Fix month creation. #name# is really used within JabRef to determine …

…a BibTeX string content (and not plain text)

- Fix tests
- Sort BibEntryTests: default constructor, keywords, all others
- Fix name of "CanonicalBibEntry" (from CanonicalBibtexEntry)
- Revert "fix jabref format of month", this also reverts commit 77c25ca.
- Format some code
- Add comments

src/main/java/org/jabref/logic/importer/fileformat/BibTeXMLImporter.java

@@ -179,7 +179,7 @@ private <T> void parse(T entryType, Map<Field, String> fields) {
                     putNumber(fields, (BigInteger) method.invoke(entryType));
                     continue;
                 } else if (method.getName().equals("getMonth")) {
-                    putMonth(fields, Month.parse((String)method.invoke(entryType)));
+                    putMonth(fields, Month.parse((String) method.invoke(entryType)));
                     continue;
                 } else if (isMethodToIgnore(method.getName())) {
                     continue;

src/main/java/org/jabref/logic/importer/fileformat/BibtexParser.java

@@ -589,7 +589,6 @@ private String parseFieldContent(Field field) throws IOException {
         int character;
 
         while (((character = peek()) != ',') && (character != '}') && (character != ')')) {
-
             if (eof) {
                 throw new IOException("Error in line " + line + ": EOF in mid-string");
             }
@@ -602,7 +601,6 @@ private String parseFieldContent(Field field) throws IOException {
                 // brackets to know when the string is finished.
                 StringBuilder text = parseBracketedTextExactly();
                 value.append(fieldContentParser.format(text, field));
-
             } else if (Character.isDigit((char) character)) { // value is a number
                 String number = parseTextToken();
                 value.append(number);
@@ -614,16 +612,11 @@ private String parseFieldContent(Field field) throws IOException {
                     throw new IOException("Error in line " + line + " or above: "
                             + "Empty text token.\nThis could be caused " + "by a missing comma between two fields.");
                 }
-                if (field != StandardField.MONTH) {
-                    value.append('#').append(textToken).append('#');
-                } else {
-                    value.append(textToken);
-                }
+                value.append('#').append(textToken).append('#');
             }
             skipWhitespace();
         }
         return value.toString();
-
     }
 
     /**

src/main/java/org/jabref/logic/importer/fileformat/IsiImporter.java

@@ -143,7 +143,7 @@ private static void processCapitalization(Map<Field, String> map) {
     public ParserResult importDatabase(BufferedReader reader) throws IOException {
         Objects.requireNonNull(reader);
 
-        List<BibEntry> bibitems = new ArrayList<>();
+        List<BibEntry> bibEntries = new ArrayList<>();
         StringBuilder sb = new StringBuilder();
 
         // Pattern fieldPattern = Pattern.compile("^AU |^TI |^SO |^DT |^C1 |^AB
@@ -233,7 +233,6 @@ public ParserResult importDatabase(BufferedReader reader) throws IOException {
                 } else if ("SO".equals(beg) || "JA".equals(beg)) {
                     hm.put(StandardField.JOURNAL, EOL_PATTERN.matcher(value).replaceAll(" "));
                 } else if ("ID".equals(beg) || "KW".equals(beg)) {
-
                     value = EOL_PATTERN.matcher(value).replaceAll(" ");
                     String existingKeywords = hm.get(StandardField.KEYWORDS);
                     if ((existingKeywords == null) || existingKeywords.contains(value)) {
@@ -242,7 +241,6 @@ public ParserResult importDatabase(BufferedReader reader) throws IOException {
                         existingKeywords += ", " + value;
                     }
                     hm.put(StandardField.KEYWORDS, existingKeywords);
-
                 } else if ("AB".equals(beg)) {
                     hm.put(StandardField.ABSTRACT, EOL_PATTERN.matcher(value).replaceAll(" "));
                 } else if ("BP".equals(beg) || "BR".equals(beg) || "SP".equals(beg)) {
@@ -271,12 +269,10 @@ public ParserResult importDatabase(BufferedReader reader) throws IOException {
                 } else if ("DI".equals(beg)) {
                     hm.put(StandardField.DOI, value);
                 } else if ("PD".equals(beg)) {
-
                     String month = IsiImporter.parseMonth(value);
                     if (month != null) {
                         hm.put(StandardField.MONTH, month);
                     }
-
                 } else if ("DT".equals(beg)) {
                     if ("Review".equals(value)) {
                         type = StandardEntryType.Article; // set "Review" in Note/Comment?
@@ -327,17 +323,19 @@ public ParserResult importDatabase(BufferedReader reader) throws IOException {
 
             b.setField(hm);
 
