[antlr-interest] "Best practice" for contructing AST

Paul J. Lucas pauljlucas at mac.com
Fri Sep 17 16:33:22 PDT 2004


	If I have, say, an XML parser in ANTLR and I want to, say,
	construct an AST for an XML element attribute list, what's
	considered "best practice" for it?

	Specifically, given:

		<foo bar="1" baz="2">Hello</foo>

	Method 1 generated AST:

		ELEMENT
		    QNAME "foo"
		    ATTRIBUTE_LIST
		        ATTRIBUTE
			    QNAME "bar"
			    VALUE 1
		        ATTRIBUTE
			    QNAME "baz"
			    VALUE 2
		    CONTENT "Hello"

	Method 2 generated AST:

		ELEMENT
		    QNAME "foo"
		    ATTRIBUTE
		        QNAME "bar"
		        VALUE 1
		    ATTRIBUTE
		        QNAME "baz"
		        VALUE 2
		    CONTENT "Hello"

	Method 2 differs from Method 1 in that there's no
	ATTRIBUTE_LIST.  It's not strictly needed since one "knows"
	that attibutes always come before element content.

	But does eliding ATTRIBUTE_LIST generally make things easier or
	harder?

	- Paul



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/antlr-interest/

<*> To unsubscribe from this group, send an email to:
    antlr-interest-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 



More information about the antlr-interest mailing list