1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
2 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
5 <title>PostgreSQL TODO List</title>
6 <meta name="generator" content="HTML::TextToHTML v2.25"/>
8 <body bgcolor="#FFFFFF" text="#000000" link="#FF0000" vlink="#A00000" alink="#0000FF">
9 <h1><a name="section_1">PostgreSQL TODO List</a></h1>
10 <p>Current maintainer: Bruce Momjian (<a href="mailto:bruce@momjian.us">bruce@momjian.us</a>)<br/>
11 Last updated: Wed Dec 12 01:59:39 EST 2007
13 <p>The most recent version of this document can be viewed at<br/>
14 <a href="http://www.postgresql.org/docs/faqs.TODO.html">http://www.postgresql.org/docs/faqs.TODO.html</a>.
16 <p><strong>A hyphen, "-", marks changes that will appear in the upcoming 8.3 release.</strong><br/>
17 <strong>A percent sign, "%", marks items that are easier to implement.</strong>
19 <p>Bracketed items, "[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?"></a>]", have more detail.
21 <p>This list contains all known PostgreSQL bugs and feature requests. If<br/>
22 you would like to work on an item, please read the Developer's FAQ<br/>
23 first. There is also a developer's wiki at<br/>
24 <a href="http://developer.postgresql.org">http://developer.postgresql.org</a>.
26 <h1><a name="section_2">Administration</a></h1>
29 <li>Allow major upgrades without dump/reload, perhaps using pg_upgrade
30 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pg_upgrade">pg_upgrade</a>]
31 </li><li>Check for unreferenced table files created by transactions that were
32 in-progress when the server terminated abruptly
33 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00096.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00096.php</a>
35 </li><li>Allow administrators to safely terminate individual sessions either
36 via an SQL function or SIGTERM
37 <p> Lock table corruption following SIGTERM of an individual backend
38 has been reported in 8.0. A possible cause was fixed in 8.1, but
39 it is unknown whether other problems exist. This item mostly
40 requires additional testing rather than of writing any new code.
42 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php</a>
44 </li><li>Set proper permissions on non-system schemas during db creation
45 <p> Currently all schemas are owned by the super-user because they are copied
46 from the template1 database. However, since all objects are inherited
47 from the template database, it is not clear that setting schemas to the db
50 </li><li>Support table partitioning that allows a single table to be stored
51 in subtables that are partitioned based on the primary key or a WHERE
53 </li><li>Add function to report the time of the most recent server reload
54 </li><li>Allow statistics collector information to be pulled from the collector
55 process directly, rather than requiring the collector to write a
56 filesystem file twice a second?
57 </li><li>Allow log_min_messages to be specified on a per-module basis
58 <p> This would allow administrators to see more detailed information from
59 specific sections of the backend, e.g. checkpoints, autovacuum, etc.
60 Another idea is to allow separate configuration files for each module,
61 or allow arbitrary SET commands to be passed to them.
63 </li><li>Simplify ability to create partitioned tables
64 <p> This would allow creation of partitioned tables without requiring
65 creation of rules for INSERT/UPDATE/DELETE, and constraints for
66 rapid partition selection. Options could include range and hash
69 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00375.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00375.php</a>
70 <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00151.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00151.php</a>
72 </li><li>Allow auto-selection of partitioned tables for min/max() operations
73 </li><li>Allow more complex user/database default GUC settings
74 <p> Currently ALTER USER and ALTER DATABASE support per-user and
75 per-database defaults. Consider adding per-user-and-database
76 defaults so things like search_path can be defaulted for a
77 specific user connecting to a specific database.
79 </li><li>Allow custom variable classes that can restrict who can set the values
80 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00911.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00911.php</a>
82 </li><li>Implement the SQL standard mechanism whereby REVOKE ROLE revokes only
83 the privilege granted by the invoking role, and not those granted
85 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-05/msg00010.php">http://archives.postgresql.org/pgsql-bugs/2007-05/msg00010.php</a>
87 </li><li>Improve replication solutions
90 <p> You can use any of the master/slave replication servers to use a
91 standby server for data warehousing. To allow read/write queries to
92 multiple servers, you need multi-master replication like pgcluster.
94 </li><li>Allow replication over unreliable or non-persistent links
96 </li><li>Configuration files
98 <li>Allow pg_hba.conf to specify host names along with IP addresses
99 <p> Host name lookup could occur when the postmaster reads the
100 pg_hba.conf file, or when the backend starts. Another
101 solution would be to reverse lookup the connection IP and
102 check that hostname against the host names in pg_hba.conf.
103 We could also then check that the host name maps to the IP
106 </li><li>%Allow postgresql.conf file values to be changed via an SQL
107 API, perhaps using SET GLOBAL
108 </li><li>Allow the server to be stopped/restarted via an SQL API
109 </li><li>Issue a warning if a change-on-restart-only postgresql.conf value
110 is modified and the server config files are reloaded
111 </li><li>Mark change-on-restart-only values in postgresql.conf
115 <li>Allow a database in tablespace t1 with tables created in
116 tablespace t2 to be used as a template for a new database created
117 with default tablespace t2
118 <p> All objects in the default database tablespace must have default
119 tablespace specifications. This is because new databases are
120 created by copying directories. If you mix default tablespace
121 tables and tablespace-specified tables in the same directory,
122 creating a new database from such a mixed directory would create a
123 new database with tables that had incorrect explicit tablespaces.
124 To fix this would require modifying pg_class in the newly copied
125 database, which we don't currently do.
127 </li><li>Allow reporting of which objects are in which tablespaces
128 <p> This item is difficult because a tablespace can contain objects
129 from multiple databases. There is a server-side function that
130 returns the databases which use a specific tablespace, so this
131 requires a tool that will call that function and connect to each
132 database to find the objects in each database for that tablespace.
134 </li><li>Allow WAL replay of CREATE TABLESPACE to work when the directory
135 structure on the recovery computer is different from the original
136 </li><li>Allow per-tablespace quotas
138 </li><li>Point-In-Time Recovery (PITR)
140 <li>Allow a warm standby system to also allow read-only statements
141 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pitr">pitr</a>]
142 <p> This is useful for checking PITR recovery.
