html2tex 0.1.1 → 0.1.2

data/README.md

@@ -5,6 +5,22 @@ This is a simple library and command-line tool to convert HTML documents into
 LaTeX. The intended use is for preparing ebooks, but it might also be useful
 for other purposes.
 
+Command-line use
+----------------
+
+    Usage: html2tex [options] input.html [output.tex]
+        -t, --title TITLE                Set (or override) the title of the document
+        -a, --author AUTHOR              Set (or override) the author of the document
+        -c, --document-class CLASS       Set the LaTeX document class to be used; default is book
+        -h, --help                       Display this screen
+
+If the output file is not specified, the program will create an output file
+based on the input filename, replacing the extension with `.tex` – or, if there
+is no extension, by appending `.tex`.
+
+The title and author will, by default, be extracted from the HTML document,
+using either Dublin Core metadata or the HTML title and an author META element.
+
 Library use
 -----------
 
@@ -18,10 +34,14 @@ Library use
       HTML2TeX.new(html).to_tex(f)
     end
 
-Command-line use
-----------------
+A hash of options can be supplied as the second argument to `to_tex`; the
+available keys are:
+
+* `:author`
+* `:title`
+* `:class`
 
-    html2tex source.html destination.tex
+See the command-line section above for an explanation of these.
 
 Dependencies
 ------------

data/bin/html2tex

@@ -1,6 +1,18 @@
 #!/usr/bin/env ruby
 require "html2tex"
+require "html2tex/cli"
 
-File.open(ARGV[1], "w") do |output|
-  HTML2TeX.new(File.read(ARGV[0])).to_tex(output)
+cli = HTML2TeX::CLI.new(ARGV)
+begin
+  cli.parse
+rescue HTML2TeX::CLI::CleanExit
+  puts cli.usage
+  exit
+rescue HTML2TeX::CLI::DirtyExit
+  puts cli.usage
+  exit 1
+end
+
+File.open(cli.output_file, "w") do |io|
+  HTML2TeX.new(File.read(cli.input_file), cli.options).to_tex(io)
 end

data/lib/html2tex/basic_processor.rb

@@ -1,10 +1,15 @@
+require "html2tex/tex"
+
 class HTML2TeX
   class BasicProcessor
-    attr_reader :scanner
+    include TeX
+
+    attr_reader :scanner, :options
 
-    def initialize(scanner)
+    def initialize(scanner, options)
       @decoder = HTMLEntities.new
       @scanner = scanner
+      @options = options
     end
 
     def to_tex(buffer="")
@@ -35,11 +40,6 @@ class HTML2TeX
       tex_escape(@decoder.decode(RubyPants.new(@decoder.decode(s)).to_html))
     end
 
-    def tex_escape(s)
-      return nil if s.nil?
-      s.gsub(/[\\{}$&#%^_~]/, '\\\\\\0')
-    end
-
     def whitespace
       if @squash_next
         @squash_next = false
@@ -65,7 +65,7 @@ class HTML2TeX
         "\\\\"
       when /^h\d/
         @squash_next = true
-        HeadingProcessor.new(scanner, t).to_tex + "\n\n"
+        HeadingProcessor.new(scanner, t, @options).to_tex + "\n\n"
       else
         ""
       end

data/lib/html2tex/cli.rb

@@ -0,0 +1,54 @@
+require "optparse"
+
+class HTML2TeX
+  class CLI
+    CleanExit = Class.new(RuntimeError)
+    DirtyExit = Class.new(RuntimeError)
+
+    attr_reader :options, :input_file, :output_file
+
+    def initialize(argv)
+      @argv    = argv
+      @options = {}
+    end
+
+    def usage
+      parser.to_s
+    end
+
+    def parse
+      parser.parse!(@argv)
+
+      unless @input_file = @argv[0]
+        raise DirtyExit
+      end
+
+      @output_file = @argv[1] || @input_file.sub(/\.[^\.]+$|$/, ".tex")
+    end
+
+  private
+    def parser
+      @parser ||= OptionParser.new{ |opts|
+        opts.banner = "Usage: #{File.basename($0)} [options] input.html [output.tex]"
+
+        opts.on "-t", "--title TITLE",
+                "Set (or override) the title of the document" do |title|
+          options[:title] = title
+        end
+        opts.on "-a", "--author AUTHOR",
+                "Set (or override) the author of the document" do |author|
+          options[:author] = author
+        end
+        opts.on "-c", "--document-class CLASS",
+                "Set the LaTeX document class to be used; default is book" do |klass|
+          options[:class] = klass
+        end
+
+        opts.on "-h", "--help",
+                "Display this screen" do
+          raise CleanExit
+        end
+      }
+    end
+  end
+end

data/lib/html2tex/heading_processor.rb

@@ -4,13 +4,14 @@ class HTML2TeX
   class HeadingProcessor < BasicProcessor
     HEADINGS = %w[part chapter section subsection subsubsection]
 
