Added support for all special and Unicode characters and checking for duplicate key groups.

README.md

@@ -61,7 +61,7 @@ A Toml object can be mapped to a custom class with the `Toml#to(Class<T>)` metho
 
 Any keys not found in both the TOML and the class are ignored.
 
-Key groups can be mapped to other custom classes or to `Map<String, Object>`.
+Key groups can be mapped to other custom classes.
 
 ````
 name = "Mwanji Ezana"
@@ -92,8 +92,8 @@ When defaults are present, a shallow merge is performed.
 
 ## TODO
 
-* Support all special characters
-* Convert Toml to custom class
+* Fail on invalid definitions
+* Support negative numbers
 
 ## License
 

src/main/java/com/moandjiezana/toml/TomlParser.java

@@ -4,7 +4,9 @@ import java.text.SimpleDateFormat;
 import java.util.ArrayList;
 import java.util.Date;
 import java.util.HashMap;
+import java.util.HashSet;
 import java.util.Map;
+import java.util.Set;
 
 import org.parboiled.BaseParser;
 import org.parboiled.Rule;
@@ -16,6 +18,7 @@ class TomlParser extends BaseParser<Object> {
 
   static class Results {
     public Map<String, Object> values = new HashMap<String, Object>();
+    public Set<String> keyGroups = new HashSet<String>();
     public StringBuilder errors = new StringBuilder();
   }
 
@@ -24,7 +27,25 @@ class TomlParser extends BaseParser<Object> {
   }
 
   Rule KeyGroup() {
-    return Sequence(KeyGroupDelimiter(), KeyGroupName(), addKeyGroup((String) pop()), KeyGroupDelimiter(), Spacing());
+    return Sequence(Sequence(KeyGroupDelimiter(), KeyGroupName(), addKeyGroup((String) pop()), KeyGroupDelimiter(), Spacing()), checkKeyGroup(match()));
+  }
+
+  Rule IllegalCharacters() {
+    return ZeroOrMore(TestNot(NewLine()), ANY);
+  }
+
+  boolean checkKeyGroup(String definition) {
+    String afterBracket = definition.substring(definition.indexOf(']') + 1);
+    for (char character : afterBracket.toCharArray()) {
+      if (character == '#') {
+        return true;
+      }
+
+      if (!Character.isWhitespace(character)) {
+        results().errors.append("Invalid key group definition: ").append(definition).append(". You may have forgotten a #");
+      }
+    }
+    return true;
   }
 
   Rule Key() {
@@ -60,7 +81,7 @@ class TomlParser extends BaseParser<Object> {
   }
 
   Rule StringValue() {
-    return Sequence(push(new StringBuilder()), '"', OneOrMore(TestNot('"'), FirstOf(SpecialCharacter(), AnyCharacter())), pushString(((StringBuilder) pop()).toString()), '"');
+    return Sequence(push(new StringBuilder()), '"', OneOrMore(TestNot('"'), FirstOf(UnicodeCharacter(), SpecialCharacter(), AnyCharacter())), pushString(((StringBuilder) pop()).toString()), '"');
   }
 
   Rule Year() {
@@ -83,6 +104,10 @@ class TomlParser extends BaseParser<Object> {
     return CharRange('a', 'z');
   }
 
+  Rule UnicodeCharacter() {
+    return Sequence(Sequence('\\', 'u', OneOrMore(FirstOf(CharRange('0', '9'), CharRange('A', 'F')))), pushCharacter(match()));
+  }
+
   Rule SpecialCharacter() {
     return Sequence(Sequence('\\', ANY), pushCharacter(match()));
   }
@@ -106,6 +131,11 @@ class TomlParser extends BaseParser<Object> {
     return ZeroOrMore(FirstOf(Comment(), AnyOf(" \t\r\n\f")));
   }
 
+  @SuppressNode
+  Rule NewLine() {
+    return AnyOf("\r\n");
+  }
+
   @SuppressNode
   Rule ArrayDelimiter() {
     return Sequence(Spacing(), ',', Spacing());
@@ -126,13 +156,20 @@ class TomlParser extends BaseParser<Object> {
     }
 
     Map<String, Object> newKeyGroup = (Map<String, Object>) getContext().getValueStack().peek();
+
+    if (!results().keyGroups.add(name)) {
+      results().errors.append("Could not create key group ").append(name).append(": key group already exists!\n");
+
+      return true;
+    }
+
     for (String splitKey : split) {
       if (!newKeyGroup.containsKey(splitKey)) {
         newKeyGroup.put(splitKey, new HashMap<String, Object>());
       }
       Object currentValue = newKeyGroup.get(splitKey);
       if (!(currentValue instanceof Map)) {
-        results().errors.append("Could not create key group ").append(name).append(": key already exists!\n");
