Issue numbers refer to the tracker at http://dev.pocoo.org/projects/pygments/.
(bugfix release, released Sep 15, 2009)
- Fixed the BBCode lexer (#435).
- Added support for new Jinja2 keywords.
- Fixed test suite failures.
- Added Gentoo-specific suffixes to Bash lexer.
(codename Brillouin, released Sep 11, 2009)
- Ported Pygments to Python 3. This needed a few changes in the way
encodings are handled; they may affect corner cases when used with
Python 2 as well.
- Lexers added:
* Antlr/Ragel, thanks to Ana Nelson
* (Ba)sh shell
* Erlang shell
* ASP.net (VB/C#)
- Fixed the LaTeX formatter's output so that output generated for one style
can be used with the style definitions of another (#384).
- Added "anchorlinenos" and "noclobber_cssfile" (#396) options to HTML
- Support multiline strings in Lua lexer.
expression literals (#403).
- When pygmentize is asked to highlight a file for which multiple lexers
match the filename, use the analyse_text guessing engine to determine the
the Scala lexer (#392), the INI lexer (#391), the Clojure lexer (#387)
and the AS3 lexer (#389).
- Fixed three Perl heredoc lexing bugs (#379, #400, #422).
- Fixed a bug in the image formatter which misdetected lines (#380).
- Fixed bugs lexing extended Ruby strings and regexes.
- Fixed a bug when lexing git diffs.
- Fixed a bug lexing the empty commit in the PHP lexer (#405).
- Fixed a bug causing Python numbers to be mishighlighted as floats (#397).
- Fixed a bug when backslashes are used in odd locations in Python (#395).
- Fixed various bugs in Matlab and S-Plus lexers, thanks to Winston Chang (#410,
#411, #413, #414) and fmarc (#419).
- Fixed a bug in Haskell single-line comment detection (#426).
- Added new-style reStructuredText directive for docutils 0.5+ (#428).
(codename Dreiundzwanzig, released Nov 23, 2008)
- Don't use join(splitlines()) when converting newlines to ``\n``,
because that doesn't keep all newlines at the end when the
``stripnl`` lexer option is False.
- Add ``-N`` option to command-line interface to get a lexer name
for a given filename.
- Add Tango style, written by Andre Roberge for the Crunchy project.
- Add Python3TracebackLexer and ``python3`` option to
- Fix a few bugs in the Haskell lexer.
- Fix PythonTracebackLexer to be able to recognize SyntaxError and
- Provide one formatter class per image format, so that surprises like::
pygmentize -f gif -o foo.gif foo.py
creating a PNG file are avoided.
- Actually use the `font_size` option of the image formatter.
- Fixed numpy lexer that it doesn't listen for `*.py` any longer.
- Fixed HTML formatter so that text options can be Unicode
- Unified Diff lexer supports the "udiff" alias now.
- Fix a few issues in Scala lexer (#367).
- RubyConsoleLexer now supports simple prompt mode (#363).
- Add Applescript lexer, thanks to Andreas Amann (#330).
- Make the codetags more strict about matching words (#368).
- NginxConfLexer is a little more accurate on mimetypes and
(released Aug 24, 2008)
- Fix a Jython compatibility issue in pygments.unistring (#358).
(codename Straußenei, released Aug 23, 2008)
Many thanks go to Tim Hatch for writing or integrating most of the bug
fixes and new features.
- Lexers added:
* Nasm-style assembly language, thanks to delroth
* YAML, thanks to Kirill Simonov
* ActionScript 3, thanks to Pierre Bourdon
* Cheetah/Spitfire templates, thanks to Matt Good
* Lighttpd config files
* Nginx config files
* Gnuplot plotting scripts
* POV-Ray scene files
* Sqlite3 interactive console sessions
* Scala source files, thanks to Krzysiek Goj
- Lexers improved:
* C lexer highlights standard library functions now and supports C99
* Bash lexer now correctly highlights heredocs without preceding
* Vim lexer now highlights hex colors properly and knows a couple
* Irc logs lexer now handles xchat's default time format (#340) and
correctly highlights lines ending in ``>``.
