-1. Re-enable DTD validation once
- https://github.com/UweSchmidt/hxt/issues/15 is fixed.
-
-2. Write a test for test/xml/Odds_XML-long-import.xml once it no
- longer takes 10 minutes to import (Postgres only?).
-
-3. We have DTDs but no sample XML for the following SportInfo types,
+1. We have DTDs but no sample XML for the following SportInfo types,
which have therefore been left unimplmented for now:
* CBASK_StealsXML.dtd
- * Cbask_Team_Scoring_XML.dtd
- * Cbask_Team_Shooting_Pct_XML.dtd
-4. The following DTD types were handled (in some form) by the old
+2. The following DTD types were handled (in some form) by the old
FeedGrabber. They are not yet handled by htsn-import (some may not
be valid):
* WNBA_Individual_Stats_XML
* WNBATeamScheduleXML
-5. Consolidate all of the make_game_time functions which take a
+3. Consolidate all of the make_game_time functions which take a
date/time and produce a combined time.
-6. Factor out test code where possible; a lot of them differ only in
+4. Factor out test code where possible; a lot of them differ only in
the filename.
-7. Combine test XML files where possible.
-
-8. Remove the -XDeriveGeneric argument in test/Doctests.hs; it
- shouldn't be necessary if doctest picks up the LANGUAGE pragmas on its
- own.
-
-9. Once generics-sop releases support for bigger tuples, we can remove
- the deriveGeneric stuff from Generics.hs. This should also fix the
- doctests.
+5. Combine test XML files where possible. We don't need to e.g. import
+ two copies of newsxml just because one has an empty attribute that
+ we want to test: we could just delete an attribute from the first
+ file.