implement a TeX lexer (closes #10)

lib/rouge.rb

@@ -18,8 +18,10 @@ def highlight(text, lexer, formatter)
 load load_dir.join('rouge/text_analyzer.rb')
 load load_dir.join('rouge/token.rb')
 load load_dir.join('rouge/lexer.rb')
+
 load load_dir.join('rouge/lexers/text.rb')
 load load_dir.join('rouge/lexers/diff.rb')
+load load_dir.join('rouge/lexers/tex.rb')
 
 load load_dir.join('rouge/lexers/shell.rb')
 

lib/rouge/lexers/tex.rb

@@ -0,0 +1,75 @@
+module Rouge
+  module Lexers
+    class TeX < RegexLexer
+      tag 'tex'
+      aliases 'TeX', 'LaTeX', 'latex'
+
+      filenames '*.tex', '*.aux', '*.toc'
+      mimetypes 'text/x-tex', 'text/x-latex'
+
+      def self.analyze_text(text)
+        return 1 if text =~ /\A\s*\\documentclass/
+        return 1 if text =~ /\A\s*\\input/
+        return 1 if text =~ /\A\s*\\documentstyle/
+        return 1 if text =~ /\A\s*\\relax/
+      end
+
+      command = /\\([a-z]+|\s+|.)/i
+
+      state :general do
+        rule /%.*$/, 'Comment'
+        rule /[{}&_^]/, 'Name.Builtin'
+      end
+
+      state :root do
+        rule /\\\[/, 'Literal.String.Backtick', :displaymath
+        rule /\\\(/, 'Literal.String', :inlinemath
+        rule /\$\$/, 'Literal.String.Backtick', :displaymath
+        rule /\$/, 'Literal.String', :inlinemath
+        rule /\\(begin|end)\{.*?\}/, 'Name.Tag'
+
+        rule /(\\verb)\b(.)/ do |m|
+          group 'Name.Builtin'
+          group 'Name.Constant'
+          delim = Regexp.escape(m[2])
+
+          push do
+            rule /#{delim}/, 'Name.Constant', :pop!
+            rule /[^#{delim}]+/m, 'Literal.String.Other'
+          end
+        end
+
+        rule command, 'Keyword', :command
+        mixin :general
+        rule /[^\\$%&_^{}]+/, 'Text'
+      end
+
+      state :math do
+        rule command, 'Name.Variable'
+        mixin :general
+        rule /[0-9]+/, 'Literal.Number'
+        rule /[-=!+*\/()\[\]]/, 'Operator'
+        rule /[^=!+*\/()\[\]\\$%&_^{}0-9-]+/, 'Name.Builtin'
+      end
+
+      state :inlinemath do
+        rule /\\\)/, 'Literal.String', :pop!
+        rule /\$/, 'Literal.String', :pop!
+        mixin :math
+      end
+
+      state :displaymath do
+        rule /\\\]/, 'Literal.String.Backtick', :pop!
+        rule /\$\$/, 'Literal.String.Backtick', :pop!
+        rule /\$/, 'Name.Builtin'
+        mixin :math
+      end
+
+      state :command do
+        rule /\[.*?\]/, 'Name.Attribute'
+        rule /\*/, 'Keyword'
+        rule(//) { pop! }
+      end
+    end
+  end
+end

spec/lexers/tex_spec.rb

@@ -0,0 +1,22 @@
+describe Rouge::Lexers::TeX do
+  let(:subject) { Rouge::Lexers::TeX.new }
+
+  describe 'guessing' do
+    include Support::Guessing
+
+    it 'guesses by filename' do
+      assert_guess :filename => 'foo.tex'
+      assert_guess :filename => 'foo.toc'
+      assert_guess :filename => 'foo.aux'
+    end
+
+    it 'guesses by mimetype' do
+      assert_guess :mimetype => 'text/x-tex'
+      assert_guess :mimetype => 'text/x-latex'
+    end
+
+    it 'guesses by source' do
+      assert_guess :source => '\\documentclass{article}'
+    end
+  end
+end

