-- Hoogle documentation, generated by Haddock
-- See Hoogle, http://www.haskell.org/hoogle/


-- | Scrape websites for changes
--   
--   This library scrapes websites and invokes callbacks when there are
--   changes, similar to a RSS reader. Sending an email by invoking
--   sendmail is a built-in callback mechanism. Of course, users can
--   provide their own callback function in addition.
--   
--   Look at <a>Network.ScrapeChanges</a> for a full working example on how
--   to use "scrape-changes".
@package scrape-changes
@version 0.1.0.5

module Network.ScrapeChanges.Domain

-- | String encoded in the standard cron format
type CronScheduleString = String

-- | Url to scrape
type Url = String

-- | Body of the HTTP request
type HttpBody = ByteString

-- | Function extracting <a>Text</a> of <a>HttpBody</a>
type Scraper = HttpBody -> Text

-- | Codomain of <a>Scraper</a>
type Text = Text

-- | Mail address for provided <a>MailConfig</a>
data MailAddr
MailAddr :: Maybe Text -> String -> MailAddr

-- | Optional name for the given <a>_mailAddr</a>
[_mailAddrName] :: MailAddr -> Maybe Text

-- | Mail address
[_mailAddr] :: MailAddr -> String
mailAddrName :: Lens' MailAddr (Maybe Text)
mailAddr :: Lens' MailAddr String
data Mail
Mail :: MailAddr -> NonEmpty MailAddr -> Text -> Text -> Mail
[_mailFrom] :: Mail -> MailAddr
[_mailTo] :: Mail -> NonEmpty MailAddr
[_mailSubject] :: Mail -> Text
[_mailBody] :: Mail -> Text
mailTo :: Lens' Mail (NonEmpty MailAddr)
mailSubject :: Lens' Mail Text
mailFrom :: Lens' Mail MailAddr
mailBody :: Lens' Mail Text
data CallbackConfig

-- | Send a mail when there's changed data at your scrape target. This
--   needs sendmail to be configured correctly on the host your program
--   runs.
MailConfig :: Mail -> CallbackConfig

-- | Just execute the provided function when there's changed data at your
--   scrape target.
OtherConfig :: (Text -> IO ()) -> CallbackConfig
_OtherConfig :: Prism' CallbackConfig (Text -> IO ())
_MailConfig :: Prism' CallbackConfig Mail
data ScrapeConfig
ScrapeConfig :: String -> CallbackConfig -> ScrapeConfig

-- | The url to be called using GET
[_scrapeInfoUrl] :: ScrapeConfig -> String

-- | The callback config to be executed when something in
--   <a>_scrapeInfoUrl</a> has changed
[_scrapeInfoCallbackConfig] :: ScrapeConfig -> CallbackConfig
scrapeInfoUrl :: Lens' ScrapeConfig String
scrapeInfoCallbackConfig :: Lens' ScrapeConfig CallbackConfig
data ScrapeResult

-- | Signals that the last execution of the provided <a>ScrapeConfig</a>
--   led to execution of <a>CallbackConfig</a>
CallbackCalled :: ScrapeResult

-- | Signals that the last execution of the provided <a>ScrapeConfig</a>
--   didn't lead to execution of <a>CallbackConfig</a>
CallbackNotCalled :: ScrapeResult
data ScrapeSchedule
ScrapeSchedule :: CronScheduleString -> ScrapeConfig -> Scraper -> ScrapeSchedule
[_scrapeScheduleCron] :: ScrapeSchedule -> CronScheduleString
[_scrapeScheduleConfig] :: ScrapeSchedule -> ScrapeConfig
[_scrapeScheduleScraper] :: ScrapeSchedule -> Scraper
scrapeScheduleScraper :: Lens' ScrapeSchedule Scraper
scrapeScheduleCron :: Lens' ScrapeSchedule CronScheduleString
scrapeScheduleConfig :: Lens' ScrapeSchedule ScrapeConfig
data ValidationError
UrlNotAbsolute :: ValidationError
UrlProtocolInvalid :: ValidationError
MailConfigInvalidMailFromAddr :: String -> ValidationError
MailConfigInvalidMailToAddr :: String -> ValidationError
CronScheduleInvalid :: String -> ValidationError
type ScrapeValidation t = AccValidation [ValidationError] t
instance GHC.Classes.Eq Network.ScrapeChanges.Domain.ValidationError
instance GHC.Show.Show Network.ScrapeChanges.Domain.ValidationError
instance GHC.Show.Show Network.ScrapeChanges.Domain.ScrapeResult
instance Data.Hashable.Class.Hashable Network.ScrapeChanges.Domain.ScrapeConfig
instance GHC.Classes.Eq Network.ScrapeChanges.Domain.ScrapeConfig
instance GHC.Classes.Eq Network.ScrapeChanges.Domain.CallbackConfig
instance Data.Hashable.Class.Hashable Network.ScrapeChanges.Domain.Mail
instance GHC.Generics.Generic Network.ScrapeChanges.Domain.Mail
instance GHC.Classes.Eq Network.ScrapeChanges.Domain.Mail
instance GHC.Show.Show Network.ScrapeChanges.Domain.Mail
instance Data.Hashable.Class.Hashable Network.ScrapeChanges.Domain.MailAddr
instance GHC.Generics.Generic Network.ScrapeChanges.Domain.MailAddr
instance GHC.Classes.Eq Network.ScrapeChanges.Domain.MailAddr
instance GHC.Show.Show Network.ScrapeChanges.Domain.MailAddr

