Class: Newstile::Converter::Latex

Inherits:
Base
  • Object
show all
Defined in:
lib/newstile/converter/latex.rb

Overview

Converts a Newstile::Document to LaTeX. This converter uses ideas from other Markdown-to-LaTeX converters like Pandoc and Maruku.

Constant Summary collapse

HEADER_TYPES = 
{
  1 => 'section',
  2 => 'subsection',
  3 => 'subsubsection',
  4 => 'paragraph',
  5 => 'subparagraph',
  6 => 'subparagraph'
}
TABLE_ALIGNMENT_CHAR = 
{:default => 'l', :left => 'l', :center => 'c', :right => 'r'}
ENTITY_CONV_TABLE =

Inspired by Maruku: entity conversion table based on the one from htmltolatex (sourceforge.net/projects/htmltolatex/), with some small adjustments/additions

{
  913 => ['$A$'],
  914 => ['$B$'],
  915 => ['$\Gamma$'],
  916 => ['$\Delta$'],
  917 => ['$E$'],
  918 => ['$Z$'],
  919 => ['$H$'],
  920 => ['$\Theta$'],
  921 => ['$I$'],
  922 => ['$K$'],
  923 => ['$\Lambda$'],
  924 => ['$M$'],
  925 => ['$N$'],
  926 => ['$\Xi$'],
  927 => ['$O$'],
  928 => ['$\Pi$'],
  929 => ['$P$'],
  931 => ['$\Sigma$'],
  932 => ['$T$'],
  933 => ['$Y$'],
  934 => ['$\Phi$'],
  935 => ['$X$'],
  936 => ['$\Psi$'],
  937 => ['$\Omega$'],
  945 => ['$\alpha$'],
  946 => ['$\beta$'],
  947 => ['$\gamma$'],
  948 => ['$\delta$'],
  949 => ['$\epsilon$'],
  950 => ['$\zeta$'],
  951 => ['$\eta$'],
  952 => ['$\theta$'],
  953 => ['$\iota$'],
  954 => ['$\kappa$'],
  955 => ['$\lambda$'],
  956 => ['$\mu$'],
  957 => ['$\nu$'],
  958 => ['$\xi$'],
  959 => ['$o$'],
  960 => ['$\pi$'],
  961 => ['$\rho$'],
  963 => ['$\sigma$'],
  964 => ['$\tau$'],
  965 => ['$\upsilon$'],
  966 => ['$\phi$'],
  967 => ['$\chi$'],
  968 => ['$\psi$'],
  969 => ['$\omega$'],
  962 => ['$\varsigma$'],
  977 => ['$\vartheta$'],
  982 => ['$\varpi$'],
  8230 => ['\ldots'],
  8242 => ['$\prime$'],
  8254 => ['-'],
  8260 => ['/'],
  8472 => ['$\wp$'],
  8465 => ['$\Im$'],
  8476 => ['$\Re$'],
  8501 => ['$\aleph$'],
  8226 => ['$\bullet$'],
  8482 => ['$^{\rm TM}$'],
  8592 => ['$\leftarrow$'],
  8594 => ['$\rightarrow$'],
  8593 => ['$\uparrow$'],
  8595 => ['$\downarrow$'],
  8596 => ['$\leftrightarrow$'],
  8629 => ['$\hookleftarrow$'],
  8657 => ['$\Uparrow$'],
  8659 => ['$\Downarrow$'],
  8656 => ['$\Leftarrow$'],
  8658 => ['$\Rightarrow$'],
  8660 => ['$\Leftrightarrow$'],
  8704 => ['$\forall$'],
  8706 => ['$\partial$'],
  8707 => ['$\exists$'],
  8709 => ['$\emptyset$'],
  8711 => ['$\nabla$'],
  8712 => ['$\in$'],
  8715 => ['$\ni$'],
  8713 => ['$\notin$'],
  8721 => ['$\sum$'],
  8719 => ['$\prod$'],
  8722 => ['$-$'],
  8727 => ['$\ast$'],
  8730 => ['$\surd$'],
  8733 => ['$\propto$'],
  8734 => ['$\infty$'],
  8736 => ['$\angle$'],
  8743 => ['$\wedge$'],
  8744 => ['$\vee$'],
  8745 => ['$\cup$'],
  8746 => ['$\cap$'],
  8747 => ['$\int$'],
  8756 => ['$\therefore$', 'amssymb'],
  8764 => ['$\sim$'],
  8776 => ['$\approx$'],
  8773 => ['$\cong$'],
  8800 => ['$\neq$'],
  8801 => ['$\equiv$'],
  8804 => ['$\leq$'],
  8805 => ['$\geq$'],
  8834 => ['$\subset$'],
  8835 => ['$\supset$'],
  8838 => ['$\subseteq$'],
  8839 => ['$\supseteq$'],
  8836 => ['$\nsubset$', 'amssymb'],
  8853 => ['$\oplus$'],
  8855 => ['$\otimes$'],
  8869 => ['$\perp$'],
  8901 => ['$\cdot$'],
  8968 => ['$\rceil$'],
  8969 => ['$\lceil$'],
  8970 => ['$\lfloor$'],
  8971 => ['$\rfloor$'],
  9001 => ['$\rangle$'],
  9002 => ['$\langle$'],
  9674 => ['$\lozenge$', 'amssymb'],
