diff --git a/epub-metadata.cabal b/epub-metadata.cabal
--- a/epub-metadata.cabal
+++ b/epub-metadata.cabal
@@ -1,5 +1,5 @@
 name:                epub-metadata
-version:             4.0
+version:             4.1
 cabal-version:       >= 1.8
 build-type:          Simple
 license:             BSD3
@@ -9,7 +9,7 @@
 maintainer:          Dino Morelli <dino@ui3.info>
 stability:           experimental
 homepage:            http://ui3.info/d/proj/epub-metadata.html
-synopsis:            Library for parsing epub document metdata
+synopsis:            Library for parsing epub document metadata
 description:         Library for parsing and manipulating epub document metadata. Supports epub versions 2 and 3.
 
                      This library was constructed by studying the IDPF specifications for epub documents found here <http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm> and here <http://www.idpf.org/epub/30/spec/epub30-publications.html>
@@ -55,6 +55,7 @@
                      Codec.Epub.Parse.Refinements
                      Codec.Epub.Parse.Spine
                      Codec.Epub.Parse.Util
+                     Codec.Epub.Util
    build-depends:    base >= 3 && < 5,
                      bytestring,
                      containers,
@@ -65,7 +66,7 @@
                      regex-compat,
                      zip-archive
    hs-source-dirs:   src
-   ghc-options:      -Wall
+   ghc-options:      -Wall -fsimpl-tick-factor=200
 
 test-suite           test-epub-metadata
    type:             exitcode-stdio-1.0
@@ -80,4 +81,4 @@
                      regex-compat,
                      zip-archive
    hs-source-dirs:   src testsuite
-   ghc-options:      -Wall
+   ghc-options:      -Wall -fsimpl-tick-factor=200
diff --git a/src/Codec/Epub/Data/Metadata.hs b/src/Codec/Epub/Data/Metadata.hs
--- a/src/Codec/Epub/Data/Metadata.hs
+++ b/src/Codec/Epub/Data/Metadata.hs
@@ -31,7 +31,7 @@
 {- | Refinements represent meta tags within the metadata section
    that refine other tags. These are used during the parsing phase
    and are discarded as their information is slotted into the data
-   they refine (the types below like Creator, Title, etc..)
+   they refine (CreatorS, TitleS, IdentifierS, etc..)
 
    This is specific to epub3
 -}
diff --git a/src/Codec/Epub/IO.hs b/src/Codec/Epub/IO.hs
--- a/src/Codec/Epub/IO.hs
+++ b/src/Codec/Epub/IO.hs
@@ -32,7 +32,9 @@
 import Text.XML.HXT.Arrow.XmlArrow ( getAttrValue, hasName, isElem )
 import Text.XML.HXT.Arrow.XmlState ( no, runX, withValidate )
 
+import Codec.Epub.Util
 
+
 locateRootFile :: (MonadIO m, MonadError String m) =>
    FilePath -> String -> m FilePath
 locateRootFile containerPath' containerDoc = do
@@ -84,7 +86,10 @@
    -}
    containerDoc <- fileFromArchive containerPath archive
 
-   rootPath <- locateRootFile containerPath containerDoc
+   let cleanedContents = removeIllegalStartChars . removeEncoding
+         . removeDoctype $ containerDoc
+
+   rootPath <- locateRootFile containerPath cleanedContents
 
    -- Now that we have the path to the .opf file, extract it
    rootContents <- fileFromArchive rootPath archive
diff --git a/src/Codec/Epub/Parse.hs b/src/Codec/Epub/Parse.hs
--- a/src/Codec/Epub/Parse.hs
+++ b/src/Codec/Epub/Parse.hs
@@ -22,7 +22,6 @@
 
 import Control.Arrow.ListArrows
 import Control.Monad.Error
-import Text.Regex
 import Text.XML.HXT.Arrow.Namespace ( propagateNamespaces )
 import Text.XML.HXT.Arrow.XmlState ( no, runX, withValidate )
 import Text.XML.HXT.Arrow.XmlState.TypeDefs
@@ -40,26 +39,7 @@
 import Codec.Epub.Parse.Package
 import Codec.Epub.Parse.Refinements
 import Codec.Epub.Parse.Spine
-
-
-{- An evil hack to remove *ILLEGAL* characters before the XML
-   declaration. Why do people write software that does this?
-   Can't they follow directions?
--}
-removeIllegalStartChars :: String -> String
-removeIllegalStartChars = dropWhile (/= '<')
-
-
--- An evil hack to remove encoding from the document
-removeEncoding :: String -> String
-removeEncoding = flip (subRegex 
-   (mkRegexWithOpts " +encoding=\"UTF-8\"" False False)) ""
-
-
--- An evil hack to remove any \<!DOCTYPE ...\> from the document
-removeDoctype :: String -> String
-removeDoctype = flip (subRegex 
-   (mkRegexWithOpts "<!DOCTYPE [^>]*>" False True)) ""
+import Codec.Epub.Util
 
 
 {- Extract the epub OPF Package data contained in the supplied 
diff --git a/src/Codec/Epub/Util.hs b/src/Codec/Epub/Util.hs
new file mode 100644
--- /dev/null
+++ b/src/Codec/Epub/Util.hs
@@ -0,0 +1,31 @@
+-- Copyright: 2010-2013 Dino Morelli
+-- License: BSD3 (see LICENSE)
+-- Author: Dino Morelli <dino@ui3.info>
+
+{- | Utility functions shared by modules that need to read the
+   contents of XML documents.
+-}
+module Codec.Epub.Util
+   where
+
+import Text.Regex
+
+
+{- | An evil hack to remove *ILLEGAL* characters before the XML
+   declaration. Why do people write software that does this?
+   Can't they follow directions?
+-}
+removeIllegalStartChars :: String -> String
+removeIllegalStartChars = dropWhile (/= '<')
+
+
+-- | An evil hack to remove encoding from the document
+removeEncoding :: String -> String
+removeEncoding = flip (subRegex 
+   (mkRegexWithOpts " +encoding=\"UTF-8\"" False False)) ""
+
+
+-- | An evil hack to remove any \<!DOCTYPE ...\> from the document
+removeDoctype :: String -> String
+removeDoctype = flip (subRegex 
+   (mkRegexWithOpts "<!DOCTYPE [^>]*>" False True)) ""