-            bibitems.add(b);
+            bibEntries.add(b);
         }
-        return new ParserResult(bibitems);
+        return new ParserResult(bibEntries);
     }
 
     private static String parsePages(String value) {
         return value.replace("-", "--");
     }
 
-    public static String parseMonth(String value) {
-
+    /**
+     * Parses the month and returns it in the JabRef format
+     */
+    static String parseMonth(String value) {
         String[] parts = value.split("\\s|\\-");
         for (String part1 : parts) {
             Optional<Month> month = Month.getMonthByShortName(part1.toLowerCase(Locale.ROOT));

src/main/java/org/jabref/logic/importer/fileformat/RepecNepImporter.java

@@ -330,22 +330,22 @@ private void parseAdditionalFields(BibEntry be, boolean multilineUrlFieldAllowed
             // skip keyword
             this.lastLine = "".equals(this.lastLine) ? "" : this.lastLine.substring(this.lastLine.indexOf(':') + 1).trim();
 
-            // parse keywords field
             if ("Keywords".equals(keyword)) {
+                // parse keywords field
                 String content = readMultipleLines(in);
                 String[] keywords = content.split("[,;]");
                 be.addKeywords(Arrays.asList(keywords),
                         importFormatPreferences.getKeywordSeparator());
-                // parse JEL field
             } else if ("JEL".equals(keyword)) {
+                // parse JEL field
                 be.setField(new UnknownField("jel"), readMultipleLines(in));
 
             } else if (keyword.startsWith("Date")) {
                 // parse date field
                 String content = readMultipleLines(in);
                 Date.parse(content).ifPresent(be::setDate);
-                // parse URL field
             } else if (keyword.startsWith("URL")) {
+                // parse URL field
                 String content;
                 if (multilineUrlFieldAllowed) {
                     content = readMultipleLines(in);
@@ -354,10 +354,8 @@ private void parseAdditionalFields(BibEntry be, boolean multilineUrlFieldAllowed
                     readLine(in);
                 }
                 be.setField(StandardField.URL, content);
-
-                // authors field
             } else if (keyword.startsWith("By")) {
-                // parse authors
+                // parse authors field
                 parseAuthors(be, in);
             } else {
                 readLine(in);

src/main/java/org/jabref/logic/xmp/DublinCoreExtractor.java

@@ -67,13 +67,11 @@ private void extractAuthor() {
 
     /**
      * Year in BibTex - Date in DublinCore is only the year information, because dc interprets empty months as January.
-     * Tries to extract the month as well.
-     * In JabRef the bibtex/month/value is prioritized.
-     * <br/>
-     * The problem is the default value of the calendar, which is always January, also if there is no month information in
-     * the xmp metdata. The idea is, to reject all information with YYYY-01-01. In cases, where xmp is written with JabRef
-     * the month property filled with jan will override this behavior and no data is lost. In the cases, where xmp
-     * is written by another service, the assumption is, that the 1st January is not a publication date at all.
+     * Tries to extract the month as well. In JabRef the bibtex/month/value is prioritized. <br/> The problem is the
+     * default value of the calendar, which is always January, also if there is no month information in the xmp metdata.
+     * The idea is, to reject all information with YYYY-01-01. In cases, where xmp is written with JabRef the month
+     * property filled with jan will override this behavior and no data is lost. In the cases, where xmp is written by
+     * another service, the assumption is, that the 1st January is not a publication date at all.
      */
     private void extractYearAndMonth() {
         List<String> dates = dcSchema.getUnqualifiedSequenceValueList("date");
@@ -87,12 +85,11 @@ private void extractYearAndMonth() {
             }
             if (calender != null) {
                 bibEntry.setField(StandardField.YEAR, String.valueOf(calender.get(Calendar.YEAR)));
+                int monthNumber = calender.get(Calendar.MONTH) + 1;
                 // not the 1st of January
-                if (!((calender.get(Calendar.MONTH) == 0) && (calender.get(Calendar.DAY_OF_MONTH) == 1))) {
-                    Optional<Month> month = Month.getMonthByNumber(calender.get(Calendar.MONTH) + 1);
-                    if (month.isPresent()) {
-                        bibEntry.setField(StandardField.MONTH, month.get().getShortName());
-                    }
+                if (!((monthNumber == 1) && (calender.get(Calendar.DAY_OF_MONTH) == 1))) {
+                    Month.getMonthByNumber(monthNumber)
+                         .ifPresent(month -> bibEntry.setMonth(month));
                 }
             }
         }
@@ -129,15 +126,13 @@ private void extractPublisher() {
     }
 
