Module: RungerEmailReplyTrimmer

Defined in:
lib/runger_email_reply_trimmer.rb,
lib/runger_email_reply_trimmer/version.rb

Constant Summary collapse

DELIMITER =
'd'
EMBEDDED =
'b'
EMPTY =
'e'
EMAIL_HEADER =
'h'
QUOTE =
'q'
SIGNATURE =
's'
TEXT =
't'
VERSION =
'0.3.0'

Class Method Summary collapse

Class Method Details

.compute_elided(text, lines) ⇒ Object



205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
# File 'lib/runger_email_reply_trimmer.rb', line 205

def compute_elided(text, lines)
  elided = []

  t = 0
  l = 0

  while t < text.size
    while l < lines.size && text[t] == lines[l]
      t += 1
      l += 1
    end
    elided << text[t]
    t += 1
  end

  elided.join("\n").strip
end

.extract_embedded_email(text) ⇒ Object



137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
# File 'lib/runger_email_reply_trimmer.rb', line 137

def extract_embedded_email(text)
  return if text.nil? || text =~ /\A[[:space:]]*\z/m

  # do some cleanup
  preprocess!(text)

  # from now on, we'll work on a line-by-line basis
  lines = text.split("\n")

  # identify content of each lines
  pattern = lines.map { |l| identify_line_content(l) }.join

  if (index = pattern =~ /(?:h[eqd]*?){3,}[tq]/)
    embedded = lines[index..].join("\n").strip
  elsif (index = pattern =~ /b(?:[eqd]*){3,}[tq]/)
    # Exception for email clients (macOS / iOS) which embed fwd emails in quotes.
    embedded = lines[index + 1..].map { |l| l.gsub(/^>\s*/, '') }.join("\n").strip
  end

  if index
    before = lines[0...(pattern[0...index] =~ /e*(b[eqd]*|b*[ed]*)$/)].join("\n").strip
    [embedded, before]
  end
end

.identify_line_content(line) ⇒ Object



22
23
24
25
26
27
28
29
30
31
# File 'lib/runger_email_reply_trimmer.rb', line 22

def identify_line_content(line)
  return EMPTY        if EmptyLineMatcher.match?(line)
  return DELIMITER    if DelimiterMatcher.match?(line)
  return SIGNATURE    if SignatureMatcher.match?(line)
  return EMBEDDED     if EmbeddedEmailMatcher.match?(line)
  return EMAIL_HEADER if EmailHeaderMatcher.match?(line)
  return QUOTE        if QuoteMatcher.match?(line)

  TEXT
end

.is_reply_at_end?(pattern) ⇒ Boolean

Returns:

  • (Boolean)


223
224
225
# File 'lib/runger_email_reply_trimmer.rb', line 223

def is_reply_at_end?(pattern)
  pattern =~ /^b[^t]+t[et]*$/
end

.preprocess!(text) ⇒ Object



162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
# File 'lib/runger_email_reply_trimmer.rb', line 162

def preprocess!(text)
  # normalize line endings
  text.gsub!("\r\n", "\n")

  # remove PGP markers
  text.gsub!(/\A-----BEGIN PGP SIGNED MESSAGE-----\n(?:Hash: \w+)?\s+/i, '')
  text.gsub!(/^-----BEGIN PGP SIGNATURE-----$[\s\S]+^-----END PGP SIGNATURE-----/, '')

  # remove unsubscribe links
  text.gsub!(/^Unsubscribe: .+@.+(\n.+http:.+)?\s*\z/i, '')

  # remove alias-style quotes marker
  text.gsub!(/^.*>{5} "[^"\n]+" == .+ writes:/, '')

  # change enclosed-style quotes format
  text.gsub!(/^>>> ?(.+) ?>>>$\n([\s\S]+?)\n^<<< ?\1 ?<<<$/) {
    ::Regexp.last_match(2).gsub(/^/, '> ')
  }
  text.gsub!(/^>{4,}[[:blank:]]*$\n([\s\S]+?)\n^<{4,}[[:blank:]]*$/) {
    ::Regexp.last_match(1).gsub(/^/, '> ')
  }

