X-Git-Url: http://gitweb.michael.orlitzky.com/?a=blobdiff_plain;f=src%2FXml.hs;h=c4d4049e2f5ec4fc62d93d80ab31d83e8d07c764;hb=53c5550fee7f8a39a7906545978f15876a06fbd1;hp=21a806847ca1761845260f1a2d4f92344d07916d;hpb=53d408ee8691bc123dd55eba64e2c00cd17c8f57;p=dead%2Fhtsn-import.git

diff --git a/src/Xml.hs b/src/Xml.hs
index 21a8068..c4d4049 100644
--- a/src/Xml.hs
+++ b/src/Xml.hs
@@ -3,24 +3,25 @@
 -- | General XML stuff.
 --
 module Xml (
-  FromXml(..),
+  DtdName(..),
+  ToFromXml(..),
   parse_opts,
-  pickle_unpickle )
+  pickle_unpickle,
+  unpickleable )
 where
 
+import Control.Exception ( SomeException(..), catch )
+import Database.Groundhog ( AutoKey )
 import Text.XML.HXT.Core (
   (>>>),
   (/>),
+  PU,
   SysConfigList,
   XmlPickler(..),
   hasName,
-  no,
   readDocument,
   runX,
-  withPreserveComment,
   withRemoveWS,
-  withSubstDTDEntities,
-  withValidate,
   xpickleVal,
   xunpickleVal,
   yes )
@@ -39,27 +40,36 @@ import Text.XML.HXT.Core (
 --   Groundhog type from the XML one. But Groundhog won't use type family
 --   instances, so here we are.
 --
-class FromXml a where
+class ToFromXml a where
   -- | Each instance a must declare its associated XML type (Xml a)
   type Xml a :: *
+  type Container a :: *
 
   -- | And provide a function for getting an (Xml a) out of an "a."
   to_xml :: a -> Xml a
 
+  -- | And provide a function for getting an "a" out of an (Xml a).
+  from_xml :: Xml a -> a
+
+  -- | Often we need to provide a key to use as a foreign key into
+  --   some container. If the instance "belongs" to some other object,
+  --   then it might need to be passed a key before it can un-XML
+  --   itself. For example, the XML version of 'NewsTeam' doesn't
+  --   contain a message ID which is part of its database type.
+  from_xml_fk :: AutoKey (Container a) -> Xml a -> a
+  from_xml_fk _ = from_xml
+
+
+-- | Represents the DTD filename ("SYSTEM") part of the DOCTYPE
+--   definition.
+newtype DtdName = DtdName String
 
 -- | A list of options passed to 'readDocument' when we parse an XML
---   document. We don't validate because the DTDs from TSN are
---   wrong. As a result, we don't want to keep useless DTDs
---   areound. Thus we disable 'withSubstDTDEntities' which, when
---   combined with "withValidate no", prevents HXT from trying to read
---   the DTD at all.
+--   document. All cosmetic whitespace should be removed, otherwise we
+--   have to parse it in each pickler.
 --
 parse_opts :: SysConfigList
-parse_opts =
-  [ withPreserveComment no,
-    withRemoveWS yes,
-    withSubstDTDEntities no,
-    withValidate no ]
+parse_opts = [ withRemoveWS yes ]
 
 
 -- | Given a root element name and a file path, return both the
@@ -73,6 +83,11 @@ parse_opts =
 --   to 'xpickle' would be ambiguous. By returning some @a@s, we allow
 --   the caller to annotate its type.
 --
+--   Note that this will happily pickle nothing to nothing and then
+--   unpickle it back to more nothing. So the fact that the
+--   before/after results from this function agree does not mean that
+--   the document was successfully unpickled!
+--
 pickle_unpickle :: XmlPickler a
                 => String
                 -> FilePath
@@ -80,7 +95,7 @@ pickle_unpickle :: XmlPickler a
 pickle_unpickle root_element filepath = do
   -- We need to check only the root message element since
   -- readDocument produces a bunch of other junk.
-    expected <- runX $ arr_getobj
+    expected <- runX arr_getobj
     actual <- runX $ arr_getobj
                      >>>
                      xpickleVal xpickle
@@ -94,3 +109,20 @@ pickle_unpickle root_element filepath = do
                    hasName root_element
                    >>>
                    xunpickleVal xpickle
+
+
+
+-- | Is the given XML file unpickleable? Unpickling will be attempted
+--   using the @unpickler@ argument. If we unilaterally used the
+--   generic 'xpickle' function for our unpickler, a type ambiguity
+--   would result. By taking the unpickler as an argument, we allow
+--   the caller to indirectly specify a concrete type.
+--
+unpickleable :: XmlPickler a => FilePath -> PU a -> IO Bool
+unpickleable filepath unpickler = do
+  xmldoc <- try_unpickle `catch` (\(SomeException _) -> return [])
+  return $ (not . null) xmldoc
+  where
+    try_unpickle = runX $ readDocument parse_opts filepath
+                          >>>
+                          xunpickleVal unpickler