agrison / jtoml Goto Github PK

The TOML GitHub page has an example of arrays of tables (which may or may not be nested) that jtoml seems not to support. The example is:

foo = "bar"

[[fruit]]
  name = "apple"

  [[fruit.variety]]
    name = "red delicious"
  [[fruit.variety]]
    name = "granny smith"

[[fruit]]
  name = "banana"

  [[fruit.variety]]
    name = "plantain"

baz = "quux"

Note: I added the foo and baz sections to check that something was parsing.

When I try this example in jtoml, it returns nothing for the key "fruit".

File f = new File("fruit.toml");
Toml t = Toml.parse(f);
System.out.println(t.getString("foo")); // prints bar
System.out.println(t.getString("baz")); // prints null
Object object = t.get("fruit");
System.out.println(object);             // prints null

Request: jtoml should support this.

Note: I used version 1.1.0 for this test.

Also, notice that the getString("baz") returns null. If I put the baz="quux" at the top of the file, it returns "quux", so it looks like the jtoml parser just stops when it hits a part of the file it considers an error and stops parsing. This seems like a defect in the parser. In my view, it should throw an exception if the TOML file is invalid. Perhaps I should file another issue?

Since the Getter throws if the value isn't convertible to a boolean, the interface should return lower-case boolean instead of nullable upper-case Boolean.

This does not work:
Toml t = Toml.parse("a = -3");
 System.out.println(t.serialize());

but if the key is more than one char it works, like this:
Toml t = Toml.parse("ab = -3.1");
 System.out.println(t.serialize());

That might be a stupid question, but from the source it seems that Toml.parse is only susceptible to throw a FileNotFoundException, not an IOException.

I understand that the later is more general than the former, but why not throwing the most precise kind?

JetBrains "IntelliJ IDEA Community 2002.2" found plenty of legacy issues, even for Java 1.6 ...hmm...
I would ignore it's hints about Regex because they looked wrong to me.

I prefer no earlier than Java 1.8 now, so even more is possible with that.

The JavaDoc build step had HTML/Javadoc warning and errors, which stopped install...
e.g. > is a reserved character in HTML, so must be added as HTML entity >.

Hey, I just started to work on an ANTLR version of jtoml, mostly for educational purposes to learn ANTLR. It's my first hands-on with a grammar, so I expect to make a lot of dumb mistakes, but maybe you're interested in my current code: https://github.com/MoriTanosuke/jtoml/tree/antlr

I try to push the branch to github whenever I have a major change in the grammar or the application code.

Caused by: java.lang.NullPointerException
	at me.grison.jtoml.impl.Toml.get(Toml.java:184)
	at me.grison.jtoml.impl.Toml.get(Toml.java:262)
	at me.grison.jtoml.impl.Toml.getAs(Toml.java:241)

Exceptions should not be used for flow control, so if the key is missing, null should be returned.
I'll have to see if I can detect the absence of just the key, as a workaround for this bug, until it's fixed.

The spec says: "64-bit minimum size expected".

The TOML spec says of floats, 64-bit (double) precision expected. Thus, the Getter should return Doubles, not Floats.

While trying to figure out some loops occuring, see other issue I found out that if you add a map to the Test Pojo the creation of the key is wrong.

public static class Foo {

    String stringKey;
    Long longKey;
    Double doubleKey;
    Boolean booleanKey;
    List<Object> listKey;
    Bar bar;
    Boolean awesome;
    Map<String, Integer> map = new HashMap<String, Integer>() {{
        put("one",1);
        put("two",2);
    }};

This will create a key with [foo.bar.map] instead of [foo.map]

The parser will fail for a file with the following content.
title = "anything é "

The problem is caused by the function Util.FileToString.read(File file) that does not support accented characters.

I fixed using the Apache CommonIO FileUtils.readFileToString but this function can solve the problem:

public static class FileToString {
        public static String read(File file)
          throws FileNotFoundException {
            String result = "";
            DataInputStream in = null;
            try {

            byte[] buffer = new byte[(int) file.length()];
            in = new DataInputStream(new FileInputStream(file));
            in.readFully(buffer);
            result = new String(buffer);
            } catch (IOException ex) {
                throw new FileNotFoundException(ex.getMessage());
            }
            return result;
        }
}

cheers

TOML_SUPPORTED should be:

    /**
     * List of types supported Natively by TOML + Map
     */
    static final Set<Class<?>> TOML_SUPPORTED = new HashSet<Class<?>>(Arrays.asList(Integer.class, Long.class, Double.class, //
            Calendar.class, Boolean.class, String.class, List.class, Map.class));

It seems that something's wrong with the deploy procedure. Please check the artifact currently at the Maven central.

If you pass a String directly into Toml.serialize with a root key it will cause a loop.

Here a test log.

20:23:05] [INFO] [TCBase] [DEBUG] buffer [[test]
]
[20:23:05] [INFO] [TCBase] [DEBUG] fields [7]
[20:23:05] [INFO] [TCBase] [DEBUG] field [value]
[20:23:05] [INFO] [TCBase] [DEBUG] type [[C]
[20:23:05] [INFO] [TCBase] [DEBUG] value [[C@7cbc11d]
[20:23:05] [INFO] [TCBase] [DEBUG] buffer [[test.value]
]
[20:23:05] [INFO] [TCBase] [DEBUG] fields [0]
[20:23:05] [INFO] [TCBase] [DEBUG] field [hash]
[20:23:05] [INFO] [TCBase] [DEBUG] type [int]
[20:23:05] [INFO] [TCBase] [DEBUG] value [0]
[20:23:05] [INFO] [TCBase] [DEBUG] buffer [[test.value.hash]
]
[20:23:05] [INFO] [TCBase] [DEBUG] fields [11]
[20:23:05] [INFO] [TCBase] [DEBUG] field [MIN_VALUE]
[20:23:05] [INFO] [TCBase] [DEBUG] type [int]
[20:23:05] [INFO] [TCBase] [DEBUG] value [-2147483648]
[20:23:05] [INFO] [TCBase] [DEBUG] buffer [[test.value.hash.MIN_VALUE]

this does not work:
Toml t = Toml.parse("[a]\ncd = -3");
 System.out.println(t.serialize());
but this does:
Toml t = Toml.parse("[a]\ncd = -3.1");
 System.out.println(t.serialize());