HTML reader now recognizes DocBook block and inline tags.

It was always possible to include raw DocBook tags in a markdown
document, but now pandoc will be able to distinguish block from
inline tags and behave accordingly. Thus, for example,

    <sidebar>
    hello
    </sidebar>

will not be wrapped in `<para>` tags.

2 files changed, 26 insertions, 6 deletions

diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs
index 2ad2c8e0f..70ea24680 100644
--- a/src/Text/Pandoc/Readers/HTML.hs
+++ b/src/Text/Pandoc/Readers/HTML.hs
@@ -503,16 +503,35 @@ blockHtmlTags = ["address", "blockquote", "body", "center", "dir", "div",
                  "dt", "frameset", "li", "tbody", "td", "tfoot",
                  "th", "thead", "tr", "script", "style"]
 
+-- We want to allow raw docbook in markdown documents, so we
+-- include docbook block tags here too.
+blockDocBookTags :: [String]
+blockDocBookTags = ["calloutlist", "bibliolist", "glosslist", "itemizedlist",
+                    "orderedlist", "segmentedlist", "simplelist",
+                    "variablelist", "caution", "important", "note", "tip",
+                    "warning", "address", "literallayout", "programlisting",
+                    "programlistingco", "screen", "screenco", "screenshot",
+                    "synopsis", "example", "informalexample", "figure",
+                    "informalfigure", "table", "informaltable", "para",
+                    "simpara", "formalpara", "equation", "informalequation",
+                    "figure", "screenshot", "mediaobject", "qandaset",
+                    "procedure", "task", "cmdsynopsis", "funcsynopsis",
+                    "classsynopsis", "blockquote", "epigraph", "msgset",
+                    "sidebar"]
+
+blockTags :: [String]
+blockTags = blockHtmlTags ++ blockDocBookTags
+
 isInlineTag :: Tag String -> Bool
-isInlineTag t = tagOpen (`notElem` blockHtmlTags) (const True) t ||
-                tagClose (`notElem` blockHtmlTags) t ||
+isInlineTag t = tagOpen (`notElem` blockTags) (const True) t ||
+                tagClose (`notElem` blockTags) t ||
                 tagComment (const True) t
 
 isBlockTag :: Tag String -> Bool
 isBlockTag t = tagOpen (`elem` blocktags) (const True) t ||
                tagClose (`elem` blocktags) t ||
                tagComment (const True) t
-                 where blocktags = blockHtmlTags ++ eitherBlockOrInline
+                 where blocktags = blockTags ++ eitherBlockOrInline
 
 isTextTag :: Tag String -> Bool
 isTextTag = tagText (const True)
@@ -547,8 +566,8 @@ t `closes` t2 |
    t `elem` ["h1","h2","h3","h4","h5","h6","dl","ol","ul","table","div","p"] &&
    t2 `elem` ["h1","h2","h3","h4","h5","h6","p" ] = True -- not "div"
 t1 `closes` t2 |
-   t1 `elem` blockHtmlTags &&
-   t2 `notElem` (blockHtmlTags ++ eitherBlockOrInline) = True
+   t1 `elem` blockTags &&
+   t2 `notElem` (blockTags ++ eitherBlockOrInline) = True
 _ `closes` _ = False
 
 --- parsers for use in markdown, textile readers
diff --git a/src/Text/Pandoc/Writers/Docbook.hs b/src/Text/Pandoc/Writers/Docbook.hs
index 29c042cf9..9f8b921e7 100644
--- a/src/Text/Pandoc/Writers/Docbook.hs
+++ b/src/Text/Pandoc/Writers/Docbook.hs
@@ -255,7 +255,8 @@ inlineToDocbook _ EnDash = text "–"
 inlineToDocbook _ (Code _ str) = 
   inTagsSimple "literal" $ text (escapeStringForXML str)
 inlineToDocbook opts (Math _ str) = inlinesToDocbook opts $ readTeXMath str
-inlineToDocbook _ (RawInline _ _) = empty
+inlineToDocbook _ (RawInline f x) | f == "html" || f == "docbook" = text x
+                                  | otherwise                     = empty
 inlineToDocbook _ LineBreak = inTagsSimple "literallayout" empty
 inlineToDocbook _ Space = space
 inlineToDocbook opts (Link txt (src, _)) =