This log summarizes the changes in each released version of rouge. The versioning scheme we use is semver, although we will often release new lexers in minor versions, as a practical matter.
https://github.com/jneen/rouge/compare/v3.2.1...v3.3.0
Release Highlight: Due to #883 with the introduction of frozen string literals, Rouge memory usage and total objects dropped quite dramatically. See #883 for more details. Thanks @ashmaroli for this PR.
- General
- Add frozen_string_literal (#883 ashmaroli)
- Mathematica Lexer (NEW)
- Support for Mathematic/Wolfram (#854 by halirutan)
- Motorola 68k Lexer (NEW)
- Add m68k assembly lexer (#909 by nguillaumin)
- SQF Lexer (NEW)
- JSP Lexer (NEW)
- Add Java Server Pages lexer (#915 by miparnisari)
- Elixir Lexer
- Add
defstruct
anddefguardp
(#960 by bjfish)
- Add
- F# / FSharp Lexer
- Add
.fsi
extension (#1002 by adam-becker)
- Add
- Kotlin Lexer
- Objective-C Lexer
- Add
objectivec
as tag/alias (#951 by revolter)
- Add
- Prolog Lexer
- Add % as single-line comment (#898 by jamesnvc)
- Puppet Lexer
- Add = as Operator in Puppet lexer (#980 by alexharv074)
- Python Lexer
- Ruby Lexer
- Add
Dangerfile
as Ruby filename (#1004 by leipert)
- Add
- Rust Lexer
- Add additional aliases for Rust (#988 by LegNeato)
- Swift Lexer
- Add
convenience
method (#950 by damian-rzeszot)
- Add
https://github.com/jneen/rouge/compare/v3.2.0...v3.2.1
- Perl Lexer
- Allow any non-whitespace character to delimit regexes (#974 by dblessing)
- Details: In specific cases where a previously unsupported regex delimiter was used, a later rule could cause a backtrack in the regex system. This resulted in Rouge hanging for an unspecified amount of time.
- Allow any non-whitespace character to delimit regexes (#974 by dblessing)
https://github.com/jneen/rouge/compare/v3.1.1...v3.2.0
- General
- Terraform Lexer (NEW - #917 by lowjoel)
- Crystal Lexer (NEW - #441 by splattael)
- Scheme Lexer
- Allow square brackets (#849 by EFanZh)
- Haskell Lexer
- Support for Quasiquotations (#868 by enolan)
- Java Lexer
- Support for Java 10
var
keyword (#888 by lc-soft)
- Support for Java 10
- VHDL Lexer
- Fix
time_vector
keyword typo (#911 by ttobsen)
- Fix
- Perl Lexer
- Recognize
.t
as valid file extension (#918 by miparnisari)
- Recognize
- Nix Lexer
- Improved escaping sequences for indented strings (#926 by veprbl)
- Fortran Lexer
- Recognize
.f
as valid file extension (#931 by veprbl)
- Recognize
- Igor Pro Lexer
- Update functions and operations for Igor Pro 8 (#921 by t-b)
- Julia Lexer
- Various improvements and fixes (#912 by ararslan)
- Kotlin Lexer
- Recognize
.kts
as valid file extension (#908 by mkobit)
- Recognize
- CSS Lexer
- Minor fixes (#916 by miparnisari)
- HTML Lexer
- Minor fixes (#916 by miparnisari)
- Javascript Lexer
- Minor fixes (#916 by miparnisari)
- Markdown Lexer
- Images may not have alt text (#904 by Himura2la)
- ERB Lexer
- Fix greedy comment matching (#902 by ananace)
https://github.com/jneen/rouge/compare/v3.1.0...v3.1.1
- Perl
- JavaScript
- Swift [Swift] Undo parsing function calls with trailing closure by dan-zheng · Pull Request #862 · jneen/rouge
- Vue
Thanks a lot for contributions; not only for the code, but also for the issues and review comments, which are vitally helpful.
