diff --git a/Text/HTML/DOM.hs b/Text/HTML/DOM.hs
--- a/Text/HTML/DOM.hs
+++ b/Text/HTML/DOM.hs
@@ -43,6 +43,10 @@
mx <- await
case fmap (entities . fmap' (decodeUtf8With lenientDecode)) mx of
Nothing -> closeStack stack
+
+ -- Ignore processing instructions (or pseudo-instructions)
+ Just (TS.TagOpen local _ _) | "?" `T.isPrefixOf` local -> go stack
+
Just (TS.TagOpen local attrs isClosed) -> do
let name = toName local
attrs' = map (toName *** return . XT.ContentText) attrs
diff --git a/html-conduit.cabal b/html-conduit.cabal
--- a/html-conduit.cabal
+++ b/html-conduit.cabal
@@ -1,5 +1,5 @@
Name: html-conduit
-Version: 1.1.0.4
+Version: 1.1.0.5
Synopsis: Parse HTML documents using xml-conduit datatypes.
Description: This package uses tagstream-conduit for its parser. It automatically balances mismatched tags, so that there shouldn't be any parse failures. It does not handle a full HTML document rendering, such as adding missing html and head tags.
Homepage: https://github.com/snoyberg/xml
diff --git a/test/main.hs b/test/main.hs
--- a/test/main.hs
+++ b/test/main.hs
@@ -54,6 +54,20 @@
]
]
in H.parseLBS html @?= doc
+ it "XHTML with doctype and \n\nfooHello World
"
+ doc = X.Document (X.Prologue [] Nothing []) root []
+ root = X.Element "html" Map.empty
+ [ X.NodeElement $ X.Element "head" Map.empty
+ [ X.NodeElement $ X.Element "title" Map.empty
+ [X.NodeContent "foo"]
+ ]
+ , X.NodeElement $ X.Element "body" Map.empty
+ [ X.NodeElement $ X.Element "p" Map.empty
+ [X.NodeContent "Hello World"]
+ ]
+ ]
+ in H.parseLBS html @?= doc
it "HTML" $
let html = "foo
Hello World
"
doc = X.Document (X.Prologue [] Nothing []) root []