1
1
TODO list for PostgreSQL
2
2
========================
3
- Last updated: Thu Jan 4 03:15:12 EST 2001
3
+ Last updated: Thu Jan 4 03:24:53 EST 2001
4
4
5
5
Current maintainer: Bruce Momjian (pgman@candle.pha.pa.us)
6
6
@@ -27,26 +27,26 @@ PARSER
27
27
[inherit]
28
28
* -Be smarter about promoting types when UNION merges different data types
29
29
* -redesign INSERT ... SELECT to have two levels of target list
30
- * have INTERSECT/EXCEPT prevent duplicates unless ALL is specified
31
- * SELECT col::DECIMAL(12,10); fails
30
+ * - have INTERSECT/EXCEPT prevent duplicates unless ALL is specified (Tom)
31
+ * - SELECT col::DECIMAL(12,10); fails
32
32
33
33
VIEWS
34
34
35
35
* -Views containing aggregates sometimes fail (Jan)
36
- * Creating view and inheriting the view causes view* to show
36
+ * - Creating view and inheriting the view causes view* to show
37
37
duplicates (inherit)
38
38
* -Disallow LOCK on view (Mark Hollomon)
39
39
40
40
MISC
41
41
42
42
* Plpgsql does not handle quoted mixed-case identifiers
43
- * Buffer reference counting bugfixes
44
- * Fix libpq bug that causes it to drop backend error message sent
43
+ * - Buffer reference counting bugfixes (Tom)
44
+ * - Fix libpq bug that causes it to drop backend error message sent
45
45
just before connection closure (ie, any FATAL error message)
46
46
* -SELECT ... UNION ... ORDER BY fails when sort expr not in result list
47
47
* -SELECT ... UNION ... GROUP BY fails if column types disagree, no type
48
48
promotion occurs
49
- * Modification of pg_class can happen while table in use by
49
+ * - Modification of pg_class can happen while table in use by
50
50
another backend. Might lead to MVCC inside of syscache
51
51
* Permission to DELETE table allows UPDATE also
52
52
57
57
58
58
* -Add OUTER joins, left and right[outer] (Tom, Thomas)
59
59
* -Allow long tuples by chaining or auto-storing outside db (TOAST) (Jan)
60
- * Fix memory leak for expressions[memory] (Tom? )
60
+ * - Fix memory leak for expressions[memory] (Tom)
61
61
* Add replication of distributed databases [replication]
62
62
o automatic fallover
63
63
o load balancing
@@ -117,7 +117,7 @@ TYPES
117
117
* Make functions more multi-byte aware, i.e. trim()
118
118
* Make n of CHAR(n)/VARCHAR(n) the number of letters, not bytes
119
119
* -Update macaddr manufacturer numbers, or remove the function macaddr_manuf()
120
- * Add btree index support for reltime, tinterval, regproc, bit, varbit
120
+ * Add btree index support for reltime, tinterval, regproc
121
121
* Add rtree index support for line, lseg, path, point
122
122
* Make binary/file in/out interface for TOAST columns
123
123
* SELECT cash_out(2) crashes because of opaque
@@ -127,10 +127,9 @@ VIEWS
127
127
* -Allow DISTINCT on views
128
128
* -Allow views of aggregate columns
129
129
* -Allow views with subselects
130
- * Create insert, update and delete rules for simple one table views
130
+ * Automatically create rules on views so they are updateable, per SQL92
131
131
* -Change elog for complex view ins|upd|del to "cannot {ins|upd|del}
132
132
* Add the functionality for "with check option" clause of create view
133
- * Automatically create rules on views so they are updateable, per SQL92
134
133
135
134
INDEXES
136
135
@@ -168,7 +167,7 @@ COMMANDS
168
167
* Allow BINARY option to SELECT, like we do with DECLARE
169
168
* MOVE 0 should not move to end of cursor
170
169
* Overhaul ACL (access control) code