  # fix all quotes formats
  text.gsub!(/^((?:[[:blank:]]*[[:alpha:]]*[>|])+)/) {
    ::Regexp.last_match(1).gsub(/([[:alpha:]]+>|\|)/, '>')
  }

  # fix embedded email markers that might span over multiple lines
  (
    EmbeddedEmailMatcher::ON_DATE_SOMEONE_WROTE_REGEXES +
    EmbeddedEmailMatcher::SOMEONE_WROTE_ON_DATE_REGEXES +
    EmbeddedEmailMatcher::DATE_SOMEONE_WROTE_REGEXES +
    [EmbeddedEmailMatcher::DATE_SOMEONE_EMAIL_REGEX]
  ).each do |r|
    text.gsub!(r) do |m|
      m.count("\n") > 4 ? m : m.gsub(/\n+[[:space:]]*/, ' ')
    end
  end

  # remove leading/trailing whitespaces
  text.strip!
end

.trim(text, split = false) ⇒ Object



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
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
# File 'lib/runger_email_reply_trimmer.rb', line 33

def trim(text, split = false)
  return if text.nil? || text =~ /\A[[:space:]]*\z/m

  # do some cleanup
  preprocess!(text)

  # from now on, we'll work on a line-by-line basis
  lines = text.split("\n")
  lines_dup = lines.dup

  # identify content of each lines
  pattern = lines.map { |l| identify_line_content(l) }.join

  # remove everything after the first delimiter
  if pattern =~ /d/
    index = pattern =~ /d/
    pattern = pattern[0...index]
    lines = lines[0...index]
  end

  # remove all mobile signatures
  while pattern =~ /s/
    index = pattern =~ /s/
    pattern.slice!(index)
    lines.slice!(index)
  end

  # when the reply is at the end of the email
  if is_reply_at_end?(pattern)
    index = pattern =~ /t[et]*$/
    pattern = ''
    lines = lines[index..]
  end

  # if there is an embedded email marker, not followed by a quote
  # then take everything up to that marker
  if pattern =~ /te*b[^q]*$/
    index = pattern =~ /te*b[^q]*$/
    pattern = pattern[0..index]
    lines = lines[0..index]
  end

  # if there is an embedded email marker, followed by a huge quote
  # then take everything up to that marker
  if pattern =~ /te*b[eqbh]*([te]*)$/ && ::Regexp.last_match(1).count('t') < 7
    index = pattern =~ /te*b[eqbh]*[te]*$/
    pattern = pattern[0..index]
    lines = lines[0..index]
  end

  # if there is some text before a huge quote ending the email,
  # then remove the quote
  if pattern =~ /t?e*[qbe]+$/
    index = pattern =~ /t?e*[qbe]+$/
    pattern = pattern[0..index]
    lines = lines[0..index]
  end

  # if there still are some embedded email markers, just remove them
  while pattern =~ /b/
    index = pattern =~ /b/
    pattern.slice!(index)
    lines.slice!(index)
  end

  # fix email headers when they span over multiple lines
  if pattern =~ /h+[hte]+h+e/
    index = pattern =~ /h+[hte]+h+e/
    size = pattern[/h+[hte]+h+e/].size
    size.times.each { |s| pattern[index + s] = EMAIL_HEADER }
  end

  # if there are at least 3 consecutive email headers,
  # take everything up to these headers
  if pattern =~ /t[eq]*h{3,}/
    index = pattern =~ /t[eq]*h{3,}/
    pattern = pattern[0..index]
    lines = lines[0..index]
  end

  # if there still are some email headers, just remove them
  while pattern =~ /h/
    index = pattern =~ /h/
    pattern.slice!(index)
    lines.slice!(index)
  end

  # remove trailing quotes when there's at least one line of text
  if pattern =~ /t/ && pattern =~ /[eq]+$/
    index = pattern =~ /[eq]+$/
    pattern = pattern[0...index]
    lines = lines[0...index]
  end

  # results
  trimmed = lines.join("\n").strip

  if split
    [trimmed, compute_elided(lines_dup, lines)]
  else
    trimmed
  end
end