+        results().errors.append("Could not create key group ").append(name).append(": key already has a value!\n");
 
         return true;
       }
@@ -140,6 +177,7 @@ class TomlParser extends BaseParser<Object> {
     }
 
     push(newKeyGroup);
+
     return true;
   }
 
@@ -213,6 +251,14 @@ class TomlParser extends BaseParser<Object> {
       sb.append('\r');
     } else if (sc.equals("\\\\")) {
       sb.append('\\');
+    } else if (sc.equals("\\/")) {
+      sb.append('/');
+    } else if (sc.equals("\\b")) {
+      sb.append('\b');
+    } else if (sc.equals("\\f")) {
+      sb.append('\f');
+    } else if (sc.startsWith("\\u")) {
+      sb.append(Character.toChars(Integer.parseInt(sc.substring(2), 16)));
     } else if (sc.startsWith("\\")) {
       results().errors.append(sc + " is a reserved special character and cannot be used!\n");
     } else {

src/test/java/com/moandjiezana/toml/TomlTest.java

@@ -10,6 +10,7 @@ import java.io.File;
 import java.util.Calendar;
 import java.util.TimeZone;
 
+import org.junit.Ignore;
 import org.junit.Test;
 
 public class TomlTest {
@@ -99,6 +100,14 @@ public class TomlTest {
     assertEquals("value", toml.getKeyGroup("group.sub").getString("key"));
   }
 
+  @Test
+  public void should_get_value_from_key_group_with_sub_key_group() throws Exception {
+    Toml toml = new Toml().parse("[a.b]\nc=1\n[a]\nd=2");
+
+    assertEquals(2, toml.getLong("a.d").intValue());
+    assertEquals(1, toml.getKeyGroup("a.b").getLong("c").intValue());
+  }
+
   @Test
   public void should_return_null_if_no_value_for_key() throws Exception {
     Toml toml = new Toml().parse("");
@@ -173,7 +182,14 @@ public class TomlTest {
   public void should_support_special_characters_in_strings() {
     Toml toml = new Toml().parse(new File(getClass().getResource("should_support_special_characters_in_strings.toml").getFile()));
 
-    assertEquals("\" \t \n \r \\", toml.getString("key"));
+    assertEquals("\" \t \n \r \\ / \b \f", toml.getString("key"));
+  }
+
+  @Test
+  public void should_support_unicode_characters_in_strings() throws Exception {
+    Toml toml = new Toml().parse("key=\"\\u00B1\"");
+
+    assertEquals("±", toml.getString("key"));
   }
 
   @Test(expected = IllegalStateException.class)
@@ -200,11 +216,21 @@ public class TomlTest {
 
   @Test(expected = IllegalStateException.class)
   public void should_fail_when_key_is_overwritten_by_key_group() {
-    new Toml().parse("[fruit]\ntype=\"apple\"\n[fruit.type]\napple=\"yes\"");
+    new Toml().parse("[a]\nb=1\n[a.b]\nc=2");
   }
 
   @Test(expected = IllegalStateException.class)
   public void should_fail_when_key_is_overwritten_by_another_key() {
     new Toml().parse("[fruit]\ntype=\"apple\"\ntype=\"orange\"");
   }
+
+  @Test(expected = IllegalStateException.class)
+  public void should_fail_when_key_group_defined_twice() throws Exception {
+    new Toml().parse("[a]\nb=1\n[a]\nc=2");
+  }
+
+  @Ignore @Test(expected = IllegalStateException.class)
+  public void should_fail_when_illegal_characters_after_key_group() throws Exception {
+    new Toml().parse("[error]   if you didn't catch this, your parser is broken");
+  }
 }

src/test/resources/com/moandjiezana/toml/hard_example.toml

@@ -19,3 +19,15 @@ test_string = "You'll hate me after this - #"          # " Annoying, isn't it?
             "]",
             # ] Oh yes I did
             ]
+
+# Each of the following keygroups/key value pairs should produce an error. Uncomment to them to test
+
+[error]   if you didn't catch this, your parser is broken
+#string = "Anything other than tabs, spaces and newline after a keygroup or key value pair has ended should produce an error unless it is a comment"   like this
+#array = [
+#         "This might most likely happen in multiline arrays",
+#         Like here,
+#         "or here,
+#         and here"
+#         ]     End of array comment, forgot the #
+#number = 3.14  pi <--again forgot the #         

src/test/resources/com/moandjiezana/toml/should_support_special_characters_in_strings.toml

@@ -1 +1 @@
-key = "\" \t \n \r \\"
+key = "\" \t \n \r \\ \/ \b \f"