DynamicLexicalPattern Types and Details

Hi Ryan,

The DynamicLexicalPatternGroup object has a PatternType property that lets you say whether it is Explciit or Regex.  Explicit means that the exact text should be used as you specify it in patterns (with the pattern group's case sensitivity option taken into account).  Regex means interpret the pattern as a regular expression.

So your "FROM" should only be added to an Explicit pattern group.  The Identifier pattern should only be added to a Regex pattern group.  There is a full guide in the documentation on the regex pattern syntax.

For your desired pattern, if it was in a regex pattern group, you could do something like:

{Word}\.{Word}

Ahh that makes perfect sense, thanks! However with the "{Word}\.{Word}" pattern, it doesn't seem to work. Depending on the pattern, it either doesn't work, or it blows up on load. Weird. I've tried bunches of different variations, and none work. Any other suggestions?

lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern(@"({Word})(\.)({Word})"));

lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern(@"({Word})({\.})({Word})"));
lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern(@"{Word}\.{Word}"));
lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern("({Word}).({Word})"));
lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern("[a-z0-9_-]\\.[a-z0-9_-]"));

It just doesn't seem to want to accept the period between two words(something.something).

Thanks,

Ryan

Sorry I think I forgot to put the quantifiers in there.  That would be a problem since it would just look for one word character, then period, then another word character.  Try this instead:

({Word})+ \. ({Word})+

Yeah I'm not sure what I'm doing wrong anymore, and I've done regex in the past....maybe it doesn't like my '@' at the beginning of the string? It does require it though to escape the backslash..it still fails by giving the red squiggily on "something.something"

lexicalPatternGroup.Patterns.Add(new DynamicLexicalPattern(@"({Word})+ \. ({Word})+"));

Thanks,

Ryan

Hi Ryan,

I'm not sure without being able to have something to debug.  Did you try using the Language Designer to build your dynamic lexer?  The benefit of doing it there is that you don't need to worry about escaping C# characters and it even has a Live Test tab where you can see how it's tokenizing text.  Maybe that would help you out here.