private void ParseText_DoWork(object sender, DoWorkEventArgs e)
{
Lexicon.Clear();
var expElementPattern = new Regex($"({PunctuationPattern.ToString()})|({WordPattern.ToString()})");
var whiteSpacePattern = new Regex(@"[\s\n\r]+", RegexOptions.Singleline | RegexOptions.Multiline);
var worker = sender as BackgroundWorker;
var text = e.Argument as string;
int progress = 0;
string state = string.Empty;
// TODO: handling of paragraph breaks and section headers, etc
foreach (Match p in ParagraphPattern.Matches(text))
{
string paragraphText = p.Value.Trim();
paragraphText = whiteSpacePattern.Replace(paragraphText, " ");
var paragraph = new Lx.Discourse();
Text.Discourse.AddLast(paragraph);
foreach (Match l in LinePattern.Matches(paragraphText))
{
//store line, section up into words and punctuation
string cleanedLine = l.Value.Trim();
//cleanedLine = whiteSpacePattern.Replace(cleanedLine, " ");
state = cleanedLine;
var expression = new Lx.Expression(cleanedLine);
paragraph.Expressions.AddLast(expression);
foreach (Match m in expElementPattern.Matches(expression.Graph))
{
if (m.Groups.Count > 0)
{
// string m => List<Glyphs>
var glyphs = Script.AddGlyphs(m.Value.ToCharArray());
// List<Glyph> => List<Grapheme>
// Pre-analysis, graphemes are 1:1 with glyphs
var graphemes = Orthography.AddGraphemes(glyphs);
// List<Grapheme> => Morpheme
if (string.IsNullOrEmpty(m.Groups[1].Value))
{
//var morph = Text.Lexicon.Add(m.Groups[2].Value);
//morph.GraphemeChain.Add(Lx.SegmentChain<Lx.Grapheme>.NewSegmentChain(graphemes));
Lx.Morpheme morph = Text.Lexicon.Add(graphemes);
expression.Sequence.AddLast(morph);
}
else
{
expression.Sequence.AddLast(Text.Paralexicon.Add(m.Groups[1].Value));
}
}
}
worker.ReportProgress(++progress, state);
}
}
UpdateLocalLexicon();
}