https://github.com/jneen/rouge/compare/v3.0.0...v3.1.0
- gemspec
- Add source code and changelog links to gemspec #785 by @timrogers
- General
- Fix #796: comments not followed by a newline are not highlighted #797 by @tyxchen
- Elem
- Add Elm language support #744 by @dmitryrogozhny
- Ruby
- Add the .erb file extension to ruby highlighting #713 by @jstumbaugh
- Hack
- Add basic Hack support #712 by @fredemmott
- F#
- Allow double backtick F# identifiers #793 by @nickbabcock
- Swift
- Python
- Add async and await keywords to Python lexer #799 by @BigChief45
- Shell
- Add missing shell commands and missing GNU coreutils executables #798 by @kernhanda
- PowerShell
- Add JEA file extensions to powershell #807 by @michaeltlombardi
- SASS / SCSS
- Don't treat
[
as a part of an attribute name in SCSS #839 by @hibariya
- Don't treat
- Haskell
- Don't treat
error
specially in Haskell #834 by @enolan
- Don't treat
- Rust
- Rust: highlight the "where" keyword #823 by @lvillani
https://github.com/jneen/rouge/compare/v2.2.1...v3.0.0
There is no breaking change in the public API, but internals' is changed.
- general:
- dropped support for Ruby 1.9, requireing Ruby v2.0.0 (#775 by gfx)
- [Internal API changes] refactored disaambiguators to removes the use of analyze_text's numeric score interface (#763 by jneen)
- See rouge-ruby#763 for details
- added
rouge guess $file
sub-command to test guessers (#773 by gfx) - added
Rouge::Lexe.guess { fallback }
interface (#777 by gfx) - removes BOM and normalizes newlines in input sources before lexing (#776 by gfx)
- kotlin:
- fix errors in generic functions (#782 by gfx; thanks to @rongi for reporting it)
- haskell:
- fix escapes in char literals (#780 by gfx; thanks to @Tosainu for reporting it)
https://github.com/jneen/rouge/compare/v2.2.0...v2.2.1
- powershell:
- Adding PowerShell builtin commands for version 5 (#757 thanks JacodeWeerd)
- general:
- Rouge::Guessers::Modeline#filter: reduce object allocations (#756 thanks @parkr)
https://github.com/jneen/rouge/compare/v2.1.1...v2.2.0
- rougify:
- trap PIPE only when platform supports it (#700 thanks @maverickwoo)
- support null formatter (
-f tokens
) (#719 thanks @abalkin)
- kotlin:
- update for companion object rename (#702 thanks @stkent)
- igorpro:
- fix igorpro lexer errrors (#706 thanks @ukos-git)
- nix:
- support nix expression language (#732 thanks @vidbina)
- q:
- fix rules for numeric literals (#717 thanks @abalkin)
- fortran:
- add missing Fortran keywords and intrinsics (#739 thanks @pbregener)
- javascript:
- Fix lexer on
<
in<script>...</script>
(#727 thanks @cpallares)
- Fix lexer on
- general:
- speed up
shebang?
check (#738 thanks @schneems) - don't default to a hash in Lexer.format (#729)
- use the token's qualname in null formatter (#730)
- speed up
- formatter:
- fix "unknown formatter: terminal256" (#735 thanks @cuihq)
- gemspec:
- fix licenses to rubygems standard (#714 thanks @nomoon)
- rougify: display help when called with no arguments
- console: bugfix for line continuations dropping characters
- make: properly handle code that doesn't end in a newline
- fix some warnings with -w, add a rubocop configuration
-
javascript:
- fix for newlines in template expressions (#570 thanks @Kovensky!)
- update keywords to ES2017 (#594 thanks @Kovensky!)
- add ES6 binary and octal literals (#619 thanks @beaufortfrancois!)
-
ruby:
- require an
=
on the=end
block comment terminator - bugfix for numeric ranges (#579 thanks @kjcpaas!)
- bugfix for constants following class/module declarations
- require an
-
shell: support based numbers in math mode
-
html-inline formatter:
- now accepts a search string, an instance, or aliases for the theme argument
-
ocaml:
- highlight variant tags
- fixes for operators vs punctuation
- fix polymorphic variants, support local open expressions, fix keywords (#643 thanks @emillon!)
-
thankful-eyes theme:
- bold operators, to distinguish between punctuation
-
rust:
- add support for range operators and type variables (#591 thanks @whitequark!)
- support rustdoc hidden lines that start with # (#652 thanks @seanmonstar!)
