New result set model and modular ranking system. Moved towards

[idzebra-moved-to-github.git] / TODO

diff --git a/TODO b/TODO
index fc1656c..120027b 100644 (file)
--- a/TODO
+++ b/TODO
@@ -1,23 +1,26 @@
-Zebra TODO $Id: TODO,v 1.2 1997-09-29 09:09:04 adam Exp $
+Zebra TODO $Id: TODO,v 1.6 1998-03-05 08:45:11 adam Exp $
 
-Better ranking in searches. Admin should be able specify initial
- weight to certain fields.
+Make regx-filter thread safe.
+
+Size of sort entries should be configurable.
+
+Use first field in sorting, i.e. author.
+
+System number sorting.
 
-Search result should hold information about hits for each term,
- especially when using ranked search.
+Configurable default sorting criteria - used when query is not ranked.
 
-Admin should be able to specify set/attr when none is specified
- (currently bib1,use=any is used).
+Configurable default search attributes.
+
+Better ranking in searches. Admin should be able specify initial
+ weight to certain fields.
 
 Explain support - including d1 to grs (d1_grs.c) in YAZ; Zebra
  to auto-generate explain information depending on data1 system.
 
-Generic character set mapping (w, p, etc.)
-
 ISAMC optimization: indirect block with pointers to all blocks
  in chain. The initial block should include the count as well.
 
-Test mode fix: when using test mode (-s) Zebra should only do
- read-locking and not write-locking thus preventing Zebra thinking
- the index is inconsistent.
- 
+Create surrogate diagnostic when ONE record is missing insead of a
+non-surrogate diagnostic.
+