     /**
-     * This method sets all fields, which are custom in bibtext and therefore supported by jabref, but which are not included in the DublinCore format.
-     * <p/>
+     * This method sets all fields, which are custom in BibTeX and therefore supported by JabRef, but which are not
+     * included in the DublinCore format.
+     * <p>
      * The relation attribute of DublinCore is abused to insert these custom fields.
      */
     private void extractBibTexFields() {
-        List<String> relationships = dcSchema.getRelations();
-
         Predicate<String> isBibTeXElement = s -> s.startsWith("bibtex/");
-
         Consumer<String> splitBibTeXElement = s -> {
             // the default pattern is bibtex/key/value, but some fields contains url etc.
             // so the value property contains additional slashes, which makes the usage of
@@ -154,15 +149,15 @@ private void extractBibTexFields() {
                 // see also DublinCoreExtractor#extractYearAndMonth
                 if (StandardField.MONTH.equals(key)) {
                     Optional<Month> parsedMonth = Month.parse(value);
-                    parsedMonth.ifPresent(month -> bibEntry.setField(key, month.getShortName()));
+                    parsedMonth.ifPresent(bibEntry::setMonth);
                 }
             }
-
         };
+        List<String> relationships = dcSchema.getRelations();
         if (relationships != null) {
             relationships.stream()
-                    .filter(isBibTeXElement)
-                    .forEach(splitBibTeXElement);
+                         .filter(isBibTeXElement)
+                         .forEach(splitBibTeXElement);
         }
     }
 
@@ -220,26 +215,27 @@ private void extractType() {
     }
 
     /**
-     * Helper function for retrieving a BibEntry from the DublinCore metadata
-     * in a PDF file.
+     * Helper function for retrieving a BibEntry from the DublinCore metadata in a PDF file.
+     * <p>
+     * To understand how to get hold of a DublinCore have a look in the test cases for XMPUtil.
+     * <p>
+     * The BibEntry is build by mapping individual fields in the dublin core (like creator, title, subject) to fields in
+     * a bibtex bibEntry. In case special "bibtex/" entries are contained, the normal dublin core fields take
+     * precedence. For instance, the dublin core date takes precedence over bibtex/month.
      *
-     * To understand how to get hold of a DublinCore have a look in the
-     * test cases for XMPUtil.
-     *
-     * The BibEntry is build by mapping individual fields in the dublin core
-     * (like creator, title, subject) to fields in a bibtex bibEntry.
-     *
-     * @return The bibtex bibEntry found in the document information.
+     * @return The bibEntry extracted from the document information.
      */
     public Optional<BibEntry> extractBibtexEntry() {
+        // first extract "bibtex/" entries
+        this.extractBibTexFields();
 
+        // then extract all "standard" dublin core entries
         this.extractEditor();
         this.extractAuthor();
         this.extractYearAndMonth();
         this.extractAbstract();
         this.extractDOI();
         this.extractPublisher();
-        this.extractBibTexFields();
         this.extractRights();
         this.extractSource();
         this.extractSubject();

src/main/java/org/jabref/model/entry/BibEntry.java

@@ -614,7 +614,7 @@ public Object clone() {
      */
     @Override
     public String toString() {
-        return CanonicalBibtexEntry.getCanonicalRepresentation(this);
+        return CanonicalBibEntry.getCanonicalRepresentation(this);
     }
 
     /**

src/main/java/org/jabref/model/entry/CanonicalBibtexEntry.java → src/main/java/org/jabref/model/entry/CanonicalBibEntry.java

@@ -11,16 +11,19 @@
 import org.jabref.model.entry.field.Field;
 import org.jabref.model.entry.field.InternalField;
 