  9824 => ['$\spadesuit$'],
  9827 => ['$\clubsuit$'],
  9829 => ['$\heartsuit$'],
  9830 => ['$\diamondsuit$'],
  38 => ['\&'],
  34 => ['"'],
  39 => ['\''],
  169 => ['\copyright'],
  60 => ['\textless{}'],
  62 => ['\textgreater{}'],
  338 => ['\OE'],
  339 => ['\oe'],
  352 => ['\v{S}'],
  353 => ['\v{s}'],
  376 => ['\"Y'],
  710 => ['\textasciicircum'],
  732 => ['\textasciitilde'],
  8211 => ['--'],
  8212 => ['---'],
  8216 => ['`'],
  8217 => ['\''],
  8220 => ['``'],
  8221 => ['\'\''],
  8224 => ['\dag'],
  8225 => ['\ddag'],
  8240 => ['\permil', 'wasysym'],
  8364 => ['\euro', 'eurosym'],
  8249 => ['\guilsinglleft'],
  8250 => ['\guilsinglright'],
  8218 => ['\quotesinglbase', 'mathcomp'],
  8222 => ['\quotedblbase', 'mathcomp'],
  402 => ['\textflorin', 'mathcomp'],
  381 => ['\v{Z}'],
  382 => ['\v{z}'],
  160 => ['\nolinebreak'],
  161 => ['\textexclamdown'],
  163 => ['\pounds'],
  164 => ['\currency', 'wasysym'],
  165 => ['\textyen', 'textcomp'],
  166 => ['\brokenvert', 'wasysym'],
  167 => ['\S'],
  171 => ['\guillemotleft'],
  187 => ['\guillemotright'],
  174 => ['\textregistered'],
  170 => ['\textordfeminine'],
  172 => ['$\neg$'],
  176 => ['$\degree$', 'mathabx'],
  177 => ['$\pm$'],
  180 => ['\''],
  181 => ['$\mu$'],
  182 => ['\P'],
  183 => ['$\cdot$'],
  186 => ['\textordmasculine'],
  162 => ['\cent', 'wasysym'],
  185 => ['$^1$'],
  178 => ['$^2$'],
  179 => ['$^3$'],
  189 => ['$\frac{1}{2}$'],
  188 => ['$\frac{1}{4}$'],
  190 => ['$\frac{3}{4}'],
  192 => ['\`A'],
  193 => ['\\\'A'],
  194 => ['\^A'],
  195 => ['\~A'],
  196 => ['\"A'],
  197 => ['\AA'],
  198 => ['\AE'],
  199 => ['\cC'],
  200 => ['\`E'],
  201 => ['\\\'E'],
  202 => ['\^E'],
  203 => ['\"E'],
  204 => ['\`I'],
  205 => ['\\\'I'],
  206 => ['\^I'],
  207 => ['\"I'],
  208 => ['$\eth$', 'amssymb'],
  209 => ['\~N'],
  210 => ['\`O'],
  211 => ['\\\'O'],
  212 => ['\^O'],
  213 => ['\~O'],
  214 => ['\"O'],
  215 => ['$\times$'],
  216 => ['\O'],
  217 => ['\`U'],
  218 => ['\\\'U'],
  219 => ['\^U'],
  220 => ['\"U'],
  221 => ['\\\'Y'],
  222 => ['\Thorn', 'wasysym'],
  223 => ['\ss'],
  224 => ['\`a'],
  225 => ['\\\'a'],
  226 => ['\^a'],
  227 => ['\~a'],
  228 => ['\"a'],
  229 => ['\aa'],
  230 => ['\ae'],
  231 => ['\cc'],
  232 => ['\`e'],
  233 => ['\\\'e'],
  234 => ['\^e'],
  235 => ['\"e'],
  236 => ['\`i'],
  237 => ['\\\'i'],
  238 => ['\^i'],
  239 => ['\"i'],
  240 => ['$\eth$'],
  241 => ['\~n'],
  242 => ['\`o'],
  243 => ['\\\'o'],
  244 => ['\^o'],
  245 => ['\~o'],
  246 => ['\"o'],
  247 => ['$\divide$'],
  248 => ['\o'],
  249 => ['\`u'],
  250 => ['\\\'u'],
  251 => ['\^u'],
  252 => ['\"u'],
  253 => ['\\\'y'],
  254 => ['\thorn', 'wasysym'],
  255 => ['\"y'],
}
TYPOGRAPHIC_SYMS = 
{
  :mdash => '---', :ndash => '--', :qdash_space => "- ", :qdash => "-", :hellip => '\ldots{}',
  :laquo_space => '\guillemotleft{}~', :raquo_space => '~\guillemotright{}',
  :laquo => '\guillemotleft{}', :raquo => '\guillemotright{}'
}
SMART_QUOTE_SYMS = 
{:lsquo => '`', :rsquo => '\'', :ldquo => '``', :rdquo => '\'\''}
ESCAPE_MAP = 
{
  "^"  => "\\^{}",
  "\\" => "\\textbackslash{}",
  "~"  => "\\ensuremath{\\sim}",
  "|"  => "\\textbar{}",
  "<"  => "\\textless{}",
  ">"  => "\\textgreater{}"
}.merge(Hash[*("{}$%&_#".scan(/./).map {|c| [c, "\\#{c}"]}.flatten)])
ESCAPE_RE = 
Regexp.union(*ESCAPE_MAP.collect {|k,v| k})