* Support more delimiters for perl regular expressions (#258).
* ObjectiveC lexer now supports 2.0 features.
- Added "Visual Studio" style.
- Updated markdown processor to Markdown 1.7.
- Support roman/sans/mono style defs and use them in the LaTeX
- The RawTokenFormatter is no longer registered to ``*.raw`` and it's
documented that tokenization with this lexer may raise exceptions.
- New option ``hl_lines`` to HTML formatter, to highlight certain
- New option ``prestyles`` to HTML formatter.
- New option *-g* to pygmentize, to allow lexer guessing based on
filetext (can be slowish, so file extensions are still checked
- ``guess_lexer()`` now makes its decision much faster due to a cache
of whether data is xml-like (a check which is used in several
versions of ``analyse_text()``. Several lexers also have more
accurate ``analyse_text()`` now.
(codename Malzeug, released May 06, 2008)
- Lexers added:
* Darcs patches
* Matlab sessions
* Python 3
* S, S-plus, R statistics languages
- In the LatexFormatter, the *commandprefix* option is now by default
'PY' instead of 'C', since the latter resulted in several collisions
with other packages. Also, the special meaning of the *arg*
argument to ``get_style_defs()`` was removed.
- Added ImageFormatter, to format code as PNG, JPG, GIF or BMP.
(Needs the Python Imaging Library.)
- Support doc comments in the PHP lexer.
- Handle format specifications in the Perl lexer.
- Fix comment handling in the Batch lexer.
- Add more file name extensions for the C++, INI and XML lexers.
- Fixes in the IRC and MuPad lexers.
- Fix function and interface name highlighting in the Java lexer.
- Fix at-rule handling in the CSS lexer.
- Handle KeyboardInterrupts gracefully in pygmentize.
- Added BlackWhiteStyle.
- Bash lexer now correctly highlights math, does not require
whitespace after semicolons, and correctly highlights boolean
- Makefile lexer is now capable of handling BSD and GNU make syntax.
(codename Herbstzeitlose, released Oct 14, 2007)
- Lexers added:
* Literate Haskell
* Common Lisp
* Various assembly languages
* Gettext catalogs
* Squid configuration
* Debian control files
* MySQL-style SQL
- Lexers improved:
* Greatly improved the Haskell and OCaml lexers.
* Improved the Bash lexer's handling of nested constructs.
* The C# and Java lexers exhibited abysmal performance with some
input code; this should now be fixed.
* The IRC logs lexer is now able to colorize weechat logs too.
* The Lua lexer now recognizes multi-line comments.
* Fixed bugs in the D and MiniD lexer.
- The encoding handling of the command line mode (pygmentize) was
enhanced. You shouldn't get UnicodeErrors from it anymore if you
don't give an encoding option.
- Added a ``-P`` option to the command line mode which can be used to
give options whose values contain commas or equals signs.
- Added 256-color terminal formatter.
- Added an experimental SVG formatter.
- Added the ``lineanchors`` option to the HTML formatter, thanks to
Ian Charnas for the idea.
- Gave the line numbers table a CSS class in the HTML formatter.
- Added a Vim 7-like style.
(released Jun 27, 2007)
- Fixed POD highlighting in the Ruby lexer.
- Fixed Unicode class and namespace name highlighting in the C# lexer.
- Fixed Unicode string prefix highlighting in the Python lexer.
- Fixed a bug in the D and MiniD lexers.
- Fixed the included MoinMoin parser.
(codename Maikäfer, released May 30, 2007)
- Lexers added:
* Haskell, thanks to Adam Blinkinsop
* Redcode, thanks to Adam Blinkinsop
* D, thanks to Kirk McDonald
* MuPad, thanks to Christopher Creutzig
* MiniD, thanks to Jarrett Billingsley
* Vim Script, by Tim Hatch
- The HTML formatter now has a second line-numbers mode in which it
will just integrate the numbers in the same ``<pre>`` tag as the
- The `CSharpLexer` now is Unicode-aware, which means that it has an
option that can be set so that it correctly lexes Unicode
identifiers allowed by the C# specs.
