Class: Eiwaji::LexerWidget

Inherits:
Qt::DockWidget
  • Object
show all
Defined in:
lib/lexer_widget.rb

Defined Under Namespace

Modules: Underscore

Constant Summary collapse

POS_IGNORE =

parts-of-speech to not provide hyperlinks for

[Ve::PartOfSpeech::Symbol]
HISTORY_STRING_LENGTH =
30
MAX_HISTORY_ITEMS =
50

Instance Method Summary collapse

Constructor Details

#initialize(parent) ⇒ LexerWidget

Returns a new instance of LexerWidget.



15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
# File 'lib/lexer_widget.rb', line 15

def initialize(parent)
  super(parent)
  
  @ui = Ui_LexerWidget.new
  @ui.setupUi(self)

  part_of_speech_colors

  connect(@ui.lexerTextBrowser, SIGNAL('anchorClicked(QUrl)'),
          self, SLOT('wordClicked(QUrl)'))

  @ui.historyBox.insertItem(0, "...")
  connect(@ui.historyBox, SIGNAL('currentIndexChanged(int)'), self, SLOT('historyItemChanged(int)'))
  connect(@ui.buttonNext, SIGNAL('clicked()'), self, SLOT('historyNext()'))
  connect(@ui.buttonPrev, SIGNAL('clicked()'), self, SLOT('historyPrev()'))
end

Instance Method Details

#event(event) ⇒ Object



36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
# File 'lib/lexer_widget.rb', line 36

def event(event)
  # provide tooltips for word part-of-speech when mousing over hrefs
  if event.type == Qt::Event::ToolTip
    cursor = @ui.lexerTextBrowser.cursorForPosition(event.pos)
    cursor.select(Qt::TextCursor::WordUnderCursor)
    fragment = cursor.selection()

    # works in this case, since the input is clearly defined in wordToHtml
    hrefRegex = /<a\s+(?:[^>]*?\s+)?href="([^"]*)"/
    resultsIndex = hrefRegex.match(fragment.toHtml)

    if not resultsIndex or resultsIndex.size != 2
      Qt::ToolTip.hideText
    else
      word = @lexer_results[resultsIndex[1].to_i]
      Qt::ToolTip.showText(event.globalPos(), word.lemma + " " + word.part_of_speech.name)
    end
  end
  super(event)
end

#historyItemChanged(index) ⇒ Object



57
58
59
60
61
# File 'lib/lexer_widget.rb', line 57

def historyItemChanged(index)
  return if index == 0
  text = @ui.historyBox.itemData(index).value
  lexText(text)
end

#historyNextObject



70
71
72
73
74
# File 'lib/lexer_widget.rb', line 70

def historyNext
  index = @ui.historyBox.currentIndex - 1
  index = 1 if index < 1
  @ui.historyBox.setCurrentIndex(index)
end

#historyPrevObject



63
64
65
66
67
68
# File 'lib/lexer_widget.rb', line 63

def historyPrev
  index = @ui.historyBox.currentIndex + 1
  index += 1 if @ui.historyBox.currentIndex == 0
  index = @ui.historyBox.count - 1 if index >= @ui.historyBox.count - 1
  @ui.historyBox.setCurrentIndex(index)
end

#lexText(text, append_to_history = false) ⇒ Object



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
# File 'lib/lexer_widget.rb', line 76

def lexText(text, append_to_history=false)
  text = text.force_encoding("UTF-8")

  if append_to_history
    @ui.historyBox.setCurrentIndex(0)
    truncated_text = text.size < HISTORY_STRING_LENGTH ? text : text[0..HISTORY_STRING_LENGTH-1] + "..."
    @ui.historyBox.insertItem(1, truncated_text, Qt::Variant.new(text)) 
    @ui.historyBox.removeItem(MAX_HISTORY_ITEMS+1) if @ui.historyBox.count > MAX_HISTORY_ITEMS+1
  end

  lines = text.split("\n")
  html_lines = []
  @lexer_results = Hash.new
  i = 0

  lines.each do |line|
    words = Ve.in(:ja).words(line)
    html_words = []

    words.each do |word|
      @lexer_results[i] = word
      html_words << wordToHtml(word, i)
      i = i + 1
    end
    html_lines << "<div style='margin-bottom: 20px'>" + html_words.join(' ') + "<\div>"
  end

  html = html_lines.join("")

  @ui.lexerTextBrowser.setText(html)
end

#part_of_speech_colorsObject

each symbol represents the class name of a Ve::PartOfSpeech



133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
# File 'lib/lexer_widget.rb', line 133

def part_of_speech_colors
  @part_of_speech_colors ||= {
    :verb => "#808080",
    :noun => "#000000",
    :proper_noun => "#0000FF",
    :symbol => "#000000",
    :postposition => "#FF0000",
    :pronoun => "#008080",
    :adjective => "#008000",
    :adverb => "#43C6DB",
    :conjunction => "#FFA500",
    :background => "#FFFFFF",
    :default => "#FF00FF"
  }
end

#searchWord(ref) ⇒ Object



112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
# File 'lib/lexer_widget.rb', line 112

def searchWord(ref)
  raise "No results from the lexer." if @lexer_results.empty?
  word = @lexer_results[ref]

  # Word lemma
  # since the lemmas of Words like "1話" become "*話", remove extra '*'s
  lemma = word.lemma.tr('*','')

  # raw token, drops "する" and other extra parts
  query = word.tokens[0][:lemma]
  query = lemma if query == "*"

  if Eiwaji::DEBUG
    p word, query, lemma
  end

  # search on the query and sort by the results most similar to the lemma
  parent.search(query, lemma)
end

#textBrowserObject



32
33
34
# File 'lib/lexer_widget.rb', line 32

def textBrowser
  @ui.textBrowser
end

#wordClicked(url) ⇒ Object



108
109
110
# File 'lib/lexer_widget.rb', line 108

def wordClicked(url)
  searchWord(url.path.to_i)
end

#wordToHtml(word, index) ⇒ Object

converts a Ve::Word and its index in the search results to a String representation



150
151
152
153
154
155
156
157
158
159
160
# File 'lib/lexer_widget.rb', line 150

def wordToHtml(word, index)
  raw = word.word
  pos = word.part_of_speech
  color = @part_of_speech_colors[pos.name.gsub(' ', '_').underscore.to_sym] || @part_of_speech_colors[:default]
  
  if POS_IGNORE.include? pos
    raw = "<font style='color: #{color}'>" + raw + "</font>"
  else
    raw = "<a href=\'#{index}\' style='color: #{color}'>" + raw + "</a>"
  end
end