Instance Method Summary collapse

Methods inherited from Base

apply_template, convert, #generate_id, get_template

Constructor Details

#initialize(doc) ⇒ Latex

Initialize the LaTeX converter with the given Newstile document doc.



36
37
38
39
40
41
 
# File 'lib/newstile/converter/latex.rb', line 36

def initialize(doc)
  super
  #TODO: set the footnote counter at the beginning of the document
  @doc.options[:footnote_nr]
  @doc.conversion_infos[:packages] = Set.new
end

Instance Method Details

#attribute_list(el) ⇒ Object 



568
569
570
571
572
 
# File 'lib/newstile/converter/latex.rb', line 568

def attribute_list(el)
  attrs = el.attr.map {|k,v| v.nil? ? '' : " #{k}=\"#{v.to_s}\""}.compact.sort.join('')
  attrs = "   % #{attrs}" if !attrs.empty?
  attrs
end

#convert(el, opts = {}) ⇒ Object 



43
44
45
 
# File 'lib/newstile/converter/latex.rb', line 43

def convert(el, opts = {})
  send("convert_#{el.type}", el, opts)
end

#convert_a(el, opts) ⇒ Object 



216
217
218
219
220
221
222
223
 
# File 'lib/newstile/converter/latex.rb', line 216

def convert_a(el, opts)
  url = el.attr['href']
  if url =~ /^#/
    "\\hyperlink{#{url[1..-1]}}{#{inner(el, opts)}}"
  else
    "\\href{#{url}}{#{inner(el, opts)}}"
  end
end

#convert_abbreviation(el, opts) ⇒ Object 



550
551
552
 
# File 'lib/newstile/converter/latex.rb', line 550

def convert_abbreviation(el, opts)
  el.value
end

#convert_blank(el, opts) ⇒ Object 



62
63
64
 
# File 'lib/newstile/converter/latex.rb', line 62

def convert_blank(el, opts)
  opts[:result] =~ /\n\n\Z|\A\Z/ ? "" : "\n"
end

#convert_blockquote(el, opts) ⇒ Object 



102
103
104
 
# File 'lib/newstile/converter/latex.rb', line 102

def convert_blockquote(el, opts)
  latex_environment(el.children.size > 1 ? 'quotation' : 'quote', el, inner(el, opts))
end