- Added a `RaiseOnErrorTokenFilter` that raises an exception when the
lexer generates an error token, and a `VisibleWhitespaceFilter` that
converts whitespace (spaces, tabs, newlines) into visible
- Fixed the `do_insertions()` helper function to yield correct
- The ReST lexer now automatically highlights source code blocks in
".. sourcecode:: language" and ".. code:: language" directive
- Improved the default style (thanks to Tiberius Teng). The old
default is still available as the "emacs" style (which was an alias
- The `get_style_defs` method of HTML formatters now uses the
`cssclass` option as the default selector if it was given.
- Improved the ReST and Bash lexers a bit.
- Fixed a few bugs in the Makefile and Bash lexers, thanks to Tim
- Fixed a bug in the command line code that disallowed ``-O`` options
when using the ``-S`` option.
- Fixed a bug in the `RawTokenFormatter`.
(released Feb 15, 2007)
- Fixed little highlighting bugs in the Python, Java, Scheme and
Apache Config lexers.
- Updated the included manpage.
- Included a built version of the documentation in the source tarball.
(codename Faschingskrapfn, released Feb 14, 2007)
- Added a MoinMoin parser that uses Pygments. With it, you get
Pygments highlighting in Moin Wiki pages.
- Changed the exception raised if no suitable lexer, formatter etc. is
found in one of the `get_*_by_*` functions to a custom exception,
`pygments.util.ClassNotFound`. It is, however, a subclass of
`ValueError` in order to retain backwards compatibility.
- Added a `-H` command line option which can be used to get the
docstring of a lexer, formatter or filter.
- Made the handling of lexers and formatters more consistent. The
aliases and filename patterns of formatters are now attributes on
- Added an OCaml lexer, thanks to Adam Blinkinsop.
- Made the HTML formatter more flexible, and easily subclassable in
order to make it easy to implement custom wrappers, e.g. alternate
line number markup. See the documentation.
- Added an `outencoding` option to all formatters, making it possible
to override the `encoding` (which is used by lexers and formatters)
when using the command line interface. Also, if using the terminal
formatter and the output file is a terminal and has an encoding
attribute, use it if no encoding is given.
- Made it possible to just drop style modules into the `styles`
subpackage of the Pygments installation.
- Added a "state" keyword argument to the `using` helper.
- Added a `commandprefix` option to the `LatexFormatter` which allows
to control how the command names are constructed.
- Added quite a few new lexers, thanks to Tim Hatch:
* Java Server Pages
* Windows batch files
* Trac Wiki markup
* Python tracebacks
* and the Befunge esoteric programming language (yay!)
- Added Mako lexers by Ben Bangert.
- Added "fruity" style, another dark background originally vim-based
- Added sources.list lexer by Dennis Kaarsemaker.
- Added token stream filters, and a pygmentize option to use them.
- Changed behavior of `in` Operator for tokens.
- Added mimetypes for all lexers.
- Fixed some problems lexing Python strings.
- Fixed tickets: #167, #178, #179, #180, #185, #201.
(codename Zimtstern, released Dec 20, 2006)
- Added option for the HTML formatter to write the CSS to an external
file in "full document" mode.
- Added RTF formatter.
- Added Bash and Apache configuration lexers (thanks to Tim Hatch).
- Improved guessing methods for various lexers.
- Added `@media` support to CSS lexer (thanks to Tim Hatch).
- Added a Groff lexer (thanks to Tim Hatch).
- License change to BSD.
- Added lexers for the Myghty template language.
- Added a Scheme lexer (thanks to Marek Kubica).
- Added some functions to iterate over existing lexers, formatters and
- The HtmlFormatter's `get_style_defs()` can now take a list as an
argument to generate CSS with multiple prefixes.
- Support for guessing input encoding added.
- Encoding support added: all processing is now done with Unicode
strings, input and output are converted from and optionally to byte
strings (see the ``encoding`` option of lexers and formatters).
- Some improvements in the C(++) lexers handling comments and line
(released Oct 30, 2006)
- Fixed traceback in ``pygmentize -L`` (thanks to Piotr Ozarowski).
(codename PyKleur, released Oct 30, 2006)
- Initial public release.