Question regarding mistake in the grammar

[andraantariksa]

There are some flaw on the library that are not following the TOML grammar rule. I have add some test from the TOML example and it's failed

# Bare keys may only contain ASCII letters, ASCII digits, underscores, and dashes (A-Za-z0-9_-). Note that bare keys are allowed to be composed of only ASCII digits, e.g. 1234, but are always interpreted as strings.
it_parses "123abc_- = 123", { "123abc_-" => 123 }
it_parses "-_123abc = 123", { "-_123abc" => 123 }

# Since v0.5.0
# Dotted keys are a sequence of bare or quoted keys joined with a dot. This allows for grouping similar properties together:
it_parses "physical.color = \"orange\"", { "physical" => { "color" => "orange" } }
it_parses "site.\"google.com\" = true", { "site" => { "google.com" => true } }

it_parses %(
    [dog."tater.man"]
    type.name = "pug"
    ),
    {"dog" => {"tater" => {"man" => {"type" => {"name" => "pug"}}}} }

It appears that the lexer read 123abc_- as 2 token, 123 as a number and abc_
as a key. And for the -_123abc, because the key started with character -, the lexer then expect the token as a negative number. I'd like to make a pull request to fix this issue.

Can you give me some pointers to fix this? I'm thinking to recreate some parts of the lexer and parser.