-
html: allow template languages to interpolate within script or style tags
-
clojure:
- finally add support for
@
- associate
*.edn
- finally add support for
-
new lexer: lasso (#584 thanks @EricFromCanada!)
-
ruby 1.9 support: Fix unescaped
@
in regex (#588 thanks @jiphex!) -
fix comments at end of files in F# and R (#590 thanks @nickbabcock!)
-
elixir: implement ruby-style sigil strings and regexes (#530 thanks @k3rni!)
-
docker: add missing keywords
-
add ruby 2.4 support
-
coffeescript: bugfix for improper multiline comments (#604 thanks @dblessing!)
-
json: make exponent signs optional (#597 thanks @HerrFolgreich!)
-
terminal256 formatter: put reset characters at the end of each line (#603 thanks @deivid-rodriguez!)
-
csharp:
- actually highlight interface names
- highlight splice literals
$""
and@$""
(#600 thanks @jmarianer!) - recognize
nameof
as a keyword (#626 thanks @drovani!)
-
new lexer: mosel (#606 thanks @germanriano!)
-
php:
- more robust ident regex that supports unicode idents
- add heuristics to determine whether to start inline
-
new lexer: q (kdb+) (#655 thanks @abalkin!)
-
new lexer: pony (#651 thanks @katafrakt!)
-
new lexer: igor-pro (#648 thanks @ukos-git!)
-
new lexer: wollok (#647 thanks @mumuki!)
-
new lexer: graphql (#634 thanks @hibariya!)
-
properties: allow hyphens in key names (#629 thanks @cezariuszmarek!)
-
HTMLPygments formatter: (breaking) wrap tokens with div.highlight
-
new lexer: console (replaces old
shell_session
lexer)- properly detects prompts and continuations
- fully configurable prompt string with
?prompt=...
- optional root-level comments with
?comments
-
new lexer: irb
-
xml: allow newlines in attribute values (#663 thanks @mojavelinux!)
-
windows fix: use
YAML.load_file
to load apache keywords -
new lexer: dot (graphviz) (#627 thanks @kAworu)
-
overhaul options handling, and treat options as untrusted user content
-
add global opt-in to debug mode (
Rouge::Lexer.enable_debug!
) to prevent debug mode from being activated by users -
shell: more strict builtins (don't highlight
cd-hello
) (#684 thanks @alex-felix!) -
new lexer: sieve (#682 thanks @kAworu!)
-
new lexer: TSX (#669 thanks @timothykang!)
-
fortran: update to 2008 (#667 thanks @pbregener!)
-
powershell: use backtick as escape instead of backslash (#660 thanks @gmckeown!)
-
new lexer: awk (#607 thanks @kAworu)
-
new lexer: hylang (#623 thanks @profitware!)
-
new lexer: plist (#622 thanks @segiddins!)
-
groovy: support shebangs (#608 thanks @hwdegroot!)
-
new lexer: pastie (#576 thanks @mojavelinux)
-
sed: bugfix for dropped characters from regexes
-
sml:
- bugfix for dropped keywords
- bugfix for mishighlighted keywords
-
gherkin, php, lua, vim, matlab: update keyword files
-
new lexer: digdag (#674 thanks @gfx!)
-
json-doc: highlight bare keys
-
HTMLTable: don't output a newline after the closing tag (#659 thanks @gpakosz!)
- haml: fix balanced braces in attribute curlies
- clojure:
- allow comments at EOF
- detect for
build.boot
(thanks @pandeiro)
- ruby 1.9.1 compat: escape @ signs (thanks @pille1842)
- c++
- add
*.tpp
as an extension (thanks @vser1) - add more C++11 keywords
- add
- new lexer: ABAP (thanks @mlaggner)
- rougify: properly handle SIGPIPE for downstream pipe closing (thanks @maverickwoo)
- tex: add
*.sty
and*.cls
extensions - html: bugfix for multiple style tags - was too greedy
- new lexer: vue
- perl: fix lexing of POD comments (thanks @kgoess)
- coq: better string escape handling
- javascript:
- add support for ES decorators
- fix multiline template strings with curlies (thanks @Kovensky)
- json: stop guessing based on curlies
- rust: support the
?
operator
- actionscript: emit correct tokens for positive numbers (thanks @JoeRobich!)