#convert_br(el, opts) ⇒ Object 



210
211
212
213
214
 
# File 'lib/newstile/converter/latex.rb', line 210

def convert_br(el, opts)
  res = "\\newline"
  res += "\n" if (c = opts[:parent].children[opts[:index]+1]) && (c.type != :text || c.value !~ /^\s*\n/)
  res
end

#convert_codeblock(el, opts) ⇒ Object 



83
84
85
86
87
88
89
90
91
92
93
94
95
 
# File 'lib/newstile/converter/latex.rb', line 83

def convert_codeblock(el, opts)
  show_whitespace = el.attr['class'].to_s =~ /\bshow-whitespaces\b/
  lang = el.attr['lang']
  if show_whitespace || lang
    result = "\\lstset{showspaces=%s,showtabs=%s}\n" % (show_whitespace ? ['true', 'true'] : ['false', 'false'])
    result += "\\lstset{language=#{lang}}\n" if lang
    result += "\\lstset{basicstyle=\\ttfamily\\footnotesize}\\lstset{columns=fixed,frame=tlbr}\n"
    attrs = attribute_list(el)
    "#{result}\\begin{lstlisting}#{attrs}\n#{el.value}\n\\end{lstlisting}#{attrs}\n"
  else
    "\\begin{verbatim}#{el.value}\\end{verbatim}\n"
  end
end

#convert_codespan(el, opts) ⇒ Object 



238
239
240
 
# File 'lib/newstile/converter/latex.rb', line 238

def convert_codespan(el, opts)
  "{\\tt #{escape(el.value)}}"
end

#convert_comment(el, opts) ⇒ Object 



206
207
208
 
# File 'lib/newstile/converter/latex.rb', line 206

def convert_comment(el, opts)
  el.value.split(/\n/).map {|l| "% #{l}"}.join("\n") + "\n"
end

#convert_dd(el, opts) ⇒ Object 



152
153
154
 
# File 'lib/newstile/converter/latex.rb', line 152

def convert_dd(el, opts)
  "#{inner(el, opts)}\n\n"
end

#convert_dl(el, opts) ⇒ Object 



140
141
142
 
# File 'lib/newstile/converter/latex.rb', line 140

def convert_dl(el, opts)
  latex_environment('description', el, inner(el, opts))
end

#convert_dt(el, opts) ⇒ Object 



148
149
150
 
# File 'lib/newstile/converter/latex.rb', line 148

def convert_dt(el, opts)
  "\\item[#{inner(el, opts)}] "
end

#convert_em(el, opts) ⇒ Object 



255
256
257
 
# File 'lib/newstile/converter/latex.rb', line 255

def convert_em(el, opts)
  "\\emph{#{inner(el, opts)}}"
end

#convert_entity(el, opts) ⇒ Object 



510
511
512
513
514
515
516
517
518
519
 
# File 'lib/newstile/converter/latex.rb', line 510

def convert_entity(el, opts)
  text, package = ENTITY_CONV_TABLE[el.value.code_point]
  if text
    @doc.conversion_infos[:packages] << package if package
    text
  else
    @doc.warnings << "Couldn't find entity in substitution table!"
    ''
  end
end

#convert_footnote(el, opts) ⇒ Object 



242
243
244
245
 
# File 'lib/newstile/converter/latex.rb', line 242

def convert_footnote(el, opts)
  @doc.conversion_infos[:packages] << 'fancyvrb'
  "\\footnote{#{inner(@doc.parse_infos[:footnotes][el.options[:name]][:content], opts).rstrip}}"
end

#convert_header(el, opts) ⇒ Object 



114
115
116
117
118
119
120
121
122
123
 
# File 'lib/newstile/converter/latex.rb', line 114

def convert_header(el, opts)
  type = HEADER_TYPES[el.options[:level]]
  if ((id = el.attr['id']) ||
      (@doc.options[:auto_ids] && (id = generate_id(el.options[:raw_text])))) &&
      (@doc.options[:toc_depth] <= 0 || el.options[:level] <= @doc.options[:toc_depth])
    "\\hypertarget{#{id}}{}\\#{type}{#{inner(el, opts)}}\\label{#{id}}\n\n"
  else
    "\\#{type}*{#{inner(el, opts)}}\n\n"
  end
end

