View Source Changelog

1.2.1 - 2024-11-07

Allow empty middle on string_like combinators.

1.2.0 - 2024-11-06

Make reparsing at end of string configurable.

1.1.2 - 2024-05-11

Do not attempt reparsing if surrounded by reaches end of string.

1.1.1 - 2023-11-08

Add dracula theme.

Add one dark theme.

Fix warnings on recent Elixir versions.

1.1.0 - 2022-02-12

Require NimbleParsec ~> 1.2.2 onwards to prepare for future deprecations.

1.0.5 - 2020-10-02

Bugfix #1

Fix serious bug in the Makeup.Lexer.Combinators.lexeme/1 when given a list of unicode characters.

Before this fix, the following combinator:

import Makeup.Lexer.Combinators

characters_lexeme =
  utf8_char([])
  |> utf8_char([])
  |> lexeme()
  |> token(:character_lexeme)

when given as input a string like "àó" would return the following invalid Unicode:

[{:character_lexeme, %{}, <<225, 242>>}]

instead of

[{:character_lexeme, %{}, "áò"}]

This was caused by the use of IO.iodata_to_binary/1 instead of (the slower) to_string(). This was a problem because IO.iodata_to_binary/1 would put any byte in a binary instead of (correctly) encoding bytes > 128 as a Unicode character. This is not a problem with IO.iodata_to_binary/1 it's a problem with using that function when we want to encode characters as Unicode strings.

Bugfix #2

Fixed an edge case in the HTML encoder in the formatter.

1.0.4 - 2020-09-25

Fix warnings and update NimbleParsec dependency.

1.0.3 - 2020-06-07

Allow styles to be given as atoms when generating stylesheets.

1.0.2 - 2020-05-28

Update NimbleParsec dependency.

1.0.1 - 2020-03-24

Remove warnings on recent Elixir versions.

1.0.0 - 2019-07-15

Upgrade the HTML formatter so that you can use different kinds of tags around the tokens.

Previous versions always used the <span> tag (e.g. <span class="n">name</span>). You can now use other HTML tags using the :highlight_tag option:

alias Makeup.Formatters.HTML.HTMLFormatter
HTMLFormatter.format_as_iolist(tokens, highlight_tag: "font")

0.8.0 - 2019-01-01

This release adds a new file extension registry for lexers. This means that it's now possible to lookup lexers by file extension. Since the previous release it was already possible to lookup lexers by language name. For example:

elixir_lexer = Makeup.Registry.fetch_lexer_by_extension!("elixir")

Now you can also do this this:

elixir_lexer = Makeup.Registry.get_lexer_by_extension!("ex")

Documentation of the registry functionality was also improved.

0.7.0 - 2018-12-30

Adds a register where lexer developers can register their lexers. This allows one to pick a lexer based on the language name.

The goal is for Makeup to be aware of the available lexers. In a project such as ExDoc this allows Makeup to support an unbounded number of languages just by declaring the lexer as a dependency.

0.6.0 - 2018-12-22

Fixes the combinators affected by compatibility breaks in nimble_parsec from 0.4.x to 0.5.x.

Pins nimble_parsec to version 0.5.0.