```
public class SemanticPipeline
extends Object
```
Do as much semantic checking as we can and fill in grammar with rules, actions, and token definitions. The only side effects are in the grammar passed to process(). We consume a bunch of memory here while we build up data structures to perform checking, but all of it goes away after this pipeline object gets garbage collected. After this pipeline finishes, we can be sure that the grammar is syntactically correct and that it's semantically correct enough for us to attempt grammar analysis. We have assigned all token types. Note that imported grammars bring in token and rule definitions but only the root grammar and any implicitly created lexer grammar get their token definitions filled up. We are treating the imported grammars like includes. The semantic pipeline works on root grammars (those that do the importing, if any). Upon entry to the semantic pipeline, all imported grammars should have been loaded into delegate grammar objects with their ASTs created. The pipeline does the BasicSemanticChecks on the imported grammar before collecting symbols. We cannot perform the simple checks such as undefined rule until we have collected all tokens and rules from the imported grammars into a single collection.

Constructor Detail
- SemanticPipeline
```
public SemanticPipeline(Grammar g)
```

Class SemanticPipeline