith", "Smith, S.");
+        assertEquals("Smith, S.", formatter.format("S Smith"));
     }
 
     @Test
     public void initialDotThenLast() {
-        expectCorrect("S. Smith", "Smith, S.");
+        assertEquals("Smith, S.", formatter.format("S. Smith"));
     }
 
     @Test
     public void initialsThenLast() {
-        expectCorrect("SH Smith", "Smith, S. H.");
+        assertEquals("Smith, S. H.", formatter.format("SH Smith"));
     }
 
     @Test
     public void lastThenJuniorThenFirst() {
-        expectCorrect("Name, della, first", "Name, della, first");
+        assertEquals("Name, della, first", formatter.format("Name, della, first"));
     }
 
-    private void expectCorrect(String input, String expected) {
-        Assert.assertEquals(expected, formatter.format(input));
+    @Test
+    public void testConcatenationOfAuthorsWithCommas() {
+        assertEquals("Ali Babar, M. and Dingsøyr, T. and Lago, P. and van der Vliet, H.", formatter.format("Ali Babar, M., Dingsøyr, T., Lago, P., van der Vliet, H."));
+        assertEquals("Ali Babar, M.", formatter.format("Ali Babar, M."));
+    }
+
+    @Test
+    public void testOddCountOfCommas() {
+        assertEquals("Ali Babar, M., Dingsøyr T. Lago P.", formatter.format("Ali Babar, M., Dingsøyr, T., Lago P."));
     }
 
     @Test
     public void formatExample() {
-        assertEquals("Einstein, Albert and Turing, Alan", formatter.format(formatter.getExampleInput()));
+        assertEquals(formatter.format(formatter.getExampleInput()), "Einstein, Albert and Turing, Alan");
+    }
+
+    @Test
+    public void testNameAffixe() {
+        assertEquals("Surname, jr, First and Surname2, First2", formatter.format("Surname, jr, First, Surname2, First2"));
+    }
+
+    @Test
+    public void testAvoidSpecialCharacter() {
+        assertEquals("Surname, {, First; Surname2, First2", formatter.format("Surname, {, First; Surname2, First2"));
+    }
+
+    @Test
+    public void testAndInName() {
+        assertEquals("Surname and , First, Surname2 First2", formatter.format("Surname, and , First, Surname2, First2"));
+    }
+
+    @Test
+    public void testMultipleNameAffixes() {
+        assertEquals("Mair, Jr, Daniel and Brühl, Sr, Daniel", formatter.format("Mair, Jr, Daniel, Brühl, Sr, Daniel"));
+    }
+
+    @Test
+    public void testCommaSeperatedNames() {
+        assertEquals("Bosoi, Cristina and Oliveira, Mariana and Sanchez, Rafael Ochoa and Tremblay, Mélanie and TenHave, Gabrie and Deutz, Nicoolas and Rose, Christopher F. and Bemeur, Chantal",
+                formatter.format("Cristina Bosoi, Mariana Oliveira, Rafael Ochoa Sanchez, Mélanie Tremblay, Gabrie TenHave, Nicoolas Deutz, Christopher F. Rose, Chantal Bemeur"));
+    }
+
+    @Test
+    public void testMultipleSpaces() {
+        assertEquals("Bosoi, Cristina and Oliveira, Mariana and Sanchez, Rafael Ochoa and Tremblay, Mélanie and TenHave, Gabrie and Deutz, Nicoolas and Rose, Christopher F. and Bemeur, Chantal",
+                formatter.format("Cristina    Bosoi,    Mariana Oliveira, Rafael Ochoa Sanchez   ,   Mélanie Tremblay  , Gabrie TenHave, Nicoolas Deutz, Christopher F. Rose, Chantal Bemeur"));
+    }
+
+    @Test
+    public void testAvoidPreposition() {
+        assertEquals("von Zimmer, Hans and van Oberbergern, Michael and zu Berger, Kevin", formatter.format("Hans von Zimmer, Michael van Oberbergern, Kevin zu Berger"));
+    }
+
+    @Test
+    public void testPreposition() {
+        assertEquals("von Zimmer, Hans and van Oberbergern, Michael and zu Berger, Kevin", formatter.format("Hans von Zimmer, Michael van Oberbergern, Kevin zu Berger"));
     }
 
+    @Test
+    public void testAvoidNameAffixes() {
+        assertEquals("der Barbar, Canon and der Große, Alexander", formatter.format("Canon der Barbar, Alexander der Große"));
+    }
+
+    @Test
+    public void testUpperCaseSensitiveList() {
+        assertEquals("der Barbar, Canon and der Große, Alexander", formatter.format("Canon der Barbar AND Alexander der Große"));
+        assertEquals("der Barbar, Canon and der Große, Alexander", formatter.format("Canon der Barbar aNd Alexander der Große"));
+        assertEquals("der Barbar, Canon and der Große, Alexander", formatter.format("Canon der Barbar AnD Alexander der Große"));
+    }
+
+    @Test
+    public void testSemiCorrectNamesWithSemicolon() {
+        assertEquals("Last, First and Last2, First2 and Last3, First3", formatter.format("Last, First; Last2, First2; Last3, First3"));
+        assertEquals("Last, Jr, First and Last2, First2", formatter.format("Last, Jr, First; Last2, First2"));
+        assertEquals("Last, First and Last2, First2 and Last3, First3 and Last4, First4", formatter.format("Last, First; Last2, First2; Last3, First3; First4 Last4"));
+        assertEquals("Last and Last2, First2 and Last3, First3 and Last4, First4", formatter.format("Last; Last2, First2; Last3, First3; Last4, First4"));
+    }
 }

src/test/java/net/sf/jabref/model/entry/AuthorListTest.java

@@ -524,6 +524,15 @@ public void testGetAuthorsLastFirstAnds() {
 
     }
 
+    @Test
+    public void testGetAuthorsLastFirstAndsCaching() {
+        // getAsLastFirstNamesWithAnd caches its results, therefore we call the method twice using the same arguments
+        Assert.assertEquals("Smith, John", AuthorList.parse("John Smith").getAsLastFirstNamesWithAnd(false));
+        Assert.assertEquals("Smith, John", AuthorList.parse("John Smith").getAsLastFirstNamesWithAnd(false));
+        Assert.assertEquals("Smith, J.", AuthorList.parse("John Smith").getAsLastFirstNamesWithAnd(true));
+        Assert.assertEquals("Smith, J.", AuthorList.parse("John Smith").getAsLastFirstNamesWithAnd(true));
+    }
+
     @Test
     public void testGetAuthorsFirstFirst() {
 
@@ -636,4 +645,5 @@ public void parseNameWithBraces() throws Exception {
         Author expected = new Author("H{e}lene", "H.", null, "Fiaux", null);
         Assert.assertEquals(new AuthorList(expected), AuthorList.parse("H{e}lene Fiaux"));
     }
+
 }