171
- * Allow ORDER BY...LIMIT in INSERT INTO ... SELECT
170
+ * - Allow ORDER BY...LIMIT in INSERT INTO ... SELECT (Tom)
172
171
* Add SHOW command to display locks
173
172
174
173
CLIENTS
@@ -193,8 +192,8 @@ REFERENTIAL INTEGRITY
193
192
* Check that primary key exists at foreign key definition time
194
193
* Prevent column dropping if column is used by foreign key
195
194
* Propagate column or table renaming to foreign key constraints
196
- * Emit a warning at foreign key creation time if no UNIQUE index
197
- exists on referenced primary key attributes
195
+ * - Emit a warning at foreign key creation time if no UNIQUE index
196
+ exists on referenced primary key attributes (Tom)
198
197
* Add deferred trigger queue file (Jan)
199
198
* Allow oid to act as a foreign key
200
199
* Implement dirty reads and use them in RI triggers
@@ -234,10 +233,10 @@ MISC
234
233
* Overhaul bufmgr/lockmgr/transaction manager
235
234
* -redesign UNION structures to have separarate target lists
236
235
* -Allow multi-level query trees for INSERT INTO ... SELECT
237
- * Use IPC_EXCL when creating shared memory and semaphores
236
+ * - Use IPC_EXCL when creating shared memory and semaphores (Tom)
238
237
* have pg_upgrade use pg_ctl to stop/start postmaster
239
238
* Encrpyt passwords in pg_shadow table using MD5 (Vince)
240
- * Use flock() to prevent multiple postmasters on the same port [flock]
239
+ * - Use flock() to prevent multiple postmasters on the same port (Tom)
241
240
* Allow Java server-side programming
242
241
243
242
PERFORMANCE
@@ -255,9 +254,10 @@ INDEXES
255
254
non-consecutive keys or OR clauses, so fewer heap accesses
256
255
* Allow SELECT * FROM tab WHERE int2col = 4 use int2col index, int8,
257
256
float4, numeric/decimal too [optimizer]
258
- * Include heap CTID in btree index keys, remove equal-key cruft from btree
257
+ * -Include heap CTID in btree index keys, remove equal-key cruft from
258
+ btree (Tom)
259
259
* Use indexes with CIDR '<<' (contains) operator
260
- * Fix LIKE indexing optimization for non-ASCII locales
260
+ * Allow LIKE indexing optimization for non-ASCII locales
261
261
262
262
CACHE
263
263
@@ -267,21 +267,19 @@ CACHE
267
267
MISC
268
268
269
269
* Allow compression of log and meta data
270
- * Allow char() not to use variable-sized header to reduce disk size
271
270
* Do async I/O to do better read-ahead of data
272
271
* Get faster regex() code from Henry Spencer <henry@zoo.utoronto.ca>
273
272
when it is available
274
273
* Use mmap() rather than SYSV shared memory(?)
275
- * Make oid use oidin/oidout not int4in/int4out in pg_type.h
274
+ * - Make oid use oidin/oidout not int4in/int4out in pg_type.h (Tom)
276
275
* Improve Subplan list handling
277
276
* Allow Subplans to use efficient joins(hash, merge) with upper variable
278
277
[subquery]
279
278
* -use fmgr_info()/fmgr_faddr() instead of fmgr() calls in high-traffic
280
279
places, like GROUP BY, UNIQUE, index processing, etc.
281
280
* improve dynamic memory allocation by introducing tuple-context memory
282
- allocation [memory]
281
+ allocation (Tom)
283
282
* -In WHERE tab1.x=3 AND tab1.x=tab2.y, add tab2.y=3
284
- * prevent labels from being output for stored rules (Tom)
285
283
* allow configuration of maximum number of open files
286
284
* Remove pg_listener index
287
285
* -Redesign ANALYZE in VACUUM so it can be run separately without locks
0 commit comments