Go through a good bit of nonsense to get it successfully parsing our three test cases.

[dead/lwn-epub.git] / src / Main.hs

diff --git a/src/Main.hs b/src/Main.hs
index 37f89674eacbc115050f65f717a6bbe8530ddf71..c6ab443a9620d20f716576eb0c188702f8ad4b5d 100644 (file)
--- a/src/Main.hs
+++ b/src/Main.hs
@@ -3,23 +3,34 @@ module Main
 where
 
 import Data.Maybe (fromJust)
-import Text.HandsomeSoup (parseHtml)
+import Text.XML.HXT.Core -- (SysConfigList, IOStateArrow, XmlTree, readDocument)
 
 import Epublishable
 import LWN.ArticlePage
 import LWN.FullPage
-import XHTML
+
+my_read :: String -> IOStateArrow s b XmlTree
+my_read =
+  readDocument [ withValidate  no,
+                 withParseHTML yes,
+                 withInputEncoding utf8,
+                 withWarnings no ]
 
 main :: IO ()
 main = do
-  article_html <- readFile "test/fixtures/501317-article.html"
-  ioap <- parse $ parseHtml article_html
+  let article_html = my_read "test/fixtures/501317-article.html"
+  ioap <- parse article_html
   let article_page :: ArticlePage = fromJust $ ioap
   epublish article_page "single_article.epub"
 
-  page_html <- readFile "test/fixtures/500848-page.html"
-  ioap_f <- parse $ parseHtml page_html
+  let page_html = my_read "test/fixtures/500848-page.html"
+  ioap_f <- parse page_html
   let full_page :: FullPage = fromJust $ ioap_f
-  --putStrLn $ to_xhtml full_page
-  --epublish full_page "full_page.epub"
+  epublish full_page "full_page.epub"
+
+  let bigpage_html = my_read "test/fixtures/50844-bigpage.html"
+  ioap_bp <- parse bigpage_html
+  let bigpage :: FullPage = fromJust $ ioap_bp
+  epublish bigpage "bigpage.epub"
+
   putStrLn "Done."