Mike Gerwitz
2fafc331a1
Non-attribute and non-empty start/end tags will have their whitespace as part of the produced span. This sets us up for a following change that will allow for deriving the name span from this span given a QName, which gives us a span that both represents the entire XIR token and allows deriving the element name. An accurate token span is necessary for parsing errors where an element was not expected, while an element name span is more appropriate for issues of grammar and semantic errors that deal not with the fact that an element was encountered, but _what_ element was encountered. DEV-7145 |
||
---|---|---|
.. | ||
test.rs |