- json: bottom-up rewrite, massively improve string performance
- markdown: don't terminate code blocks unless there's a newline before the terminator
- tulip: rewrite lexer with updated features
- swift: update for swift 3 (thanks @radex!)
- fortran: correctly lex exponent floats (thanks @jschwab!)
- bugfix: escape
\@
for ruby 1.9.x - verilog: recognize underscores and question marks (thanks @whitequark!)
- common lisp: recognize .asd files for ASDF
- new lexer: mxml (thanks @JoeRobich!)
- new lexer: 1c (thanks @karnilaev!)
- new lexer: turtle/trig (thanks @jakubklimek!)
- new lexer: vhdl (thanks @ttobsen!)
- new lexer: jsx
- new lexer: prometheus (thanks @dblessing!)
- bugfix: don't spam stdout from the yaml lexer
- new lexer: docker (thanks @KitaitiMakoto!)
- new lexer: fsharp (thanks @raymens!)
- python: improve string escapes (thanks @di!)
- yaml: highlight keys differently than values
- guessing: ambiguous guesses now raise
Rouge::Guesser::Ambiguous
instead of a mysterious class inside a metaclass. - praat: various fixes for unconventional names (thanks @jjatria!)
- workaround for rdoc parsing bug that should fix
gem install
with rdoc parsing on. - ruby:
- best effort colon handling
- fix for heredocs with method calls at the end
- tulip: rewrite from the ground up
- markdown: fix improper greediness of backticks
- tooling: improve the debug output, and properly highlight the legend
- liquid: support variables ending in question marks (thanks @brettg!)
- new lexer: IDL (thanks @sappjw!)
- javascript:
- fix bug causing
:
error tokens (#497) - support for ES6 string interpolation with backticks (thanks @iRath96!)
- fix bug causing
- csharp: allow comments at EOF
- java: allow underscored numeric literals (thanks @vandiedakaf!)
- terminal formatter: theme changes had broken this formatter, this is fixed.
- shell: support "ansi strings" -
$'some-string\n'
- Bugfix for
Formatter#token_lines
without a block
- new formatters! see README.md for documentation, use
Rouge::Formatters::HTMLLegacy
for the old behavior.
- new guesser infrastructure, support for emacs and vim modelines (#489)
- javascript bugfix for nested objects with quoted keys (#496)
- new theme: Gruvbox (thanks @jamietanna!)
- praat: lots of improvements (thanks @jjatria)
- fix for rougify error when highlighting from stdin (#493)
- new lexer: kotlin (thanks @meleyal!)
- new lexer: cfscript (thanks @mjclemente!)
- groovy:
- remove pathological regexes and add basic support for triple-quoted strings (#485)
- add the "trait" keyword and fix project url (thanks @glaforge! #378)
- new lexer: coq (thanks @gmalecha! #389)
- gemspec license now more accurate (thanks @connorshea! #484)
- swift:
- properly support nested comments (thanks @dblessing! #479)
- support swift 2.2 features (thanks @radex #376 and @wokalski #442)
- add indirect declaration (thanks @nRewik! #326)
- new lexer: verilog (thanks @Razer6! #317)
- new lexer: typescript (thanks @Seikho! #400)
- new lexers: jinja and twig (thanks @robin850! #402)
- new lexer: pascal (thanks @alexcu!)
- css: support attribute selectors (thanks @skoji! #426)
- new lexer: shell session (thanks @sio4! #481)
- ruby: add support for <<~ heredocs (thanks @tinci! #362)
- recognize comments at EOF in SQL, Apache, and CMake (thanks @julp! #360)
- new lexer: phtml (thanks @Igloczek #366)
- recognize comments at EOF in CoffeeScript (thanks @rdavila! #370)
- c/c++:
- support c11/c++11 features (thanks @Tosainu! #371)
- Allow underscores in identifiers (thanks @coverify! #333)
- rust: add more builtin types (thanks @RalfJung! #372)
- ini: allow hyphen keys (thanks @KrzysiekJ! #380)
- r: massively improve lexing quality (thanks @klmr! #383)
- c#:
- add missing keywords (thanks @BenVlodgi #384 and @SLaks #447)
- diff: do not require newlines at the ends (thanks @AaronLasseigne! #387)
- new lexer: ceylon (thanks @bjansen! #414)
- new lexer: biml (thanks @japj! #415)
- new lexer: TAP - the test anything protocol (thanks @mblayman! #409)
- rougify bugfix: treat input as utf8 (thanks @japj! #417)
- new lexer: jsonnet (thanks @davidzchen! #420)
- clojure: associate
*.cljc
for cross-platform clojure (thanks @alesguzik! #423) - new lexer: D (thanks @nikibobi! #435)
- new lexer: smarty (thanks @tringenbach! #427)
- apache:
- add directives for v2.4 (thanks @stanhu!)
