Normally, the parser search indefinitely for the entity end
character (i.e. ';'). When this flag is set, the
characters '&', '<', '>', '"',
''', and any whitespace breaks the search for the entity
end, and the entity text is then ignored, i.e. treated as
data.