-public class CanonicalBibtexEntry {
+public class CanonicalBibEntry {
 
-    private CanonicalBibtexEntry() {
+    private CanonicalBibEntry() {
     }
 
     /**
-     * This returns a canonical BibTeX serialization. Special characters such as "{" or "&" are NOT escaped, but written
-     * as is
+     * This returns a canonical BibTeX serialization. Serializes all fields, even the JabRef internal ones. Does NOT
+     * serialize "KEY_FIELD" as field, but as key
      *
-     * Serializes all fields, even the JabRef internal ones. Does NOT serialize "KEY_FIELD" as field, but as key
+     * <ul>
+     *     <li>Special characters such as "{" or "&" are NOT escaped, but written as</li>
+     *     <li>String constants are not handled. That means, <code>month = apr</code> in a bib file gets <code>month = {#apr#}</code>. This indicates that the month field is correctly stored</li>
+     * </ul>
      */
     public static String getCanonicalRepresentation(BibEntry entry) {
         StringBuilder sb = new StringBuilder();
@@ -50,7 +53,7 @@ public static String getCanonicalRepresentation(BibEntry entry) {
         // generate field entries
         StringJoiner sj = new StringJoiner(",\n", "", "\n");
         for (String fieldName : sortedFields) {
-            String line = String.format("  %s = {%s}", fieldName, String.valueOf(mapFieldToValue.get(fieldName)).replaceAll("\\r\\n","\n"));
+            String line = String.format("  %s = {%s}", fieldName, String.valueOf(mapFieldToValue.get(fieldName)).replaceAll("\\r\\n", "\n"));
             sj.add(line);
         }
         sb.append(sj);
@@ -59,5 +62,4 @@ public static String getCanonicalRepresentation(BibEntry entry) {
         sb.append('}');
         return sb.toString();
     }
-
 }

src/main/java/org/jabref/model/entry/Month.java

@@ -168,13 +168,14 @@ public String getShortName() {
      * Returns the month in JabRef format. The format is the short 3-digit name surrounded by a '#'.
      * Example: #jan#, #feb#, etc.
      *
-     * See https://github.com/JabRef/jabref/issues/263#issuecomment-151246595 for a discussion on that thing.
-     * This seems to be an <em>invalid</em> format in terms of plain BiBTeX, but a <em>valid</em> format in the case of JabRef
+     * See <a href="https://github.com/JabRef/jabref/issues/263#issuecomment-151246595">Issue 263</a> for a discussion on that thing.
+     * This seems to be an <em>invalid</em> format in terms of plain BiBTeX, but a <em>valid</em> format in the case of JabRef.
+     * The documentation is available at the <a href="http://help.jabref.org/en/Strings">Strings help</a> of JabRef.
      *
      * @return Month in JabRef format
      */
     public String getJabRefFormat() {
-        return String.format("%s", shortName);
+        return String.format("#%s#", shortName);
     }
 
     /**

src/test/java/org/jabref/logic/bibtex/BibEntryWriterTest.java

@@ -416,7 +416,7 @@ void monthFieldSpecialSyntax() throws IOException {
         // modify month field
         Set<Field> fields = entry.getFields();
         assertTrue(fields.contains(StandardField.MONTH));
-        assertEquals("mar", entry.getField(StandardField.MONTH).get());
+        assertEquals("#mar#", entry.getField(StandardField.MONTH).get());
 
         //write out bibtex string
         StringWriter stringWriter = new StringWriter();

src/test/java/org/jabref/logic/cleanup/CleanupWorkerTest.java

@@ -183,7 +183,7 @@ void cleanupMonthChangesNumberToBibtex() {
         entry.setField(StandardField.MONTH, "01");
 
         worker.cleanup(preset, entry);
-        assertEquals(Optional.of("jan"), entry.getField(StandardField.MONTH));
+        assertEquals(Optional.of("#jan#"), entry.getField(StandardField.MONTH));
     }
 
     @Test

src/test/java/org/jabref/logic/cleanup/ConvertToBibtexCleanupTest.java

@@ -27,7 +27,7 @@ public void cleanupMovesDateToYearAndMonth() {
 
         assertEquals(Optional.empty(), entry.getField(StandardField.DATE));
         assertEquals(Optional.of("2011"), entry.getField(StandardField.YEAR));
-        assertEquals(Optional.of("jan"), entry.getField(StandardField.MONTH));
+        assertEquals(Optional.of("#jan#"), entry.getField(StandardField.MONTH));
     }
 
     @Test

src/test/java/org/jabref/logic/formatter/bibtexfields/NormalizeMonthFormatterTest.java

@@ -19,6 +19,6 @@ public void setUp() {
 
     @Test
     public void formatExample() {
-        assertEquals("dec", formatter.format(formatter.getExampleInput()));
+        assertEquals("#dec#", formatter.format(formatter.getExampleInput()));
     }
 }

src/test/java/org/jabref/logic/importer/fileformat/BibtexParserTest.java

@@ -1784,4 +1784,11 @@ void bibTeXConstantAprilIsDisplayedAsConstant() throws ParseException {
 
         assertEquals("#apr#", result.get().getField(StandardField.MONTH).get());
     }
+
+    @Test
+    void bibTeXConstantAprilIsParsedAsStringMonthAprilWhenReadingTheField() throws ParseException {
+        Optional<BibEntry> result = parser.parseSingleEntry("@Misc{m, month = apr }" );
+
+        assertEquals(Optional.of("#apr#"), result.get().getField(StandardField.MONTH));
+    }
 }