- various improvements (thanks @julp! #301)
- faster keyword lookups
- fix nil error on unknown directive (cf #246, #300)
- properly manage case-insensitive names (cf #246)
- properly handle windows CRLF
- objective-c:
- support literal dictionaries and block arguments (thanks @BenV! #443 and #444)
- Fix error tokens when defining interfaces (thanks @meleyal! #477)
- new lexer: NASM (thanks @sraboy! #457)
- new lexer: gradle (thanks @nerro! #468)
- new lexer: API Blueprint (thanks @kylef! #261)
- new lexer: ActionScript (thanks @honzabrecka! #241)
- terminal256 formatter: stop confusing token names (thanks @julp! #367)
- new lexer: julia (thanks @mpeteuil! #331)
- new lexer: cmake (thanks @julp! #302)
- new lexer: eiffel (thanks @Conaclos! #323)
- new lexer: protobuf (thanks @fqqb! #327)
- new lexer: fortran (thanks @CruzR! #328)
- php: associate
*.phpt
files (thanks @Razer6!) - python: support
raise from
andyield from
(thanks @mordervomubel! #324) - new VimL example (thanks @tpope! #315)
- diff: fix deleted lines which were not being highlighted (thanks @DouweM)
- fix warnings on files being loaded multiple times
- swift: (thanks @radex)
- new keywords
- support all
@
-prefixed attributes - add support for
try!
and#available(...)
- bugfix: Properly manage
#style_for
precedence for terminal and inline formatters (thanks @mojavelinux) - visual basic: recognize
*.vb
files (thanks @naotaco) - common-lisp:
- add
elisp
as an alias (todo: make a real elisp lexer) (thanks @tejasbubane) - bugfix: fix crash on array and structure literals
- add
- new lexer: praat (thanks @jjatria)
- rust: stop recognizing
*.rc
(thanks @maximd) - matlab: correctly highlight
'
(thanks @miicha)
- new lexer: powershell (thanks @aaroneg!)
- new lexer: tulip
- bugfix: pass opts through so lex(continue: true) retains them (thanks @stanhu!)
- c#: bugfix: don't error on unknown states in the C# lexer
- php: match drupal file extensions (thanks @rumpelsepp!)
- prolog: allow camelCase atoms (thanks @mumuki!)
- c: bugfix: was dropping text in function declarations (thanks @JonathonReinhart!)
- groovy: bugfix: allow comments at eof without newline
-
objc: add array literals (thanks @mehowte)
-
slim: reset ruby and html lexers, be less eager with guessing, detect html entities (thanks @elstgav)
-
js: add
yield
as a keyword (thanks @honzabrecka) -
elixir: add alias
exs
(thanks @ismaelga) -
json: lex object keys as
Name::Tag
(thanks @morganjbruce) -
swift: add support for
@noescape
and@autoclosure(escaping)
(thanks @radex) and makeas?
andas!
look better -
sass/scss: add support for
@each
,@return
,@media
, and@function
(thanks @i-like-robots) -
diff: make the whole thing more forgiving and less buggy (thanks @rumpelsepp)
-
c++: add arduino file mappings and also Berksfile (thanks @Razer6)
-
liquid: fix #237 which was dropping content (thanks @RadLikeWhoa)
-
json: add json-api mime type (thanks @brettchalupa)
-
new lexer: glsl (thanks @sriharshachilakapati)
-
new lexer: json-doc, which is like JSON but supports comments and ellipsis (thanks @textshell)
-
add documentation to the
--formatter
option inrougify help
(thanks @mjbshaw) -
new website! http://rouge.jneen.net/ (thanks @edwardloveall!)