144 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00050.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00050.php</a>
146 </li><li>%Create dump tool for write-ahead logs for use in determining
147 transaction id for point-in-time recovery
148 </li><li>Allow the PITR process to be debugged and data examined
149 </li><li>Allow recovery.conf to allow the same syntax as
150 postgresql.conf, including quoting
151 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00497.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00497.php</a>
155 <h1><a name="section_3">Data Types</a></h1>
158 <li>Change NUMERIC to enforce the maximum precision
159 </li><li>Reduce storage space for small NUMERICs
160 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg01331.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg01331.php</a>
161 <a href="http://archives.postgresql.org/pgsql-patches/2007-02/msg00505.php">http://archives.postgresql.org/pgsql-patches/2007-02/msg00505.php</a>
163 </li><li>Fix data types where equality comparison isn't intuitive, e.g. box
164 </li><li>Allow user-defined types to specify a type modifier at table creation
166 </li><li>Add support for public SYNONYMs
167 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php</a>
169 </li><li>Fix CREATE CAST on DOMAINs
170 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php">http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php</a>
171 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php</a>
173 </li><li>Add support for SQL-standard GENERATED/IDENTITY columns
174 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php</a>
175 <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00038.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00038.php</a>
176 <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00344.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00344.php</a>
177 <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00076.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00076.php</a>
179 </li><li>Improve XML support
180 <p> <a href="http://developer.postgresql.org/index.php/XML_Support">http://developer.postgresql.org/index.php/XML_Support</a>
182 </li><li>Consider placing all sequences in a single table, or create a system
184 </li><li>Allow all data types to cast to and from TEXT
185 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00017.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00017.php</a>
187 </li><li>Dates and Times
189 <li>Allow infinite dates and intervals just like infinite timestamps
190 </li><li>Merge hardwired timezone names with the TZ database; allow either
191 kind everywhere a TZ name is currently taken
192 </li><li>Allow TIMESTAMP WITH TIME ZONE to store the original timezone
193 information, either zone name or offset from UTC [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?timezone">timezone</a>]
194 <p> If the TIMESTAMP value is stored with a time zone name, interval
195 computations should adjust based on the time zone rules.
197 </li><li>Fix SELECT '0.01 years'::interval, '0.01 months'::interval
198 </li><li>Add a GUC variable to allow output of interval values in ISO8601
200 </li><li>Have timestamp subtraction not call justify_hours()?
201 <p> <a href="http://archives.postgresql.org/pgsql-sql/2006-10/msg00059.php">http://archives.postgresql.org/pgsql-sql/2006-10/msg00059.php</a>
203 </li><li>Improve timestamptz subtraction to be DST-aware
204 <p> Currently subtracting one date from another that crosses a
205 daylight savings time adjustment can return '1 day 1 hour', but
206 adding that back to the first date returns a time one hour in
207 the future. This is caused by the adjustment of '25 hours' to
208 '1 day 1 hour', and '1 day' is the same time the next day, even
209 if daylight savings adjustments are involved.
211 </li><li>Fix interval display to support values exceeding 2^31 hours
212 </li><li>Add overflow checking to timestamp and interval arithmetic
213 </li><li>Extend timezone code to allow 64-bit values so we can
214 represent years beyond 2038
215 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01363.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01363.php</a>
217 </li><li>Use LC_TIME for localized weekday/month names, rather than
219 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00390.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00390.php</a>
221 </li><li>Add ISO INTERVAL handling
224 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-01/msg00250.php">http://archives.postgresql.org/pgsql-hackers/2006-01/msg00250.php</a><br/>
225 <a href="http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php">http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php</a>
228 <li>Support ISO INTERVAL syntax if units cannot be determined from
229 the string, and are supplied after the string
230 <p> The SQL standard states that the units after the string
231 specify the units of the string, e.g. INTERVAL '2' MINUTE
232 should return '00:02:00'. The current behavior has the units
233 restrict the interval value to the specified unit or unit
234 range, INTERVAL '70' SECOND returns '00:00:10'.
236 <p> For syntax that isn't uniquely ISO or PG syntax, like '1' or
237 '1:30', treat as ISO if there is a range specification clause,
238 and as PG if there no clause is present, e.g. interpret '1:30'
239 MINUTE TO SECOND as '1 minute 30 seconds', and interpret
240 '1:30' as '1 hour, 30 minutes'.
242 <p> This makes common cases like SELECT INTERVAL '1' MONTH
243 SQL-standard results. The SQL standard supports a limited
244 number of unit combinations and doesn't support unit names in
245 the string. The PostgreSQL syntax is more flexible in the
246 range of units supported, e.g. PostgreSQL supports '1 year 1
247 hour', while the SQL standard does not.
249 </li><li>Add support for year-month syntax, INTERVAL '50-6' YEAR TO MONTH
250 </li><li>Interpret INTERVAL '1 year' MONTH as CAST (INTERVAL '1 year' AS
251 INTERVAL MONTH), and this should return '12 months'
252 </li><li>Round or truncate values to the requested precision, e.g.
253 INTERVAL '11 months' AS YEAR should return one or zero
254 </li><li>Support precision, CREATE TABLE foo (a INTERVAL MONTH(3))
258 <li>Delay resolution of array expression's data type so assignment
259 coercion can be performed on empty array expressions
260 </li><li>Add support for arrays of domains
261 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00114.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00114.php</a>
263 </li><li>Allow single-byte header storage for array elements
267 <li>Improve vacuum of large objects, like /contrib/vacuumlo?
268 </li><li>Add security checking for large objects
269 </li><li>Auto-delete large objects when referencing row is deleted
270 <p> /contrib/lo offers this functionality.
272 </li><li>Allow read/write into TOAST values like large objects
273 <p> This requires the TOAST column to be stored EXTERNAL.
275 </li><li>Add API for 64-bit large object access
276 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php</a>
279 </li><li>MONEY data type
281 <li>Add locale-aware MONEY type, and support multiple currencies
282 <p> <a href="http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php">http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php</a>
283 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01181.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01181.php</a>
285 </li><li>MONEY dumps in a locale-specific format making it difficult to
286 restore to a system with a different locale
287 </li><li>Allow MONEY to be cast to/from other numeric data types
291 <h1><a name="section_4">Functions</a></h1>
294 <li>Allow INET subnet tests using non-constants to be indexed
295 </li><li>Allow to_date() and to_timestamp() accept localized month names
296 </li><li>Fix to_date()-related functions to consistently issue errors
297 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00915.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00915.php</a>
299 </li><li>Add missing parameter handling in to_char()
300 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php">http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php</a>
302 </li><li>Allow functions to have a schema search path specified at creation time
303 </li><li>Allow substring/replace() to get/set bit values
304 </li><li>Allow to_char() on interval values to accumulate the highest unit
306 <p> Some special format flag would be required to request such
307 accumulation. Such functionality could also be added to EXTRACT.
308 Prevent accumulation that crosses the month/day boundary because of
309 the uneven number of days in a month.