#convert_hr(el, opts) ⇒ Object 



125
126
127
128
 
# File 'lib/newstile/converter/latex.rb', line 125

def convert_hr(el, opts)
  attrs = attribute_list(el)
  "\\begin{center}#{attrs}\n\\rule{3in}{0.4pt}\n\\end{center}#{attrs}\n"
end

#convert_html_element(el, opts) ⇒ Object 



156
157
158
159
160
161
162
163
164
165
 
# File 'lib/newstile/converter/latex.rb', line 156

def convert_html_element(el, opts)
  if el.value == 'i'
    "\\emph{#{inner(el, opts)}}"
  elsif el.value == 'b'
    "\\emph{#{inner(el, opts)}}"
  else
    @doc.warnings << "Can't convert HTML element"
    ''
  end
end

#convert_img(el, opts) ⇒ Object 



225
226
227
228
229
230
231
232
233
234
235
236
 
# File 'lib/newstile/converter/latex.rb', line 225

def convert_img(el, opts)
  if el.attr['src'] =~ /^(https?|ftps?):\/\//
    @doc.warnings << "Cannot include non-local image"
    ''
  elsif !el.options.attr['src'].empty?
    @doc.conversion_infos[:packages] << 'graphicx'
    "\\includegraphics{#{el.attr['src']}}"
  else
    @doc.warnings << "Cannot include image with empty path"
    ''
  end
end

#convert_li(el, opts) ⇒ Object 



144
145
146
 
# File 'lib/newstile/converter/latex.rb', line 144

def convert_li(el, opts)
  "\\item #{inner(el, opts).sub(/\n+\Z/, '')}\n"
end

#convert_math(el, opts) ⇒ Object 



537
538
539
540
541
542
543
544
545
546
547
548
 
# File 'lib/newstile/converter/latex.rb', line 537

def convert_math(el, opts)
  @doc.conversion_infos[:packages] += %w[amssymb amsmath amsthm amsfonts]
  if el.options[:category] == :block
    if el.value =~ /\A\s*\\begin\{/
      el.value
    else
      latex_environment('displaymath', el, el.value)
    end
  else
    "$#{el.value}$"
  end
end

#convert_p(el, opts) ⇒ Object 



70
71
72
73
74
75
76
 
# File 'lib/newstile/converter/latex.rb', line 70

def convert_p(el, opts)
  if el.children.size == 1 && el.children.first.type == :img && !(img = convert_img(el.children.first, opts)).empty?
    convert_standalone_image(el, opts, img)
  else
    "#{inner(el, opts)}\n\n"
  end
end

#convert_raw(el, opts) ⇒ Object 



247
248
249
250
251
252
253
 
# File 'lib/newstile/converter/latex.rb', line 247

def convert_raw(el, opts)
  if !el.options[:type] || el.options[:type].empty? || el.options[:type].include?('latex')
    el.value + (el.options[:category] == :block ? "\n" : '')
  else
    ''
  end
end

#convert_root(el, opts) ⇒ Object 



58
59
60
 
# File 'lib/newstile/converter/latex.rb', line 58

def convert_root(el, opts)
  inner(el, opts)
end

#convert_smart_quote(el, opts) ⇒ Object 



531
532
533
534
535
 
# File 'lib/newstile/converter/latex.rb', line 531

def convert_smart_quote(el, opts)
  res = SMART_QUOTE_SYMS[el.value]
  res += "{}" if (nel = opts[:parent].children[opts[:index]+1]) && nel.type == :smart_quote
  res
end

#convert_standalone_image(el, opts, img) ⇒ Object 



78
79
80
81
 
# File 'lib/newstile/converter/latex.rb', line 78

def convert_standalone_image(el, opts, img)
  attrs = attribute_list(el)
  "\\begin{figure}#{attrs}\n\\begin{center}\n#{img}\n\\end{center}\n\\caption{#{escape(el.children.first.attr['alt'])}}\n\\end{figure}#{attrs}\n"
end

