Eclipse Community Forums: Dynamic Languages Toolkit (DLTK)

Help

Home

Home » Eclipse Projects » Dynamic Languages Toolkit (DLTK) » Code completion and broken AST

Show: Today's Messages :: Show Polls :: Message Navigator

Code completion and broken AST [message #541968]

Wed, 23 June 2010 12:17

Jan F.

Messages: 6
Registered: June 2010

Junior Member

What is the DLTK's typical approach for working with AST that is broken by interim edits, such as typing "this." to invoke code completion?

Is there a concept of using the previous cached AST, or should the parser rather be made to be resistent to any interim editing and provide the structure down to method/fields/lcoal vars in all possible cases?

Report message to a moderator

Re: Code completion and broken AST [message #543524 is a reply to message #541968]

Tue, 29 June 2010 16:46

Alex Panchenko

Messages: 342
Registered: July 2009

Senior Member

Hi Jan,

The typical approach is making the parser tolerant to the errors, so it can recover after the error and continue on the next statement, etc.

Regards,
Alex

Report message to a moderator