From 9b64dc0f38c3540823fbf82c24d81464f742405d Mon Sep 17 00:00:00 2001
From: simonmar <unknown>
Date: Thu, 25 Mar 2004 15:20:55 +0000
Subject: [haddock @ 2004-03-25 15:20:55 by simonmar] Update

---
 TODO | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/TODO b/TODO
index b98944bd..f61daf7c 100644
--- a/TODO
+++ b/TODO
@@ -1,10 +1,6 @@
 -----------------------------------------------------------------------------
 -- bugs
 
-* Ralf Laemmel's isntance bug (email: "haddock 0.6 issue")
-
-* Pick the preferred export names as per the paper.
-
 * A module re-export should only reference the target module if the
   target module is imported without hiding any of its exports (otherwise
   we should inline just the exported bits).
@@ -42,6 +38,13 @@
 -----------------------------------------------------------------------------
 -- features
 
+* In theory there's no reason why we have to only link to modules which
+  are imported by the current module.  This is an artificial
+  restriction.  We might try doing a two-pass strategy: first find out
+  what everything exports, and then resolve all the links in the
+  documentation.  We could mark some modules as "definitive" destinations
+  for links to their entities, or even give a priority for each module.
+
 * nested itemized and enumerated lists.
 
 * There ought to be a way to include some structure in the "description"
@@ -92,6 +95,8 @@
 -----------------------------------------------------------------------------
 -- cosmetic
 
+* Allow more parts of the documentation to be squashed and expanded?
+
 * for a constructor, don't fill its entire cell with a grey background.
 
 * switch to longer rendering form for datatypes when necessary?
-- 
cgit v1.2.3