spec/visual/samples/tex

@@ -0,0 +1,132 @@
+\documentclass{article}
+
+\begin{document}
+
+
+
+\centerline{\sc \large A Simple Sample \LaTeX\ File}
+\vspace{.5pc}
+\centerline{\sc Stupid Stuff I Wish Someone Had Told Me Four Years Ago}
+\centerline{\it (Read the .tex file along with this or it won't 
+            make much sense)}
+\vspace{2pc}
+
+The first thing to realize about \LaTeX\ is that it is not ``WYSIWYG''. 
+In other words, it isn't a word processor; what you type into your 
+.tex file is not what you'll see in your .dvi file.  For example, 
+\LaTeX\ will      completely     ignore               extra
+spaces    within                             a line of your .tex file.
+Pressing return
+in 
+the 
+middle 
+of
+a
+line
+will not register in your .dvi file. However, a double carriage-return
+is read as a paragraph break.
+
+Like this.  But any carriage-returns after the first two will be 
+completely ignored; in other words, you 
+
+
+can't 
+
+add
+
+
+
+
+
+
+more 
+
+
+
+
+space 
+
+
+between 
+
+
+
+
+lines, no matter how many times you press return in your .tex file.
+
+In order to add vertical space you have to use ``vspace''; for example, 
+you could add an inch of space by typing \verb|\vspace{1in}|, like this:
+\vspace{1in}
+
+To get three lines of space you would type \verb|\vspace{3pc}|
+(``pc'' stands for ``pica'', a font-relative size), like this:
+\vspace{3pc}
+
+Notice that \LaTeX\ commands are always preceeded by a backslash.  
+Some commands, like \verb|\vspace|, take arguments (here, a length) in
+curly brackets.  
+
+The second important thing to notice about \LaTeX\ is that you type 
+in various ``environments''...so far we've just been typing regular 
+text (except for a few inescapable usages of \verb|\verb| and the
+centered, smallcaps, large title).  There are basically two ways that 
+you can enter and/or exit an environment;
+\vspace{1pc}
+
+\centerline{this is the first way...}
+
+\begin{center}
+this is the second way.
+\end{center}
+
+\noindent Actually there is one more way, used above; for example, 
+{\sc this way}.  The way that you get in and out of environment varies
+depending on which kind of environment you want; for example, you use 
+\verb|\underline| ``outside'', but \verb|\it| ``inside''; 
+notice \underline{this} versus {\it this}.
+
+The real power of \LaTeX\ (for us) is in the math environment. You 
+push and pop out of the math environment by typing \verb|$|. For 
+example, $2x^3 - 1 = 5$ is typed between dollar signs as
+\verb|$2x^3 - 1 = 5$|. Perhaps a more interesting example is
+$\lim_{N \to \infty} \sum_{k=1}^N f(t_k) \Delta t$.
+
+You can get a fancier, display-style math 
+environment by enclosing your equation with double dollar signs.  
+This will center your equation, and display sub- and super-scripts in 
+a more readable fashion:
+
+$$\lim_{N \to \infty} \sum_{k=1}^N f(t_k) \Delta t.$$
+
+If you don't want your equation to be centered, but you want the nice 
+indicies and all that, you can use \verb|\displaystyle| and get your 
+formula ``in-line''; using our example this is 
+$\displaystyle \lim_{N \to \infty} \sum_{k=1}^N f(t_k) \Delta t.$  Of 
+course this can screw up your line spacing a little bit.
+
+There are many more things to know about \LaTeX\ and we can't 
+possibly talk about them all here.
+You can use \LaTeX\ to get tables, commutative diagrams, figures, 
+aligned equations, cross-references, labels, matrices, and all manner 
+of strange things into your documents.  You can control margins, 
+spacing, alignment, {\it et cetera} to higher degrees of accuracy than 
+the human eye can percieve.  You can waste entire days typesetting 
+documents to be ``just so''.  In short, \LaTeX\ rules.
+
+The best way to learn \LaTeX\ is by example. Get yourself a bunch
+of .tex files, see what kind of output they produce, and figure out how
+to modify them to do what you want.  There are many template and 
+sample files on the department \LaTeX\ page and in real life in the 
+big binder that should be in the computer lab somewhere.  Good luck!
+
+
+
+
+
+
+
+
+
+
+
+\end{document}