Module: HtmlToPlainText

Included in:
Premailer
Defined in:
lib/premailer/html_to_plain_text.rb

Overview

Support functions for Premailer

Instance Method Summary collapse

Instance Method Details

#convert_to_text(html, line_length = 65, from_charset = 'UTF-8') ⇒ Object

Returns the text in UTF-8 format with all HTML tags removed

TODO: add support for DL, OL



10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
 
# File 'lib/premailer/html_to_plain_text.rb', line 10

def convert_to_text(html, line_length = 65, from_charset = 'UTF-8')
  txt = html

  # strip text ignored html. Useful for removing
  # headers and footers that aren't needed in the
  # text version
  txt.gsub!(/<!-- start text\/html -->.*?<!-- end text\/html -->/m, '')

  # replace images with their alt attributes
  # for img tags with "" for attribute quotes
  # with or without closing tag
  # eg. the following formats:
  # <img alt="" />
  # <img alt="">
  txt.gsub!(/<img.+?alt=\"([^\"]*)\"[^>]*\>/i, '\1')

  # for img tags with '' for attribute quotes
  # with or without closing tag
  # eg. the following formats:
  # <img alt='' />
  # <img alt=''>
  txt.gsub!(/<img.+?alt=\'([^\']*)\'[^>]*\>/i, '\1')

  # links
  txt.gsub!(/<a\s[^\n]*?href=["'](mailto:)?([^"']*)["'][^>]*>(.*?)<\/a>/im) do |s|
    if $3.empty?
      ''
    elsif $3.strip.downcase == $2.strip.downcase
      $3.strip
    else
      $3.strip + ' ( ' + $2.strip + ' )'
    end
  end

  # handle headings (H1-H6)
  txt.gsub!(/(<\/h[1-6]>)/i, "\n\\1") # move closing tags to new lines
  txt.gsub!(/[\s]*<h([1-6]+)[^>]*>[\s]*(.*)[\s]*<\/h[1-6]+>/i) do |s|
    hlevel = $1.to_i

    htext = $2
    htext.gsub!(/<br[\s]*\/?>/i, "\n") # handle <br>s
    htext.gsub!(/<\/?[^>]*>/i, '') # strip tags

    # determine maximum line length
    hlength = 0
    htext.each_line { |l| llength = l.strip.length; hlength = llength if llength > hlength }
    hlength = line_length if hlength > line_length

    case hlevel
      when 1   # H1, asterisks above and below
        htext = ('*' * hlength) + "\n" + htext + "\n" + ('*' * hlength)
      when 2   # H1, dashes above and below
        htext = ('-' * hlength) + "\n" + htext + "\n" + ('-' * hlength)
      else     # H3-H6, dashes below
        htext = htext + "\n" + ('-' * hlength)
    end

    "\n\n" + htext + "\n\n"
  end

  # wrap spans
  txt.gsub!(/(<\/span>)[\s]+(<span)/mi, '\1 \2')

  # lists -- TODO: should handle ordered lists
  txt.gsub!(/[\s]*(<li[^>]*>)[\s]*/i, '* ')
  # list not followed by a newline
  txt.gsub!(/<\/li>[\s]*(?![\n])/i, "\n")

  # paragraphs and line breaks
  txt.gsub!(/<\/p>/i, "\n\n")
  txt.gsub!(/<br[\/ ]*>/i, "\n")

  # strip remaining tags
  txt.gsub!(/<\/?[^>]*>/, '')

  # decode HTML entities
  he = HTMLEntities.new
  txt = he.decode(txt)

  # word wrap
  txt = word_wrap(txt, line_length)

  # remove linefeeds (\r\n and \r -> \n)
  txt.gsub!(/\r\n?/, "\n")

  # strip extra spaces
  txt.gsub!(/[ \t]*\302\240+[ \t]*/, " ") # non-breaking spaces -> spaces
  txt.gsub!(/\n[ \t]+/, "\n") # space at start of lines
  txt.gsub!(/[ \t]+\n/, "\n") # space at end of lines

  # no more than two consecutive newlines
  txt.gsub!(/[\n]{3,}/, "\n\n")

  # the word messes up the parens
  txt.gsub!(/\(([ \n])(http[^)]+)([\n ])\)/) do |s|
    ($1 == "\n" ? $1 : '' ) + '( ' + $2 + ' )' + ($3 == "\n" ? $1 : '' )
  end

  txt.strip
end

#word_wrap(txt, line_length) ⇒ Object

Taken from Rails' word_wrap helper (http://api.rubyonrails.org/classes/ActionView/Helpers/TextHelper.html#method-i-word_wrap)



112
113
114
115
116
 
# File 'lib/premailer/html_to_plain_text.rb', line 112

def word_wrap(txt, line_length)
  txt.split("\n").collect do |line|
    line.length > line_length ? line.gsub(/(.{1,#{line_length}})(\s+|$)/, "\\1\n").strip : line
  end * "\n"
end