312 <li>to_char(INTERVAL '1 hour 5 minutes', 'MI') => 65
313 </li><li>to_char(INTERVAL '43 hours 20 minutes', 'MI' ) => 2600
314 </li><li>to_char(INTERVAL '43 hours 20 minutes', 'WK:DD:HR:MI') => 0:1:19:20
315 </li><li>to_char(INTERVAL '3 years 5 months','MM') => 41
317 </li><li>Implement inlining of set-returning functions defined in SQL
318 </li><li>Allow SQL-language functions to return results from RETURNING queries
319 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00665.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00665.php</a>
321 </li><li>Allow SQL-language functions to reference parameters by parameter name
322 <p> Currently SQL-language functions can only refer to dollar parameters,
325 </li><li>Add SPI_gettypmod() to return the typemod for a TupleDesc
326 </li><li>Enforce typmod for function inputs, function results and parameters for
327 spi_prepare'd statements called from PLs
328 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01403.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01403.php</a>
330 </li><li>Allow holdable cursors in SPI
331 </li><li>Tighten function permission checks
332 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00568.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00568.php</a>
334 </li><li>Fix IS OF so it matches the ISO specification, and add documentation
335 <p> <a href="http://archives.postgresql.org/pgsql-patches/2003-08/msg00060.php">http://archives.postgresql.org/pgsql-patches/2003-08/msg00060.php</a>
336 <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00060.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00060.php</a>
338 </li><li>Add missing operators for geometric data types
339 <p> Some geometric types do not have the full suite of geometric operators,
342 </li><li>Implement Boyer-Moore searching in strpos()
343 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-08/msg00012.php">http://archives.postgresql.org/pgsql-patches/2007-08/msg00012.php</a>
346 <h1><a name="section_5">Multi-Language Support</a></h1>
349 <li>Add NCHAR (as distinguished from ordinary varchar),
350 </li><li>Allow locale to be set at database creation
351 <p> Currently locale can only be set during initdb. No global tables have
352 locale-aware columns. However, the database template used during
353 database creation might have locale-aware indexes. The indexes would
354 need to be reindexed to match the new locale.
356 </li><li>Allow encoding on a per-column basis optionally using the ICU library:
357 <p> Right now only one encoding is allowed per database. [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
359 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php">http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php</a>
360 <a href="http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php">http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php</a>
361 <a href="http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php">http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php</a>
362 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php</a>
364 </li><li>Add CREATE COLLATE? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
365 </li><li>Support multiple simultaneous character sets, per SQL92
366 </li><li>Improve UTF8 combined character handling?
367 </li><li>Add octet_length_server() and octet_length_client()
368 </li><li>Make octet_length_client() the same as octet_length()?
369 </li><li>Fix problems with wrong runtime encoding conversion for NLS message files
370 </li><li>Add URL to more complete multi-byte regression tests
371 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php">http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php</a>
373 </li><li>Fix ILIKE and regular expressions to handle case insensitivity
374 properly in multibyte encodings
375 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php">http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php</a>
376 <a href="http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php">http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php</a>
378 </li><li>Set client encoding based on the client operating system encoding
379 <p> Currently client_encoding is set in postgresql.conf, which
380 defaults to the server encoding.
382 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg01696.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg01696.php</a>
384 </li><li>Fix cases where invalid byte encodings are accepted by the database,
385 but throw an error on SELECT
386 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00767.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00767.php</a>
389 <h1><a name="section_6">Views / Rules</a></h1>
392 <li>Automatically create rules on views so they are updateable, per SQL99
393 <p> We can only auto-create rules for simple views. For more complex
394 cases users will still have to write rules manually.
396 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php</a>
397 <a href="http://archives.postgresql.org/pgsql-patches/2006-08/msg00255.php">http://archives.postgresql.org/pgsql-patches/2006-08/msg00255.php</a>
399 </li><li>Add the functionality for WITH CHECK OPTION clause of CREATE VIEW
400 </li><li>Allow NOTIFY in rules involving conditionals
401 </li><li>Allow VIEW/RULE recompilation when the underlying tables change
402 <p> Another issue is whether underlying table changes should be reflected
403 in the view, e.g. should SELECT * show additional columns if they
404 are added after the view is created.
406 </li><li>Make it possible to use RETURNING together with conditional DO INSTEAD
407 rules, such as for partitioning setups
408 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-09/msg00577.php">http://archives.postgresql.org/pgsql-hackers/2007-09/msg00577.php</a>
411 <h1><a name="section_7">SQL Commands</a></h1>
414 <li>Add CORRESPONDING BY to UNION/INTERSECT/EXCEPT
415 </li><li>Add ROLLUP, CUBE, GROUPING SETS options to GROUP BY
416 </li><li>%Allow SET CONSTRAINTS to be qualified by schema/table name
417 </li><li>%Add a separate TRUNCATE permission
418 <p> Currently only the owner can TRUNCATE a table because triggers are not
419 called, and the table is locked in exclusive mode.
421 </li><li>Allow PREPARE of cursors
422 </li><li>Allow finer control over the caching of prepared query plans
423 <p> Currently queries prepared via the libpq API are planned on first
424 execute using the supplied parameters --- allow SQL PREPARE to do the
425 same. Also, allow control over replanning prepared queries either
426 manually or automatically when statistics for execute parameters
427 differ dramatically from those used during planning.
429 </li><li>Improve logging of prepared statements recovered during startup
430 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00092.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00092.php</a>
432 </li><li>Allow LISTEN/NOTIFY to store info in memory rather than tables?
433 <p> Currently LISTEN/NOTIFY information is stored in pg_listener. Storing
434 such information in memory would improve performance.
436 </li><li>Add optional textual message to NOTIFY
437 <p> This would allow an informational message to be added to the notify
438 message, perhaps indicating the row modified or other custom
441 </li><li>Add a GUC variable to warn about non-standard SQL usage in queries
442 </li><li>Add SQL-standard MERGE command, typically used to merge two tables
443 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
444 <p> This is similar to UPDATE, then for unmatched rows, INSERT.
445 Whether concurrent access allows modifications which could cause
446 row loss is implementation independent.
448 </li><li>Add REPLACE or UPSERT command that does UPDATE, or on failure, INSERT
449 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
450 <p> To implement this cleanly requires that the table have a unique index
451 so duplicate checking can be easily performed. It is possible to
452 do it without a unique index if we require the user to LOCK the table
455 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00501.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00501.php</a>
456 <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00536.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00536.php</a>
458 </li><li>Add NOVICE output level for helpful messages like automatic sequence/index
460 </li><li>Add GUC to issue notice about statements that use unjoined tables
461 </li><li>Allow EXPLAIN to identify tables that were skipped because of
463 </li><li>Allow EXPLAIN output to be more easily processed by scripts, perhaps XML
464 </li><li>Enable standard_conforming_strings
465 </li><li>Make standard_conforming_strings the default in 8.4?