#convert_strong(el, opts) ⇒ Object 



259
260
261
 
# File 'lib/newstile/converter/latex.rb', line 259

def convert_strong(el, opts)
  "\\textbf{#{inner(el, opts)}}"
end

#convert_table(el, opts) ⇒ Object 



179
180
181
182
183
 
# File 'lib/newstile/converter/latex.rb', line 179

def convert_table(el, opts)
  align = el.options[:alignment].map {|a| TABLE_ALIGNMENT_CHAR[a]}.join('|')
  attrs = attribute_list(el)
  "\\begin{tabular}{|#{align}|}#{attrs}\n\\hline\n#{inner(el, opts)}\\hline\n\\end{tabular}#{attrs}\n\n"
end

#convert_tbody(el, opts) ⇒ Object 



189
190
191
 
# File 'lib/newstile/converter/latex.rb', line 189

def convert_tbody(el, opts)
  inner(el, opts)
end

#convert_td(el, opts) ⇒ Object Also known as: convert_th 



201
202
203
 
# File 'lib/newstile/converter/latex.rb', line 201

def convert_td(el, opts)
  inner(el, opts)
end

#convert_text(el, opts) ⇒ Object 



66
67
68
 
# File 'lib/newstile/converter/latex.rb', line 66

def convert_text(el, opts)
  escape(el.value)
end

#convert_tfoot(el, opts) ⇒ Object 



193
194
195
 
# File 'lib/newstile/converter/latex.rb', line 193

def convert_tfoot(el, opts)
  "\\hline \\hline \n#{inner(el, opts)}"
end

#convert_thead(el, opts) ⇒ Object 



185
186
187
 
# File 'lib/newstile/converter/latex.rb', line 185

def convert_thead(el, opts)
  "#{inner(el, opts)}\\hline\n"
end

#convert_tr(el, opts) ⇒ Object 



197
198
199
 
# File 'lib/newstile/converter/latex.rb', line 197

def convert_tr(el, opts)
  el.children.map {|c| send("convert_#{c.type}", c, opts)}.join(' & ') + "\\\\\n"
end

#convert_typographic_sym(el, opts) ⇒ Object 



526
527
528
 
# File 'lib/newstile/converter/latex.rb', line 526

def convert_typographic_sym(el, opts)
  TYPOGRAPHIC_SYMS[el.value]
end

#convert_ul(el, opts) ⇒ Object Also known as: convert_ol 



130
131
132
133
134
135
136
137
 
# File 'lib/newstile/converter/latex.rb', line 130

def convert_ul(el, opts)
  if !@doc.conversion_infos[:has_toc] && (el.options[:ial][:refs].include?('toc') rescue nil)
    @doc.conversion_infos[:has_toc] = true
    '\tableofcontents'
  else
    latex_environment(el.type == :ul ? 'itemize' : 'enumerate', el, inner(el, opts))
  end
end

#convert_xml_comment(el, opts) ⇒ Object 



167
168
169
 
# File 'lib/newstile/converter/latex.rb', line 167

def convert_xml_comment(el, opts)
  el.value.split(/\n/).map {|l| "% #{l}"}.join("\n") + "\n"
end

#convert_xml_pi(el, opts) ⇒ Object Also known as: convert_html_doctype 



171
172
173
174
 
# File 'lib/newstile/converter/latex.rb', line 171

def convert_xml_pi(el, opts)
  @doc.warnings << "Can't convert XML PI/HTML document type"
  ''
end

#escape(str) ⇒ Object 



564
565
566
 
# File 'lib/newstile/converter/latex.rb', line 564

def escape(str)
  str.gsub(ESCAPE_RE) {|m| ESCAPE_MAP[m]}
end

#inner(el, opts) ⇒ Object 



47
48
49
50
51
52
53
54
55
56
 
# File 'lib/newstile/converter/latex.rb', line 47

def inner(el, opts)
  result = ''
  options = opts.dup.merge(:parent => el)
  el.children.each_with_index do |inner_el, index|
    options[:index] = index
    options[:result] = result
    result << send("convert_#{inner_el.type}", inner_el, options)
  end
  result
end

#latex_environment(type, el, text) ⇒ Object 



97
98
99
100
 
# File 'lib/newstile/converter/latex.rb', line 97

def latex_environment(type, el, text)
  attrs = attribute_list(el)
  "\\begin{#{type}}#{attrs}\n#{text.rstrip}\n\\end{#{type}}#{attrs}\n"
end