OpenEuphoria

1. dot pre-processor

• Posted by jiri Mar 26, 2009
• 862 views

Thanks, Jeremy, for reviving David's pre-processor. Has it really taken 11 years? I hope it will take much less time to be hard-wired into the interpreter/compiler!

jiri

3. Re: dot pre-processor

• Posted by Kat Mar 26, 2009
• 883 views

And adding this DOT processor to the EUINC bat file i call now is easy. Only drawback is the trace window won't display what i actually wrote.

useless

5. Re: dot pre-processor

• Posted by jeremy (admin) Mar 26, 2009
• 855 views

It's not connected to OOP, however, if OOP comes to Euphoria, it may use a similar syntax, thus causing a conflict, that's all. I think we need discuss the hard details of OOP and Euphoria before making the interpreter do anything remotely OOP like. For instance, DOT syntax does this:

sequence name = "John" 
? name.length() 
 
-- Or 
enum FirstName, LastName 
sequence person = { "John", "Doe" } 
printf(1, "Hello, %s! Your last name is %s?", { person.FirstName, person.LastName }) 

Both of those could very much conflict with OOP, if it were to come to Euphoria.

Jeremy

7. Re: dot pre-processor

• Posted by DerekParnell (admin) Mar 26, 2009
• 853 views

I don't think that there is any necessary conflict. Therefore your phrase "thus causing a conflict" should be worded more like "thus might cause a conflict", IMO. However, I agree that a lot of discussion needs to happen before we add built-in OOP facilities into the language.

sequence name = "John" 
? name.length() 
 
-- Or 
enum FirstName, LastName 
sequence person = { "John", "Doe" } 
printf(1, "Hello, %s! Your last name is %s?", { person.FirstName, person.LastName }) 

It is possible, I suppose, but in these examples, the 'name' and 'person' objects are sequences and that could be detected at parser time and thus the correct IL code could be generated, and this would not be in conflict with dot notation used in other manners.