LXSH: Lexing & Syntax Highlighting in Lua

LXSH is a collection of [lexers] lexing and [syntax highlighters] highlighting written in Lua lua using the excellent pattern-matching library LPeg lpeg. Several syntaxes are currently supported: Lua, C, BibTeX and shell script. The syntax highlighters support three output formats: HTML html designed to be easily embedded in web pages, LaTeX latex which can be used to generate high quality PDF files and RTF rtf which can be used in graphical text editors like Microsoft Word and LibreOffice (formerly OpenOffice). Three predefined color schemes are included. Here are some examples of the supported input languages, output formats and color schemes:

	Earendel	Slate	Wiki
Lua:	HTML · PDF · RTF	HTML · PDF · RTF	HTML · PDF · RTF
Lua (interactive prompt):	HTML · PDF · RTF	HTML · PDF · RTF	HTML · PDF · RTF
C:	HTML · PDF · RTF	HTML · PDF · RTF	HTML · PDF · RTF
BibTeX:	HTML · PDF · RTF	HTML · PDF · RTF	HTML · PDF · RTF
Shell script:	HTML · PDF · RTF	HTML · PDF · RTF	HTML · PDF · RTF

As you may have noticed in the above examples, the syntax highlighters replace standard library identifiers (and then some) with hyperlinks to the relevant documentation. You can also try switching between style sheets while staying on the same web page by using your web browser's View → Page styles menu (this works using so-called "alternate style sheets").

Installation

The easiest way to download and install LXSH is using LuaRocks luarocks:

$ luarocks install lxsh

If you don't have LuaRocks installed you can [download the latest release] zipball directly from GitHub as a ZIP archive. To install create an lxsh directory in your [$LUA_PATH] lua_path and copy the contents of the src directory from the ZIP archive to the lxsh directory so that you end up with the following structure:

• $LUA_PATH/lxsh/init.lua
• $LUA_PATH/lxsh/lexers/*.lua
• $LUA_PATH/lxsh/highlighters/*.lua
• $LUA_PATH/lxsh/formatters/*.lua
• $LUA_PATH/lxsh/colors/*.lua
• $LUA_PATH/lxsh/docs/*.lua

Usage

If you want to call a lexer or access an LPeg pattern defined by a lexer you can do so as follows (this example demonstrates the Lua lexer but the other lexers work the same way):

> -- Load the LXSH module.
> require 'lxsh'

> -- Run the lexer on a string of source code.
> for kind, text, lnum, cnum in lxsh.lexers.lua.gmatch 'i = i + 1\n-- example' do
>>  print(string.format('%s: %q (%i:%i)', kind, text, lnum, cnum))
>> end
identifier: "i" (1:1)
whitespace: " " (1:2)
operator:   "=" (1:3)
whitespace: " " (1:4)
identifier: "i" (1:5)
whitespace: " " (1:6)
operator:   "+" (1:7)
whitespace: " " (1:8)
number:     "1" (1:9)
whitespace: "\n" (1:10)
comment:    "-- example" (2:1)

> -- Use one of the patterns defined by the lexer.
> lxsh.lexers.lua.patterns.comment:match '--[=[ this is a long comment ]=]'

Note that you only need to load the main LXSH module with require(), the lexer and highlighter submodules are automatically loaded as they're first used. Lexers define the following functions:

• lexer.find(subject [, init [, options ]]) takes a string and optional starting position, matches a single token (anchored) and returns two values: the token kind and the last matched character
• lexer.match(subject [, init [, options ]]) takes a string and optional starting position, matches a single token (anchored) and returns two values: the token kind and the matched text
• lexer.gmatch(subject [, options]) returns an iterator that produces four values on each iteration: the kind of token (see below), the matched text, the starting line number and the starting column number (line and column numbers start at 1)

When options is given it should be a table of options that can be used to configure lexers. Currently only one option is defined: When you pass join_identifiers=true to the Lua lexer, expressions like io.write will be matched as a single identifier instead of the sequence (identifier io, operator ., identifier write).

The syntax highlighters can be used as follows:

> print(lxsh.highlighters.lua("require 'lpeg'", { formatter = lxsh.formatters.html, external = true }))
<pre class="sourcecode lua">
<a href="http://www.lua.org/manual/5.1/manual.html#pdf-require" class="library">require</a>
<span class="constant">'lpeg'</span>
</pre>

You can customize the output of the highlighters by passing a table with one or more of the following options:

• encodews: Instruct the HTML highlighter to replace newlines with <br> elements and ordinary spaces with non-breaking spaces so that whitespace is preserved when the highlighted code isn't embedded in a <pre> block
• external: By default the HTML highlighter generates inline CSS which makes it easier to use the output directly but it also bloats the size significantly. If you want to reduce the size and don't mind including an external style sheet you can set this option to true. You'll need to make sure the required styles are loaded, e.g. by embedding the output of lxsh.highlighters.html.preamble(preferred, includeswitcher) in the <head> of your HTML document (the preferred argument indicates the default style sheet and if you pass includeswitcher as true then an interactive style sheet switcher using JavaScript is included)
• colors: The color scheme to use, one of the following:
  • lxsh.colors.earendel based on the [Vim color scheme Earendel] earendel by Georg Dahn (this is the default)
  • lxsh.colors.slate based on the [Vim color scheme Slate] slate by Ralph Amissah
  • lxsh.colors.wiki based on the style of the [lua-users wiki] lua_wiki

Tokens produced by the lexers

The Lua lexer produces the following tokens:

• comment
• constant (true, false and nil)
• error (invalid input)
• identifier
• keyword
• number
• operator
• string
• whitespace

The C lexer produces the following tokens:

• comment
• character (literals like 'C')
• string (literals like "Lua")
• error (invalid input)
• identifier
• keyword
• number
• operator
• preprocessor
• whitespace

The BibTeX lexer produces the following tokens:

• entry (e.g. @Book)
• field (e.g. author)
• identifier
• string
• number
• operator
• delimiter
• whitespace
• error (invalid input)

The shell script lexer produces the following tokens:

• comment
• number
• string
• variable
• operator
• keyword
• command
• error (invalid input)

Contact

If you have questions, bug reports, suggestions, etc. the author can be contacted at [email protected]. The latest version is available at http://peterodding.com/code/lua/lxsh/ and http://github.com/xolox/lua-lxsh.

License

This software is licensed under the [MIT license] mit.
© 2011 Peter Odding <[email protected]>.