module Network.ScrapeChanges.Internal

-- | Helper constructor for <a>ScrapeConfig</a> containing
--   <a>MailConfig</a> callback.
mailScrapeConfig :: ScrapeInfoUrl -> MailFromAddr -> NonEmpty MailToAddr -> ScrapeConfig

-- | Helper constructor for <a>ScrapeConfig</a> containing
--   <a>OtherConfig</a> callback.
otherScrapeConfig :: ScrapeInfoUrl -> (Text -> IO ()) -> ScrapeConfig
validateScrapeConfig :: ScrapeConfig -> ScrapeValidation ScrapeConfig
validateCronSchedule :: CronScheduleString -> ScrapeValidation CronScheduleString
readLatestHash :: (Hashable t) => t -> IO (Maybe Hash)
saveHash :: (Hashable t) => t -> Hash -> IO ()
executeCallbackConfig :: ScrapeConfig -> Text -> IO ()
removeHash :: (Hashable t) => t -> IO ()
removeHashes :: IO ()
hash' :: Hashable t => t -> String
type ScrapeInfoUrl = String
type MailFromAddr = MailAddr
type MailToAddr = MailAddr
type Hash = String
loggerName :: String
httpExceptionHandler :: ScrapeConfig -> HttpException -> IO t


-- | Main module, reexports everything you need to use "scrape-changes".
--   Full working example:
--   
--   <pre>
--   {-# LANGUAGE OverloadedStrings #-}
--   
--   import Data.ByteString (isInfixOf)
--   import Data.ByteString.Lazy (ByteString, toStrict)
--   import Data.Text.Lazy.Encoding (decodeUtf8With)
--   import Data.Foldable (find)
--   import Data.Maybe (fromMaybe)
--   import Text.HTML.TagSoup (Tag(..), (~==), (~/=), parseTags, fromAttrib) 
--   import Data.List.NonEmpty (NonEmpty ((:|)))
--   import qualified System.Log.Logger as Logger
--   import qualified System.Log.Handler.Syslog as Syslog
--   import Data.Monoid ((&lt;&gt;))
--   import Control.Monad (forever)
--   import Network.ScrapeChanges
--   
--   main :: IO ()
--   main = do
--     _ &lt;- configureLogging
--     _ &lt;- (Logger.errorM thisLogger . show) <a>either</a> id $ scrapeChangesJobs 
--     putStrLn "scrape-changes examples executable. Just look at the example source code."
--     -- |Simplest way to block the main thread forever. Good enough for the use cases of 'scrape-changes'
--     _ &lt;- forever getLine
--     -- |Will never be executed in this case
--     clearAllScrapeConfigs
--   
--   -- |Google logo scrape function using the tagsoup library
--   scrapeGoogleLogo :: ByteString -&gt; Text
--   scrapeGoogleLogo byteString =   
--     let tags                 = parseTags byteString
--         divWithBackgroundUrl = find (~/= TagClose ("div" :: ByteString)) $
--                                dropWhile (not . isDivWithBackgroundUrl) tags 
--         resultMaybe          = decodeUtf8Lenient . styleAttribContent <a>$</a> divWithBackgroundUrl
--     in fromMaybe "" resultMaybe 
--     where decodeUtf8Lenient = decodeUtf8With $ const . const . Just $ <tt>?</tt>
--           isDivWithBackgroundUrl t = 
--             let containsBackgroundUrl = isInfixOf "background:url" . toStrict
--             in t ~== TagOpen ("div" :: ByteString) [] &amp;&amp; containsBackgroundUrl (styleAttribContent t)
--           styleAttribContent = fromAttrib "style"
--   
--   scrapeChangesJobs :: Either [(Url, [ValidationError])] (IO ())
--   scrapeChangesJobs = repeatScrapeAll [
--       -- Checks each minute for changes and sends a mail if there are any
--       ScrapeSchedule {
--         _scrapeScheduleCron = "* * * * *" -- std cron format