466 <p> When this is done, backslash-quote should be prohibited in non-E''
467 strings because of possible confusion over how such strings treat
468 backslashes. Basically, '' is always safe for a literal single
469 quote, while \' might or might not be based on the backslash
472 </li><li>Simplify dropping roles that have objects in several databases
473 </li><li>Allow COMMENT ON to accept an expression rather than just a string
474 </li><li>Allow the count returned by SELECT, etc to be to represent as an int64
475 to allow a higher range of values
476 </li><li>Add SQL99 WITH clause to SELECT
477 </li><li>Add SQL:2003 WITH RECURSIVE (hierarchical) queries to SELECT
478 </li><li>Add DEFAULT .. AS OWNER so permission checks are done as the table
480 <p> This would be useful for SERIAL nextval() calls and CHECK constraints.
482 </li><li>Allow DISTINCT to work in multiple-argument aggregate calls
483 </li><li>Add column to pg_stat_activity that shows the progress of long-running
484 commands like CREATE INDEX and VACUUM
485 </li><li>Implement SQL:2003 window functions
486 </li><li>Improve failure message when DROP DATABASE is used on a database that
487 has prepared transactions
488 </li><li>Allow INSERT/UPDATE ... RETURNING inside a SELECT 'FROM' clause
489 <p> <a href="http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php">http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php</a>
490 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php</a>
492 </li><li>Increase locking when DROPing objects so dependent objects cannot
493 get dropped while the DROP operation is happening
494 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00937.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00937.php</a>
498 <li>Allow CREATE TABLE AS to determine column lengths for complex
499 expressions like SELECT col1 || col2
500 </li><li>Use more reliable method for CREATE DATABASE to get a consistent
502 </li><li>Fix transaction restriction checks for CREATE DATABASE and
504 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00133.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00133.php</a>
506 </li><li>Have WITH CONSTRAINTS also create constraint indexes
507 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-04/msg00149.php">http://archives.postgresql.org/pgsql-patches/2007-04/msg00149.php</a>
512 <li>Allow UPDATE tab SET ROW (col, ...) = (SELECT...)
513 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php</a>
514 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00865.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00865.php</a>
516 </li><li>Research self-referential UPDATEs that see inconsistent row versions
517 in read-committed mode
518 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00507.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00507.php</a>
519 <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00016.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00016.php</a>
521 </li><li>Allow GLOBAL temporary tables to exist as empty by default in
523 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-07/msg00006.php">http://archives.postgresql.org/pgsql-hackers/2007-07/msg00006.php</a>
528 <li>%Have ALTER TABLE RENAME rename SERIAL sequence names
529 </li><li>Have ALTER SEQUENCE RENAME rename the sequence name stored
530 in the sequence table
531 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-09/msg00092.php">http://archives.postgresql.org/pgsql-bugs/2007-09/msg00092.php</a>
532 <a href="http://archives.postgresql.org/pgsql-bugs/2007-10/msg00007.php">http://archives.postgresql.org/pgsql-bugs/2007-10/msg00007.php</a>
534 </li><li>Add ALTER DOMAIN to modify the underlying data type
535 </li><li>%Allow ALTER TABLE ... ALTER CONSTRAINT ... RENAME
536 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php</a>
538 </li><li>%Allow ALTER TABLE to change constraint deferrability and actions
539 </li><li>Add missing object types for ALTER ... SET SCHEMA
540 </li><li>Allow ALTER TABLESPACE to move to different directories
541 </li><li>Allow databases to be moved to different tablespaces
542 </li><li>Allow moving system tables to other tablespaces, where possible
543 <p> Currently non-global system tables must be in the default database
544 tablespace. Global system tables can never be moved.
546 </li><li>Prevent parent tables from altering or dropping constraints
547 like CHECK that are inherited by child tables unless CASCADE
549 </li><li>%Prevent child tables from altering or dropping constraints
550 like CHECK that were inherited from the parent table
551 </li><li>Have ALTER INDEX update the name of a constraint using that index
552 </li><li>Add ALTER TABLE RENAME CONSTRAINT, update index name also
553 </li><li>Allow column display reordering by recording a display,
554 storage, and permanent id for every column?
555 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00782.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00782.php</a>
560 <li>Automatically maintain clustering on a table
561 <p> This might require some background daemon to maintain clustering
562 during periods of low usage. It might also require tables to be only
563 partially filled for easier reorganization. Another idea would
564 be to create a merged heap/index data file so an index lookup would
565 automatically access the heap data too. A third idea would be to
566 store heap rows in hashed groups, perhaps using a user-supplied
568 <a href="http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php">http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php</a>
570 </li><li>%Add default clustering to system tables
571 <p> To do this, determine the ideal cluster index for each system
572 table and set the cluster setting during initdb.
574 </li><li>%Add VERBOSE option to report tables as they are processed,
579 <li>Allow COPY to report error lines and continue
580 <p> This requires the use of a savepoint before each COPY line is
581 processed, with ROLLBACK on COPY failure.
583 </li><li>Allow COPY on a newly-created table to skip WAL logging
584 <p> On crash recovery, the table involved in the COPY would
585 be removed or have its heap and index files truncated. One
586 issue is that no other backend should be able to add to
587 the table at the same time, which is something that is
591 </li><li>GRANT/REVOKE
593 <li>Allow column-level privileges
594 </li><li>%Allow GRANT/REVOKE permissions to be applied to all schema objects
596 <p> The proposed syntax is:
597 </p><p> GRANT SELECT ON ALL TABLES IN public TO phpuser;
598 GRANT SELECT ON NEW TABLES IN public TO phpuser;
600 </li><li>Allow GRANT/REVOKE permissions to be inherited by objects based on
602 </li><li>Allow SERIAL sequences to inherit permissions from the base table?
606 <li>Prevent DROP TABLE from dropping a row referenced by its own open
611 <li>Allow INSERT/UPDATE of the system-generated oid value for a row
612 </li><li>In rules, allow VALUES() to contain a mixture of 'old' and 'new'
617 <li>Add SET PERFORMANCE_TIPS option to suggest INDEX, VACUUM, VACUUM
619 </li><li>Add SET PATH for schemas?
620 <p> This is basically the same as SET search_path.
623 </li><li>Referential Integrity
625 <li>Add MATCH PARTIAL referential integrity
626 </li><li>Change foreign key constraint for array -> element to mean element
628 </li><li>Enforce referential integrity for system tables
629 </li><li>Fix problem when cascading referential triggers make changes on
630 cascaded tables, seeing the tables in an intermediate state
631 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
632 <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
634 </li><li>Allow DEFERRABLE and end-of-statement UNIQUE constraints?