- css: fix "empty range in char class" bug and improve id/class name matches (#227/#228). Thanks @elstgav!
- swift: add
@objc_block
and fix eof comments (#226). Thanks @radex! - new lexer: liquid (#224). Thanks @RadLikeWhoa!
- cli: add
-v
flag to print version (#225). Thanks @RadLikeWhoa! - ruby: add
alias_method
as a builtin (#223). Thanks @kochd! - more conservative guessing for R, eliminate the
.S
extension - new theme: molokai (#220). Thanks @kochd!
- allow literate haskell that doesn't end in eof
- add human-readable "title" attribute to lexers (#215). Thanks @edwardloveall!
- swift: add support for preprocessor macros (#201). Thanks @mipsitan!
- fix previous bad release: actually add yaml files to the gem
lexer fixes and tweaks:
- javascript: fix function literals in object literals (reported by @taye)
- css: fix for percentage values and add more units (thanks @taye)
- ruby: highlight
require_relative
as a builtin (thanks @NARKOZ)
new lexers:
- nim (thanks @singularperturbation)
- apache (thanks @iiska)
new filetype associations:
- highlight PKGBUILD as shell (thanks @rumpelsepp)
- highlight Podspec files as ruby (thanks @NARKOZ)
other:
- lots of doc work in the README (thanks @rumpelsepp)
- clojure: hotfix for namespaced keywords with
::
- background fix: add css class to pre tag instead of code tag (#191)
- new name in readme and license
- new contributor code of conduct
- ruby: hotfix for symbols in method calling position (rubyyyyy.......)
- http: add PATCH as a verb
- new lexer: Dart (thanks @R1der!)
- null formatter now prints token names and values
- ruby: hotfix for division with no space
- ruby: hotfix for the
/=
operator
- ruby: give up on trying to highlight capitalized builtin methods
- swift: updates for beta 6 (thanks @radex!) (#174, #172)
- support ASCII-8BIT encoding found on macs, as it's a subset of UTF-8 (#178)
- redcarpet plugin [BREAKING]: change
#rouge_formatter
's override pattern- it is now a method that takes a lexer and returns a formatter, instead of a hash of generated options. (thanks @vince-styling!)
- java: stop erroneously highlighting keywords within words (thanks @koron!) (#177)
- html: dash is allowed in tag names (thanks @tjgrathwell!) (#173)
- swift: updates for beta 5 (thanks @radex!)
- hotfix release for common lisp, php, objective c, and qml lexers
- haml: balance braces in interpolation
- new lexer: slim (thanks @knutaldrin and @greggroth!)
- javascript: inner tokens in regexes are now lexed, as well as improvments to the block / object distinction.
- ruby bugfixes for symbol edgecases and one-letter constants
- utf-8 all of the things
- update all builtins
- rust: add
box
keyword and associated builtins
- new lexer: swift (thanks @totocaster!)
- update elixir for new upstream features (thanks @splattael!)
- ruby bugfixes:
- add support for method calls with trailing dots
- fix for
foo[bar] / baz
being highlighted as a regex
- terminal256 formatter: re-style each line - some platforms reset on each line
- breaking: wrap code in
<pre ...><code>...</code></pre>
if:wrap
is not overridden (thanks @Arcovion) - Allow passing a theme name as a string to
:inline_theme
(thanks @Arcovion) - Add
:start_line
option for html line numbers (thanks @sencer) - List available themes in
rougify help style
- New lexers:
- QML (thanks @seanchas116)
- Applescript (thanks @joshhepworth)
- Properties (thanks @pkuczynski)
- Ruby bugfix for
{ key: /regex/ }
(#134) - JSON bugfix: properly highlight null (thanks @zsalzbank)
- Implement a noop formatter for perf testing (thanks @splattael)
- prolog bugfix: was raising an error on some inputs (#126)
- python bugfix: was inconsistently highlighting keywords/builtins mid-word (#127)
- html formatter: always end output with a newline (#125)
- Now tested in Ruby 2.1
- C family bugfix: allow exponential floats without decimals (
1e-2
) - cpp: allow single quotes as digit separators (
100'000'000
) - ruby: highlight
%=
as an operator in the right context
- fill in some lexer descriptions and add the behat alias for gherkin
- assorted CLI bugfixes: better error handling, CGI-style options, no loadpath munging
- html: support multiline doctypes
- ocaml: bugfix for OO code: allows
#
as an operator - inline some styles in tableized output instead of relying on the theme
- redcarpet: add overrideable
#rouge_formatter
for custom formatting options
- New lexers:
- MATLAB (thanks @adambard!)