--       , _scrapeScheduleConfig = mailScrapeConfig "<a>http://www.google.co.uk"</a> -- to scrape
--                                                  (MailAddr Nothing "max<tt>mustermann.de") -- from
--                                                  (MailAddr Nothing "receiver</tt>scrape-changes.com" :| []) -- to
--       , _scrapeScheduleScraper = scrapeGoogleLogo --scrape function
--       }
--       -- Checks each minute for changes and notifies to syslog if there are any
--     , ScrapeSchedule {
--         _scrapeScheduleCron = "* * * * *"
--       , _scrapeScheduleConfig = otherScrapeConfig "<a>http://www.google.co.uk"</a> 
--                                                   (text -&gt; Logger.infoM thisLogger . show $ 
--                                                             "Change detected: " &lt;&gt; text)
--       , _scrapeScheduleScraper = scrapeGoogleLogo
--       }
--     ]
--   
--   configureLogging :: IO ()
--   configureLogging = do
--     syslogHandler &lt;- Syslog.openlog thisLogger [] Syslog.DAEMON Logger.DEBUG
--     let logConfig = flip Logger.updateGlobalLogger (Logger.addHandler syslogHandler . Logger.setLevel Logger.DEBUG)
--     sequence_ $ logConfig <a>$</a> [<a>Network.ScrapeChanges</a>, thisLogger]
--   
--   thisLogger :: String
--   thisLogger = "scrape-changes-examples"
--   </pre>
module Network.ScrapeChanges

-- | The basic scrape function. It fires a GET request against the url
--   defined within the provided <a>ScrapeConfig</a>. The body is passed to
--   the provided <a>Scraper</a>. The result <a>Text</a> of the latter is
--   used to determine whether something has changed on the respective
--   website. If so, the callback configured in <a>ScrapeConfig</a> is
--   executed and <a>CallbackCalled</a> is returned. Otherwise
--   <a>CallbackNotCalled</a> is returned.
scrape :: ScrapeConfig -> Scraper -> Either [ValidationError] (IO ScrapeResult)

-- | Repeat executing <a>scrape</a> by providing a
--   <a>CronScheduleString</a>. The returned IO action is non blocking
repeatScrape :: CronScheduleString -> ScrapeConfig -> Scraper -> Either [ValidationError] (IO ())

-- | Execute a list of <a>ScrapeSchedule</a> in parallel. If validation of
--   any <a>ScrapeSchedule</a> fails, <a>Left</a> containing
--   <a>ValidationError</a> indexed by the corresponding <a>Url</a> is
--   returned.
repeatScrapeAll :: [ScrapeSchedule] -> Either [(Url, [ValidationError])] (IO ())

-- | Execute a list of <a>ScrapeConfig</a> in sequence using <a>scrape</a>
--   and collect the results in a map containing the respective <a>Url</a>
--   as key.
scrapeAll :: [(ScrapeConfig, Scraper)] -> [(Url, Either [ValidationError] (IO ScrapeResult))]

-- | Helper constructor for <a>ScrapeConfig</a> containing
--   <a>MailConfig</a> callback.
mailScrapeConfig :: ScrapeInfoUrl -> MailFromAddr -> NonEmpty MailToAddr -> ScrapeConfig

-- | Helper constructor for <a>ScrapeConfig</a> containing
--   <a>OtherConfig</a> callback.
otherScrapeConfig :: ScrapeInfoUrl -> (Text -> IO ()) -> ScrapeConfig

-- | Clear all mutable state associated with the provided
--   <a>ScrapeConfig</a>
clearScrapeConfig :: ScrapeConfig -> IO ()

-- | Clear all mutable state ever used by "scrape-changes"
clearAllScrapeConfigs :: IO ()