635 <p> This would allow UPDATE tab SET col = col + 1 to work if col has
636 a unique index. Currently, uniqueness checks are done while the
637 command is being executed, rather than at the end of the statement
640 <p> <a href="http://people.planetpostgresql.org/greg/index.php?/archives/2006/06/10.html">http://people.planetpostgresql.org/greg/index.php?/archives/2006/06/10.html</a>
641 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php</a>
644 </li><li>Server-Side Languages
648 <li>Fix RENAME to work on variables other than OLD/NEW
649 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2002-03/msg00591.php">http://archives.postgresql.org/pgsql-hackers/2002-03/msg00591.php</a>
650 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01615.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01615.php</a>
651 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01587.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01587.php</a>
653 </li><li>Allow function parameters to be passed by name,
654 get_employee_salary(12345 AS emp_id, 2001 AS tax_year)
655 </li><li>Add Oracle-style packages (Pavel)
656 <p> A package would be a schema with session-local variables,
657 public/private functions, and initialization functions. It
658 is also possible to implement these capabilities
659 in all schemas and not use a separate "packages"
662 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php</a>
664 </li><li>Allow handling of %TYPE arrays, e.g. tab.col%TYPE[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?"></a>]
665 </li><li>Allow listing of record column names, and access to
666 record columns via variables, e.g. columns := r.(*),
668 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php</a>
669 <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php</a>
670 <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php</a>
672 </li><li>Add single-step debugging of functions
673 </li><li>Add support for SCROLL cursors
674 </li><li>Add support for WITH HOLD cursors
675 </li><li>Allow RETURN to return row or record functions
676 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-11/msg00045.php">http://archives.postgresql.org/pgsql-patches/2005-11/msg00045.php</a>
677 <a href="http://archives.postgresql.org/pgsql-patches/2006-08/msg00397.php">http://archives.postgresql.org/pgsql-patches/2006-08/msg00397.php</a>
678 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg00388.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg00388.php</a>
680 </li><li>Fix problems with RETURN NEXT on tables with
681 dropped/added columns after function creation
682 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00165.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00165.php</a>
684 </li><li>Allow row and record variables to be set to NULL constants,
685 and allow NULL tests on such variables
686 <p> Because a row is not scalar, do not allow assignment
687 from NULL-valued scalars.
689 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00070.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00070.php</a>
694 <li>Add table function support to pltcl, plpythonu
695 </li><li>Add support for polymorphic arguments and return types to
696 languages other than PL/PgSQL
697 </li><li>Add capability to create and call PROCEDURES
698 </li><li>Add support for OUT and INOUT parameters to languages other
700 </li><li>Add PL/PythonU tracebacks
701 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php</a>
703 </li><li>Allow data to be passed in native language formats, rather
705 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00289.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00289.php</a>
710 <h1><a name="section_8">Clients</a></h1>
713 <li>Have pg_ctl look at PGHOST in case it is a socket directory?
714 </li><li>Allow pg_ctl to work properly with configuration files located outside
716 <p> pg_ctl can not read the pid file because it isn't located in the
717 config directory but in the PGDATA directory. The solution is to
718 allow pg_ctl to read and understand postgresql.conf to find the
719 data_directory value.
721 </li><li>Make consistent use of long/short command options --- pg_ctl needs
722 long ones, pg_config doesn't have short ones, postgres doesn't have
723 enough long ones, etc.
726 <li>Have psql show current values for a sequence
727 </li><li>Move psql backslash database information into the backend, use
728 mnemonic commands? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?psql">psql</a>]
729 <p> This would allow non-psql clients to pull the same information out
730 of the database as psql.
732 </li><li>Fix psql's \d commands more consistent
733 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
734 <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
736 </li><li>Consistently display privilege information for all objects in psql
737 </li><li>Add auto-expanded mode so expanded output is used if the row
738 length is wider than the screen width.
739 <p> Consider using auto-expanded mode for backslash commands like \df+.
741 </li><li>Prevent tab completion of SET TRANSACTION from querying the
742 database and therefore preventing the transaction isolation
743 level from being set.
744 <p> Currently SET <tab> causes a database lookup to check all
745 supported session variables. This query causes problems
746 because setting the transaction isolation level must be the
747 first statement of a transaction.
749 </li><li>Consider parsing the -c string into individual queries so each
750 is run in its own transaction
751 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00291.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00291.php</a>
753 </li><li>Add a \set variable to control whether \s displays line numbers
754 <p> Another option is to add \# which lists line numbers, and
755 allows command execution.
757 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00255.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00255.php</a>
762 <li>%Add dumping of comments on index columns and composite type columns
763 </li><li>%Add full object name to the tag field. eg. for operators we need
764 '=(integer, integer)', instead of just '='.
765 </li><li>Add pg_dumpall custom format dumps?
766 </li><li>Remove unnecessary function pointer abstractions in pg_dump source
768 </li><li>Allow selection of individual object(s) of all types, not just
770 </li><li>In a selective dump, allow dumping of an object and all its
772 </li><li>Add options like pg_restore -l and -L to pg_dump
773 </li><li>Stop dumping CASCADE on DROP TYPE commands in clean mode
774 </li><li>Allow pg_dump --clean to drop roles that own objects or have
776 </li><li>Change pg_dump so that a comment on the dumped database is
777 applied to the loaded database, even if the database has a
778 different name. This will require new backend syntax, perhaps
779 COMMENT ON CURRENT DATABASE.
784 <p> Document differences between ecpg and the SQL standard and
785 information about the Informix-compatibility module.
787 </li><li>Solve cardinality > 1 for input descriptors / variables?
788 </li><li>Add a semantic check level, e.g. check if a table really exists
789 </li><li>fix handling of DB attributes that are arrays
790 </li><li>Use backend PREPARE/EXECUTE facility for ecpg where possible
791 </li><li>Implement SQLDA
792 </li><li>Fix nested C comments
793 </li><li>%sqlwarn[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?6">6</a>] should be 'W' if the PRECISION or SCALE value specified
794 </li><li>Make SET CONNECTION thread-aware, non-standard?
795 </li><li>Allow multidimensional arrays
796 </li><li>Add internationalized message strings
797 </li><li>Implement COPY FROM STDIN
801 <li>Add PQescapeIdentifierConn()
802 </li><li>Prevent PQfnumber() from lowercasing unquoted the column name
803 <p> PQfnumber() should never have been doing lowercasing, but
804 historically it has so we need a way to prevent it
806 </li><li>Allow statement results to be automatically batched to the client
807 <p> Currently all statement results are transferred to the libpq
808 client before libpq makes the results available to the
809 application. This feature would allow the application to make
810 use of the first result rows while the rest are transferred, or
811 held on the server waiting for them to be requested by libpq.
812 One complexity is that a statement like SELECT 1/col could error
813 out mid-way through the result set.