- Scala (thanks @sgrif!)
- Standard ML (sml)
- OCaml
- Major performance overhaul, now ~2x faster (see #114) (thanks @korny!)
- Deprecate
RegexLexer#group
(internal). Use#groups
instead. - Updated PHP builtins
- CLI now responds to
rougify --version
- For tableized line numbers, the table is no longer surrounded by a
<pre>
tag, which is invalid HTML. This was previously causing issues with HTML post-processors such as loofah. This may break some stylesheets, as it changes the generated markup, but stylesheets only referring to the scope passed to the formatter should be unaffected. - New lexer: moonscript (thanks @nilnor!)
- New theme: monokai, for real this time! (thanks @3100!)
- Fix intermittent loading errors for good with
Lexer.load_const
, which closes the long-standing #66
- lua: encoding bugfix, and a performance tweak for string literals
- The Big 1.0! From now on, strict semver will apply, and new lexers and features will be introduced in minor releases, reserving patch releases for bugfixes.
- Cleaned up stray invalid error tokens
- Fix C++/objc loading bug in
rougify
- Guessing alg tweaks: don't give up if no filename or mimetype matches
- Rebuilt the CLI without thor (removed the thor dependency)
- objc: Bugfix for
:forward_classname
error tokens
- Critical bugfixes (#98 and #99) for Ruby and Markdown. Some inputs would throw errors. (thanks @hrysd!)
- Bugfixes for C/C++
- Major bugfix: YAML was in a broken state :\ (thanks @hrysd!)
- Implement lexer subclassing, with
append
andprepend
- new lexer: objective c (!)
- Fix non-default themes (thanks @tiroc!)
- Minor lexing bugfixes in ruby
- Various performance optimizations
- javascript:
- quoted object keys were not being highlighted correctly
- multiline comments were not being highlighted
- common lisp: fix commented forms
- golang: performance bump
- ruby: fix edge case for
def-@
- c: fix a pathological performance case
- fix line number alignment on non-newline-terminated code (#91)
Rouge::Lexers::Text
renamed toRouge::Lexers::PlainText
- Tokens are now constants, rather than strings. This only affects you if you've written a custom lexer, formatter, or theme.
- Add the
:inline_theme
option toFormatters::HTML
for environments that don't support stylesheets (like super-old email clients) - Improve documentation of
Formatters::HTML
options - bugfix: don't include subsequent whitespace in an elixir keyword. In certain fonts/themes, this can cause inconsistent indentation if bold spaces are wider than non-bold spaces. (thanks @splattael!)
- Add the
license
key in the gemspec - new lexer: R
- new lexers:
- elixir (thanks @splattael!)
- racket (thanks @greghendershott!)
- new lexers:
- erlang! (thanks @potomak!)
- http (with content-type based delegation)
- bugfix: highlight true and false in JSON
- bugfix: Add the local lib dir to the path in ./bin/rougify
so the internal
require
works properly. - php: Properly lex variables in double-quoted strings and provide the correct token for heredocs (thanks @hrysd!)
- Add a
:wrap
option to the html formatter (default true) to provide the<pre>
wrapper. This allows skipping the wrapper entirely for postprocessing. (thanks @cjohansen!)
- fixed bad release that included unfinished D and wdiff lexers :\
- Added a github theme (thanks @simonc!) (#75)
- Correctly highlight ruby 1.9-style symbols and %i() syntax (thanks @simonc!) (#74)
- Fixed a performance bug in the C++ lexer (#73) reported by @jeffgran
- New lexer: go (thanks @hashmal!)
- Clojure bugfix: allow # in keywords and symbols
- Basic prompt support in the shell lexer
- Add CSS3 attributes to CSS/Sass/SCSS lexers
- Bugfix for a crash in the vim lexer
- Another hotfix release for the Sass/SCSS lexers, because I am being dumb
- Hotfix release: fix errors loading the SCSS lexer on some systems.
