In the RS spec we say:

In addition, when processing XML documents, a reading system MUST NOT resolve 
[external identifiers](https://www.w3.org/TR/xml/#NT-ExternalID) in DOCTYPE, 
ENTITY and NOTATION declarations [[xml](https://www.w3.org/TR/epub-rs-33/#bib-xml)].

However, EPUB check will report an error for such content. For instance, when run against the test for this it will produce this error:

EPUB/content_001.xhtml:4, 36: External entities are not allowed in EPUB v3 documents. 
   External entity declaration found: xxe.
EPUB/foo.xhtml:1, 7: Error while parsing file: element "span" not allowed here; 
   expected element "html" (with xmlns="https://www.w3.org/1999/xhtml")

It is unclear if a RS that rejects this content is conformant. Must it process the document and display the &xxe; in text? There may be other cases where EPUB check reports an error but the spec gives guidance on how to process the document. For content that comes from a publisher, it seems like failing is the best thing to do, as it forces a fix to the content. Should we add some sort of blanket statement that allows an RS to reject such content? And do we need to specify that that is an acceptable outcome for the tests?