814 <li>Fix SSL retry to avoid useless repeated connection attempts and
815 ensuing misleading error messages
816 </li><li>Consider disallowing multiple queries in PQexec() as an
817 additional barrier to SQL injection attacks
818 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00184.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00184.php</a>
820 </li><li>Add PQexecf() that allows complex parameter substitution
821 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01803.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01803.php</a>
825 <h1><a name="section_9">Triggers</a></h1>
828 <li>Add deferred trigger queue file
829 <p> Right now all deferred trigger information is stored in backend
830 memory. This could exhaust memory for very large trigger queues.
831 This item involves dumping large queues into files.
833 </li><li>Allow triggers to be disabled in only the current session.
834 <p> This is currently possible by starting a multi-statement transaction,
835 modifying the system tables, performing the desired SQL, restoring the
836 system tables, and committing the transaction. ALTER TABLE ...
837 TRIGGER requires a table lock so it is not ideal for this usage.
839 </li><li>With disabled triggers, allow pg_dump to use ALTER TABLE ADD FOREIGN KEY
840 <p> If the dump is known to be valid, allow foreign keys to be added
841 without revalidating the data.
843 </li><li>Allow statement-level triggers to access modified rows
844 </li><li>Support triggers on columns
845 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php</a>
847 </li><li>Allow AFTER triggers on system tables
848 <p> System tables are modified in many places in the backend without going
849 through the executor and therefore not causing triggers to fire. To
850 complete this item, the functions that modify system tables will have
853 </li><li>Tighten trigger permission checks
854 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00564.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00564.php</a>
856 </li><li>Allow BEFORE INSERT triggers on views
857 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-02/msg01466.php">http://archives.postgresql.org/pgsql-general/2007-02/msg01466.php</a>
860 <h1><a name="section_10">Indexes</a></h1>
863 <li>Add UNIQUE capability to non-btree indexes
864 </li><li>Prevent index uniqueness checks when UPDATE does not modify the column
865 <p> Uniqueness (index) checks are done when updating a column even if the
866 column is not modified by the UPDATE.
868 </li><li>Allow the creation of on-disk bitmap indexes which can be quickly
869 combined with other bitmap indexes
870 <p> Such indexes could be more compact if there are only a few distinct values.
871 Such indexes can also be compressed. Keeping such indexes updated can be
874 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php</a>
875 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01107.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01107.php</a>
877 </li><li>Allow accurate statistics to be collected on indexes with more than
878 one column or expression indexes, perhaps using per-index statistics
879 <p> <a href="http://archives.postgresql.org/pgsql-performance/2006-10/msg00222.php">http://archives.postgresql.org/pgsql-performance/2006-10/msg00222.php</a>
880 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01131.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01131.php</a>
882 </li><li>Consider compressing indexes by storing key values duplicated in
883 several rows as a single index entry
884 <p> This is difficult because it requires datatype-specific knowledge.
886 </li><li>Add REINDEX CONCURRENTLY, like CREATE INDEX CONCURRENTLY
887 <p> This is difficult because you must upgrade to an exclusive table lock
888 to replace the existing index file. CREATE INDEX CONCURRENTLY does not
889 have this complication. This would allow index compaction without
892 </li><li>Allow multiple indexes to be created concurrently, ideally via a
893 single heap scan, and have a restore of a pg_dump somehow use it
894 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-05/msg01274.php">http://archives.postgresql.org/pgsql-general/2007-05/msg01274.php</a>
898 <li>Allow inherited tables to inherit indexes, UNIQUE constraints,
899 and primary/foreign keys
900 </li><li>Honor UNIQUE INDEX on base column in INSERTs/UPDATEs
901 on inherited table, e.g. INSERT INTO inherit_table
902 (unique_index_col) VALUES (dup) should fail
903 <p> The main difficulty with this item is the problem of
904 creating an index that can span multiple tables.
906 </li><li>Allow SELECT ... FOR UPDATE on inherited tables
910 <li>Add more GIST index support for geometric data types
911 </li><li>Allow GIST indexes to create certain complex index types, like
912 digital trees (see Aoki)
915 <a href="http://archives.postgresql.org/pgsql-hackers/2007-09/msg00051.php">http://archives.postgresql.org/pgsql-hackers/2007-09/msg00051.php</a>
917 <li>Pack hash index buckets onto disk pages more efficiently
918 <p> Currently only one hash bucket can be stored on a page. Ideally
919 several hash buckets could be stored on a single page and greater
920 granularity used for the hash algorithm.
922 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-06/msg00168.php">http://archives.postgresql.org/pgsql-hackers/2004-06/msg00168.php</a>
924 </li><li>Consider sorting hash buckets so entries can be found using a
925 binary search, rather than a linear scan
926 </li><li>In hash indexes, consider storing the hash value with or instead
928 </li><li>Add WAL logging for crash recovery
929 </li><li>Allow multi-column hash indexes
930 </li><li>During index creation, pre-sort the tuples to improve build speed
931 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01199.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01199.php</a>
935 <h1><a name="section_11">Fsync</a></h1>
938 <li>Improve commit_delay handling to reduce fsync()
939 </li><li>Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options
940 <p> Ideally this requires a separate test program that can be run
941 at initdb time or optionally later. Consider O_SYNC when
944 </li><li>%Add an option to sync() before fsync()'ing checkpoint files
945 </li><li>Add program to test if fsync has a delay compared to non-fsync
947 <h1><a name="section_12">Cache Usage</a></h1>
950 <li>Speed up COUNT(*)
951 <p> We could use a fixed row count and a +/- count to follow MVCC
952 visibility rules, or a single cached value could be used and
953 invalidated if anyone modifies the table. Another idea is to
954 get a count directly from a unique index, but for this to be
955 faster than a sequential scan it must avoid access to the heap
956 to obtain tuple visibility information.
958 </li><li>Provide a way to calculate an "estimated COUNT(*)"
959 <p> Perhaps by using the optimizer's cardinality estimates or random
962 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php</a>
964 </li><li>Allow data to be pulled directly from indexes
965 <p> Currently indexes do not have enough tuple visibility information
966 to allow data to be pulled from the index without also accessing
967 the heap. One way to allow this is to set a bit on index tuples
968 to indicate if a tuple is currently visible to all transactions
969 when the first valid heap lookup happens. This bit would have to
970 be cleared when a heap tuple is expired.
972 <p> Another idea is to maintain a bitmap of heap pages where all rows
973 are visible to all backends, and allow index lookups to reference
974 that bitmap to avoid heap lookups, perhaps the same bitmap we might
975 add someday to determine which heap pages need vacuuming. Frequently
976 accessed bitmaps would have to be stored in shared memory. One 8k
977 page of bitmaps could track 512MB of heap pages.
979 <p> A third idea would be for a heap scan to check if all rows are visible
980 and if so set a per-table flag which can be checked by index scans.
981 Any change to the table would have to clear the flag. To detect
982 changes during the heap scan a counter could be set at the start and
983 checked at the end --- if it is the same, the table has not been
984 modified --- any table change would increment the counter.