- Refactor source guessing to return fewer false positives, and
to be better at disambiguating between filename matches (such as
nginx.conf
vs.*.conf
, or*.pl
for both prolog and perl) - Added
Lexer.guesses
which can return multiple or zero results for a guess. - Fix number literals in C#
- New lexers:
- Gherkin (cucumber)
- Prolog (@coffeejunk)
- LLVM (@coffeejunk)
- New lexer: lua (thanks, @nathany!)
- Add extra filetypes that map to Ruby (
Capfile
,Vagrantfile
,*.ru
and*.prawn
) (@nathany) - Bugfix: add demos for ini and toml
- The
thankful_eyes
theme now colorsLiteral.Date
- No more gigantic load list in
lib/rouge.rb
- New lexers:
- puppet
- literate coffeescript
- literate haskell
- ini
- toml (@coffeejunk)
- clojure:
cljs
alias, and make it more visually balanced by usingName
instead ofName.Variable
. - Stop trying to read /etc/bash.bashrc in the specs (@coffeejunk)
- Highlight ClojureScipt files (
*.cljs
) as Clojure (@blom) - README and doc enhancements (plus an actual wiki!) (@robin850)
- Don't open
Regexp
, especially if we're not adding anything to it.
- Python: bugfix for lone quotes in triple-quoted strings
- Ruby: bugfix for
#
in%
-delimited strings
- New lexer: C# (csharp)
- rust: better macro handling
- Python bugfix for "'" and '"' (@garybernhardt)
- New lexer: rust (rust-lang.org)
- Include rouge.gemspec with the built gem
- Update the PHP builtins
- New lexers: io, sed, conf, and nginx
- fixed an error on numbers in the shell lexer
- performance bumps for shell and ruby by prioritizing more common patterns
- (@korny) Future-proofed the regexes in the Perl lexer
rougify
now streams the formatted text to stdout as it's available instead of waiting for the lex to be done.
- Bugfix for tableized line numbers when the code doesn't end with a newline.
- Major performance improvements. 80% running time reduction for some files since v0.2.5 (thanks again @korny!)
- Deprecated
postprocess
for performance reasons - it wasn't that useful in the first place. - The shell lexer should now recognize .bashrc, .profile and friends
- coffeescript: don't yield error tokens for keywords as attributes
- add the
--scope=SELECTOR
option torougify style
- Add the
:line_numbers
option to the HTML formatter to get line numbers! The styling for the line numbers is determined by the theme's styling for'Generic.Lineno'
- Massive performance improvements by reducing calls to
option
and toRegexp#source
(@korny)
- hotfix: ship the demos with the gem.
- Several improvements to the javasript and scheme lexers
- Lexer.demo, with small demos for each lexer
- Rouge.highlight takes a string for the formatter
- Formatter.format delegates to the instance
- sass: Support the @extend syntax, fix new-style attributes, and support 3.2 placeholder syntax
- Fixed several postprocessing-related bugs
- New lexers: coffeescript, sass, smalltalk, handlebars/mustache
- In terminal256, stop highlighting backgrounds of text-like tokens
- Fix a bug which was breaking guessing with filenames beginning with .
- Fix the require path for redcarpet in the README (@JustinCampbell)
- New lexers: clojure, groovy, sass, scss
- YAML: detect files with the %YAML directive
- Fail fast for non-UTF-8 strings
- Formatter#render deprecated, renamed to Formatter#format. To be removed in v0.3.
- Lexer#tag delegates to the class
- Better keyword/builtin highlighting for CSS
- Add the
:token
option to the text lexer
- Began the changelog
- Removed several unused methods and features from Lexer and RegexLexer
- Added a lexer for SQL
- Added a lexer for VimL, along with
rake builtins:vim
- Added documentation for RegexLexer, TextAnalyzer, and the formatters
- Refactored
rake phpbuiltins
- renamed torake builtins:php
- Fixed a major bug in the Ruby lexer that prevented highlighting the
module
keyword. - Changed the default formatter for the
rougify
executable toterminal256
. - Implemented
rougify list
, and added short descriptions to all of the lexers. - Fixed a bug in the C lexer that was yielding error tokens in case statements.