Only fetch login cookies once.

[dead/lwn-epub.git] / src / LWN / HTTP.hs

diff --git a/src/LWN/HTTP.hs b/src/LWN/HTTP.hs
index 1b6bcfd3c966a060ea326a06245f67811f66ba9a..63079fd127aa853e6b69eef96588153dace02d30 100644 (file)
--- a/src/LWN/HTTP.hs
+++ b/src/LWN/HTTP.hs
@@ -4,12 +4,12 @@ module LWN.HTTP
 where
 
 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(..),
   CurlResponse,
-  URLString,
   do_curl_,
   initialize,
   respBody,
@@ -19,11 +19,14 @@ 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 LWN.URI (filename)
+import qualified Configuration as C (Cfg(..))
+import LWN.Article (real_article_path)
+import LWN.URI (URL, filename)
 
-login_url :: URLString
+login_url :: URL
 login_url = "https://lwn.net/login"
 
 username_field :: String
@@ -40,7 +43,7 @@ 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,
 
@@ -48,10 +51,7 @@ default_curl_opts =
     CurlFollowLocation True,
 
     -- Give it a little time...
-    CurlTimeout 45,
-
-    -- For debugging.
-    CurlVerbose True ]
+    CurlTimeout 45 ]
 
 
 make_cookie_jar :: IO FilePath
@@ -62,7 +62,7 @@ make_cookie_jar = do
   hClose out_handle -- We just want to create it for now.
   return out_path
 
-get_page :: Maybe FilePath -> URLString -> IO (Either String String)
+get_page :: Maybe FilePath -> URL -> IO (Either String String)
 get_page cookie_file url =
   withCurlDo $ do
     -- Create a curl instance.
@@ -93,9 +93,8 @@ get_page cookie_file url =
 
 
 -- | Log in using curl. Store the resulting session cookies in the
---   supplied file.Warning: This returns an error if the function
---   fails!
-log_in :: FilePath -> String -> String -> IO (Maybe String)
+--   supplied file.
+log_in :: FilePath -> String -> String -> IO (Either String String)
 log_in cookie_jar username password =
   withCurlDo $ do
     -- Create a curl instance.
@@ -111,8 +110,8 @@ log_in cookie_jar username password =
 
     return $
       case code of
-        CurlOK -> Nothing
-        error_code -> Just $ "HTTP Error: " ++ (show error_code)
+        CurlOK -> Right (respBody resp)
+        error_code -> Left $ "HTTP Error: " ++ (show error_code)
         -- If an error occurred, we want to dump as much information as
         -- possible. If this becomes a problem, we can use respGetInfo to
         -- query the response object for more information
@@ -135,7 +134,7 @@ log_in cookie_jar username password =
         CurlCookieJar cookie_jar,
         CurlPost True,
         CurlPostFields post_data ]
-        
+
     curl_opts :: [CurlOption]
     curl_opts = default_curl_opts ++ post_opts
 
@@ -147,7 +146,7 @@ log_in cookie_jar username password =
 --   We need to be able to parse the filename out of the URL
 --   so that when we stick our image in the document, the reader
 --   knows that type (jpg, png, etc.) it is.
-save_image :: URLString -> IO (Maybe FilePath)
+save_image :: URL -> IO (Maybe FilePath)
 save_image url = do
   it_exists <- doesFileExist url
   if it_exists then do
@@ -168,3 +167,67 @@ save_image url = do
           Right bs -> do
             B.hPut out_handle bs
             return $ Just out_path
+
+
+
+
+-- | Map absolute image URLs to local system file paths where the
+--   image referenced by the URL is stored.
+type ImageMap = Map.Map URL FilePath
+
+download_image_urls :: [URL] -> IO ImageMap
+download_image_urls image_urls = do
+  files <- mapM save_image image_urls
+  let pairs = zip image_urls files
+  return $ foldl my_insert empty_map pairs
+  where
+    empty_map = Map.empty :: ImageMap
+
+    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