986 </li><li>Consider automatic caching of statements at various levels:
988 <li>Parsed query tree
989 </li><li>Query execute plan
990 </li><li>Query results
992 </li><li>Consider increasing internal areas when shared buffers is increased
993 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php">http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php</a>
995 </li><li>Consider decreasing the amount of memory used by PrivateRefCount
996 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00797.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00797.php</a>
997 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php</a>
1000 <h1><a name="section_13">Vacuum</a></h1>
1003 <li>Improve speed with indexes
1004 <p> For large table adjustments during VACUUM FULL, it is faster to cluster
1005 or reindex rather than update the index. Also, index updates can bloat
1008 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00024.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00024.php</a>
1009 <a href="http://archives.postgresql.org/pgsql-performance/2007-05/msg00296.php">http://archives.postgresql.org/pgsql-performance/2007-05/msg00296.php</a>
1011 </li><li>Reduce lock time during VACUUM FULL by moving tuples with read lock,
1012 then write lock and truncate table
1013 <p> Moved tuples are invisible to other backends so they don't require a
1014 write lock. However, the read lock promotion to write lock could lead
1015 to deadlock situations.
1017 </li><li>Auto-fill the free space map by scanning the buffer cache or by
1018 checking pages written by the background writer
1019 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php">http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php</a>
1020 <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php</a>
1022 </li><li>Create a bitmap of pages that need vacuuming
1023 <p> Instead of sequentially scanning the entire table, have the background
1024 writer or some other process record pages that have expired rows, then
1025 VACUUM can look at just those pages rather than the entire table. In
1026 the event of a system crash, the bitmap would probably be invalidated.
1027 One complexity is that index entries still have to be vacuumed, and
1028 doing this without an index scan (by using the heap values to find the
1029 index entry) might be slow and unreliable, especially for user-defined
1032 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01188.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01188.php</a>
1033 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00121.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00121.php</a>
1035 </li><li>Allow FSM to return free space toward the beginning of the heap file,
1036 in hopes that empty pages at the end can be truncated by VACUUM
1037 </li><li>Allow FSM page return free space based on table clustering, to assist
1038 in maintaining clustering?
1039 </li><li>Improve dead row detection during multi-statement transactions usage
1040 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-03/msg00358.php">http://archives.postgresql.org/pgsql-patches/2007-03/msg00358.php</a>
1042 </li><li>Prevent long-lived temporary tables from causing frozen-xid advancement
1045 <p> The problem is that autovacuum cannot vacuum them to set frozen xids;<br/>
1046 only the session that created them can do that.
1051 <li>Use free-space map information to guide refilling
1052 </li><li>%Issue log message to suggest VACUUM FULL if a table is nearly
1054 </li><li>Consider logging activity either to the logs or a system view
1055 </li><li>Improve control of auto-vacuum
1056 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00876.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00876.php</a>
1058 </li><li>Prevent long-lived temporary tables from causing frozen-xid
1059 advancement starvation
1060 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-06/msg01645.php">http://archives.postgresql.org/pgsql-general/2007-06/msg01645.php</a>
1064 <h1><a name="section_14">Locking</a></h1>
1067 <li>Fix priority ordering of read and write light-weight locks (Neil)
1068 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00893.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00893.php</a>
1069 <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00905.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00905.php</a>
1071 </li><li>Fix problem when multiple subtransactions of the same outer transaction
1072 hold different types of locks, and one subtransaction aborts
1073 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg01011.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg01011.php</a>
1074 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00001.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00001.php</a>
1075 <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00435.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00435.php</a>
1077 </li><li>Allow UPDATEs on only non-referential integrity columns not to conflict
1078 with referential integrity locks
1079 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00073.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00073.php</a>
1081 </li><li>Add idle_in_transaction_timeout GUC so locks are not held for long
1084 <h1><a name="section_15">Startup Time Improvements</a></h1>
1087 <li>Experiment with multi-threaded backend for backend creation [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?thread">thread</a>]
1088 <p> This would prevent the overhead associated with process creation. Most
1089 operating systems have trivial process creation time compared to
1090 database startup overhead, but a few operating systems (Win32,
1091 Solaris) might benefit from threading. Also explore the idea of
1092 a single session using multiple threads to execute a statement faster.
1094 </li><li>Experiment with multi-threaded backend better resource utilization
1095 <p> This would allow a single query to make use of multiple CPU's or
1096 multiple I/O channels simultaneously. One idea is to create a
1097 background reader that can pre-fetch sequential and index scan
1098 pages needed by other backends. This could be expanded to allow
1099 concurrent reads from multiple devices in a partitioned table.
1101 </li><li>Add connection pooling
1102 <p> It is unclear if this should be done inside the backend code or done
1103 by something external like pgpool. The passing of file descriptors to
1104 existing backends is one of the difficulties with a backend approach.
1107 <h1><a name="section_16">Write-Ahead Log</a></h1>
1110 <li>Eliminate need to write full pages to WAL before page modification [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?wal">wal</a>]
1111 <p> Currently, to protect against partial disk page writes, we write
1112 full page images to WAL before they are modified so we can correct any
1113 partial page writes during recovery. These pages can also be
1114 eliminated from point-in-time archive files.
1117 <li>When off, write CRC to WAL and check file system blocks
1119 <p> If CRC check fails during recovery, remember the page in case
1120 a later CRC for that page properly matches.
1122 </li><li>Write full pages during file system write and not when
1123 the page is modified in the buffer cache
1124 <p> This allows most full page writes to happen in the background
1125 writer. It might cause problems for applying WAL on recovery
1126 into a partially-written page, but later the full page will be
1130 </li><li>Allow WAL traffic to be streamed to another server for stand-by
1132 </li><li>Reduce WAL traffic so only modified values are written rather than
1134 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01589.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01589.php</a>
1136 </li><li>Allow WAL information to recover corrupted pg_controldata
1137 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php</a>
1139 </li><li>Find a way to reduce rotational delay when repeatedly writing
1141 <p> Currently fsync of WAL requires the disk platter to perform a full
1142 rotation to fsync again. One idea is to write the WAL to different
1143 offsets that might reduce the rotational delay.
1145 </li><li>Allow WAL logging to be turned off for a table, but the table
1146 might be dropped or truncated during crash recovery [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
1147 <p> Allow tables to bypass WAL writes and just fsync() dirty pages on
1148 commit. This should be implemented using ALTER TABLE, e.g. ALTER
1149 TABLE PERSISTENCE [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo? DROP | TRUNCATE | DEFAULT "> DROP | TRUNCATE | DEFAULT </a>]. Tables using
1150 non-default logging should not use referential integrity with
1151 default-logging tables. A table without dirty buffers during a
1152 crash could perhaps avoid the drop/truncate.