-    def initialize(scanner, label)
+    def initialize(scanner, label, options)
       @label = label
-      super(scanner)
+      super(scanner, options)
     end
 
-    def to_tex(*args)
-      wrap(super(*args))
+    def to_tex(buffer="")
+      buffer << wrap(super(""))
+      buffer
     end
 
   private

data/lib/html2tex/preamble_processor.rb

@@ -0,0 +1,78 @@
+require "html2tex/basic_processor"
+require "nokogiri"
+
+class HTML2TeX
+  class PreambleProcessor < BasicProcessor
+
+    def to_tex(buffer="")
+      read_html_head
+      [ tex(:documentclass, options[:class]),
+        tex(:title, title),
+        tex(:author, author),
+        tex(:begin, "document"),
+        tex(:maketitle),
+        ""
+      ].each do |s|
+        buffer << s << "\n"
+      end
+      buffer
+    end
+
+  private
+    def read_html_head
+      scanner.scan %r{\s*}
+      scanner.scan %r{<!doctype[^>]*>\s*}i
+      scanner.scan %r{<html[^>]*>\s*}i
+      if head = scanner.scan(%r{<head[^>]*>.*?</head>}im)
+        h = metadata_hash(head)
+        @author = h["dc.creator"] || h["author"]
+        @title  = h["dc.title"] || head[%r{<title[^>]*>\s*(.*?)\s*</title>}im, 1]
+      end
+    end
+
+    def author
+      @options[:author] || @author || "AUTHOR"
+    end
+
+    def title
+      @options[:title] || @title || "TITLE"
+    end
+
+    def parse_meta_element(meta)
+      data = {}
+      s = StringScanner.new(meta)
+      s.scan(/<meta\s+/i)
+      until s.eos? do
+        if s.scan(/\s*\/?>/)
+          return data
+        else
+          key = s.scan(/[^=]+/).downcase
+          s.scan(/=/)
+        end
+
+        if s.scan(/"/)
+          data[key] = s.scan(/[^"]+/)
+          s.scan(/"/)
+        elsif s.scan(/'/)
+          data[key] = s.scan(/[^']+/)
+          s.scan(/'/)
+        else
+          data[key] = s.scan(/\S+/)
+        end
+        s.scan(/\s+/)
+      end
+    end
+
+    def metadata_hash(head)
+      hash = {}
+      head.scan(%r{<meta\s[^>]+>}im).each do |meta|
+        m = parse_meta_element(meta)
+        if (name = m["name"]) && (content = m["content"])
+          hash[name.downcase] = content
+        end
+      end
+      hash
+    end
+  end
+end
+

data/lib/html2tex/tex.rb

@@ -0,0 +1,14 @@
+class HTML2TeX
+  module TeX
+    def tex(name, param=nil)
+      directive = "\\" + name.to_s
+      directive << "{" << tex_escape(param) << "}" if param
+      directive
+    end
+
+    def tex_escape(s)
+      return nil if s.nil?
+      s.gsub(/[\\{}$&#%^_~]/, '\\\\\\0')
+    end
+  end
+end

data/lib/html2tex/version.rb

@@ -1,3 +1,3 @@
 class HTML2TeX
-  VERSION = "0.1.1"
+  VERSION = "0.1.2"
 end

data/lib/html2tex.rb

@@ -2,13 +2,36 @@ require "htmlentities"
 require "rubypants"
 require "html2tex/version"
 require "html2tex/basic_processor"
+require "html2tex/preamble_processor"
+require "html2tex/tex"
 
 class HTML2TeX
-  def initialize(html)
-    @html = html
+  include TeX
+
+  DEFAULT_OPTIONS = {
+    :document => true,
+    :class    => "book"
+  }
+
+  def initialize(html, options={})
+    @html    = html
+    @options = DEFAULT_OPTIONS.merge(options)
   end
 
   def to_tex(buffer="")
-    BasicProcessor.new(StringScanner.new(@html)).to_tex(buffer)
+    scanner = StringScanner.new(@html)
+
+    if @options[:document]
+      PreambleProcessor.new(scanner, @options).to_tex(buffer)
+    end
+
+    BasicProcessor.new(scanner, @options).to_tex(buffer)
+
+    if @options[:document]
+      buffer << "\n\n" << tex(:end, "document")
+    end
+
+    buffer << "\n"
+    buffer
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: html2tex
 version: !ruby/object:Gem::Version 
-  version: 0.1.1
+  version: 0.1.2
 platform: ruby
 authors: 
 - Paul Battley
@@ -56,7 +56,10 @@ files:
 - lib/html2tex.rb
 - lib/html2tex/version.rb
 - lib/html2tex/basic_processor.rb
+- lib/html2tex/cli.rb
+- lib/html2tex/preamble_processor.rb
 - lib/html2tex/heading_processor.rb
+- lib/html2tex/tex.rb
 - README.md
 - COPYING
 has_rdoc: true