X-Git-Url: http://gitweb.michael.orlitzky.com/?a=blobdiff_plain;f=src%2FLWN%2FHTTP.hs;h=ef4fc5aeebf14cbefd25be3cf7084052eeac4025;hb=d7eb43dabd083ff2e12e9cfbf9bc2b6fcaa55e0b;hp=a8a1980f273cfdc352911bb7bcd3b55b58137cdf;hpb=f3321e2ce7d7645ad562dc8f6620bfd561edc75d;p=dead%2Flwn-epub.git

diff --git a/src/LWN/HTTP.hs b/src/LWN/HTTP.hs
index a8a1980..ef4fc5a 100644
--- a/src/LWN/HTTP.hs
+++ b/src/LWN/HTTP.hs
@@ -3,8 +3,10 @@
 module LWN.HTTP
 where
 
+import Control.Concurrent.ParallelIO (parallel)
 import qualified Data.ByteString as B (hPut)
 import qualified Data.Map as Map (Map, empty, insert)
+import Data.Maybe (fromJust, isNothing)
 import Network.Curl (
   CurlCode(..),
   CurlOption(..),
@@ -18,8 +20,11 @@ import Network.Curl (
 import Network.Curl.Download (openURI)
 import System.Directory (doesFileExist, getTemporaryDirectory)
 import System.IO (hClose, hPutStrLn, stderr)
+import qualified System.IO.UTF8 as Utf8 (readFile)
 import System.IO.Temp (openBinaryTempFile, openTempFile)
 
+import qualified Configuration as C (Cfg(..))
+import LWN.Article (real_article_path)
 import LWN.URI (URL, filename)
 
 login_url :: URL
@@ -39,7 +44,7 @@ default_curl_opts :: [CurlOption]
 default_curl_opts =
   [ -- The Global cache is not thread-friendly.
     CurlDNSUseGlobalCache False,
-    
+
     -- And we don't want to use a DNS cache anyway.
     CurlDNSCacheTimeout 0,
 
@@ -130,7 +135,7 @@ log_in cookie_jar username password =
         CurlCookieJar cookie_jar,
         CurlPost True,
         CurlPostFields post_data ]
-        
+
     curl_opts :: [CurlOption]
     curl_opts = default_curl_opts ++ post_opts
 
@@ -173,7 +178,7 @@ type ImageMap = Map.Map URL FilePath
 
 download_image_urls :: [URL] -> IO ImageMap
 download_image_urls image_urls = do
-  files <- mapM save_image image_urls
+  files <- parallel $ map save_image image_urls
   let pairs = zip image_urls files
   return $ foldl my_insert empty_map pairs
   where
@@ -182,3 +187,48 @@ download_image_urls image_urls = do
     my_insert :: ImageMap -> (URL, Maybe FilePath) -> ImageMap
     my_insert dict (_, Nothing)  = dict
     my_insert dict (k, Just v) = Map.insert k v dict
+
+
+
+
+
+get_login_cookie :: C.Cfg -> IO C.Cfg
+get_login_cookie cfg
+  | isNothing (C.username cfg) = return cfg
+  | isNothing (C.password cfg) = return cfg
+  | otherwise = do
+      let uname = fromJust $ C.username cfg
+      let pword = fromJust $ C.password cfg
+      cj <- make_cookie_jar
+      li_result <- log_in cj uname pword
+
+      case li_result of
+        Left err -> do
+          let msg = "Failed to log in. " ++ err
+          hPutStrLn stderr msg
+        Right response_body -> do
+          hPutStrLn stderr response_body
+
+      return $ cfg { C.cookie_jar = Just cj }
+
+
+-- | Try to parse the given article using HXT. We try a few different
+--   methods; if none of them work, we return 'Nothing'.
+get_article_contents :: C.Cfg -> URL -> IO (Maybe String)
+get_article_contents cfg article_name = do
+  my_article <- real_article_path article_name
+  is_file <- doesFileExist my_article
+  case is_file of
+    True -> do
+      contents <- Utf8.readFile my_article
+      return $ Just $ contents
+    False -> do
+      -- Download the URL and try to parse it.
+      html <- get_page (C.cookie_jar cfg) my_article
+
+      case html of
+        Left err -> do
+          let msg = "Failed to retrieve page. " ++ err
+          hPutStrLn stderr msg
+          return Nothing
+        Right h -> return $ Just h