1154 </li><li>Allow WAL logging to be turned off for a table, but the table would
1155 avoid being truncated/dropped [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
1156 <p> To do this, only a single writer can modify the table, and writes
1157 must happen only on new pages so the new pages can be removed during
1158 crash recovery. Readers can continue accessing the table. Such
1159 tables probably cannot have indexes. One complexity is the handling
1160 of indexes on TOAST tables.
1163 <h1><a name="section_17">Optimizer / Executor</a></h1>
1166 <li>Improve selectivity functions for geometric operators
1167 </li><li>Precompile SQL functions to avoid overhead
1168 </li><li>Create utility to compute accurate random_page_cost value
1169 </li><li>Improve ability to display optimizer analysis using OPTIMIZER_DEBUG
1170 </li><li>Have EXPLAIN ANALYZE issue NOTICE messages when the estimated and
1171 actual row counts differ by a specified percentage
1172 </li><li>Consider using hash buckets to do DISTINCT, rather than sorting
1173 <p> This would be beneficial when there are few distinct values. This is
1174 already used by GROUP BY.
1176 </li><li>Log statements where the optimizer row estimates were dramatically
1177 different from the number of rows actually found?
1178 </li><li>Consider compressed annealing to search for query plans
1179 <p> This might replace GEQO, <a href="http://sixdemonbag.org/Djinni">http://sixdemonbag.org/Djinni</a>.
1181 </li><li>Improve merge join performance by allowing mark/restore of
1183 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00096.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00096.php</a>
1186 <h1><a name="section_18">Miscellaneous Performance</a></h1>
1189 <li>Do async I/O for faster random read-ahead of data
1190 <p> Async I/O allows multiple I/O requests to be sent to the disk with
1191 results coming back asynchronously.
1193 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00820.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00820.php</a>
1195 </li><li>Use mmap() rather than SYSV shared memory or to write WAL files?
1196 <p> This would remove the requirement for SYSV SHM but would introduce
1197 portability issues. Anonymous mmap (or mmap to /dev/zero) is required
1198 to prevent I/O overhead.
1200 </li><li>Consider mmap()'ing files into a backend?
1201 <p> Doing I/O to large tables would consume a lot of address space or
1202 require frequent mapping/unmapping. Extending the file also causes
1203 mapping problems that might require mapping only individual pages,
1204 leading to thousands of mappings. Another problem is that there is no
1205 way to <u>prevent</u> I/O to disk from the dirty shared buffers so changes
1206 could hit disk before WAL is written.
1208 </li><li>Add a script to ask system configuration questions and tune postgresql.conf
1209 </li><li>Consider ways of storing rows more compactly on disk
1211 <li>Reduce the row header size?
1212 </li><li>Consider reducing on-disk varlena length from four bytes to
1213 two because a heap row cannot be more than 64k in length
1215 </li><li>Consider increasing NUM_CLOG_BUFFERS
1216 </li><li>Consider having the background writer update the transaction status
1217 hint bits before writing out the page
1218 </li><li>Allow user configuration of TOAST thresholds
1219 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00213.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00213.php</a>
1221 </li><li>Allow configuration of backend priorities via the operating system
1222 <p> Though backend priorities make priority inversion during lock
1223 waits possible, research shows that this is not a huge problem.
1225 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-02/msg00493.php">http://archives.postgresql.org/pgsql-general/2007-02/msg00493.php</a>
1227 </li><li>Consider reducing memory used for shared buffer reference count
1228 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php</a>
1231 <h1><a name="section_19">Source Code</a></h1>
1234 <li>Add use of 'const' for variables in source tree
1235 </li><li>Move some things from /contrib into main tree
1236 </li><li>%Remove warnings created by -Wcast-align
1237 </li><li>Move platform-specific ps status display info from ps_status.c to ports
1238 </li><li>Add optional CRC checksum to heap and index pages
1239 </li><li>Improve documentation to build only interfaces (Marc)
1240 </li><li>Remove or relicense modules that are not under the BSD license, if possible
1241 </li><li>%Remove memory/file descriptor freeing before ereport(ERROR)
1242 </li><li>Acquire lock on a relation before building a relcache entry for it
1243 </li><li>%Promote debug_query_string into a server-side function current_query()
1244 </li><li>Allow cross-compiling by generating the zic database on the target system
1245 </li><li>Improve NLS maintenance of libpgport messages linked onto applications
1246 </li><li>Allow ecpg to work with MSVC and BCC
1247 </li><li>Add xpath_array() to /contrib/xml2 to return results as an array
1248 </li><li>Clean up casting in /contrib/isn
1249 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00245.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00245.php</a>
1251 </li><li>Allow building in directories containing spaces
1252 <p> This is probably not possible because 'gmake' and other compiler tools
1253 do not fully support quoting of paths with spaces.
1255 </li><li>Fix sgmltools so PDFs can be generated with bookmarks
1256 </li><li>Use UTF8 encoding for NLS messages so all server encodings can
1258 </li><li>Update Bonjour to work with newer cross-platform SDK
1259 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg02238.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg02238.php</a>
1260 <a href="http://archives.postgresql.org/pgsql-patches/2006-10/msg00048.php">http://archives.postgresql.org/pgsql-patches/2006-10/msg00048.php</a>
1262 </li><li>Split out libpq pgpass and environment documentation sections to make
1263 it easier for non-developers to find
1264 </li><li>Consider detoasting keys before sorting
1265 </li><li>Consider GnuTLS if OpenSSL license becomes a problem
1266 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php</a>
1267 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01213.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01213.php</a>
1269 </li><li>Use strlcpy() rather than our StrNCpy() macro
1270 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg02108.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg02108.php</a>
1272 </li><li>Consider changing documentation format from SGML to XML
1273 <p> <a href="http://archives.postgresql.org/pgsql-docs/2006-12/msg00152.php">http://archives.postgresql.org/pgsql-docs/2006-12/msg00152.php</a>
1275 </li><li>Consider making NAMEDATALEN more configurable in future releases
1276 </li><li>Update our code to handle 64-bit timezone files to match the zic
1277 source code, which now uses them
1278 </li><li>Have configure choose integer datetimes by default
1279 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00046.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00046.php</a>
1281 </li><li>Support scoped IPv6 addresses
1282 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-05/msg00111.php">http://archives.postgresql.org/pgsql-bugs/2007-05/msg00111.php</a>
1284 </li><li>Consider allowing 64-bit integers and floats to be passed by value on
1286 <p> Also change 32-bit floats (float4) to be passed by value at the same
1291 <li>Remove configure.in check for link failure when cause is found
1292 </li><li>Remove readdir() errno patch when runtime/mingwex/dirent.c rev
1294 </li><li>Remove psql newline patch when we find out why mingw outputs an
1296 </li><li>Allow psql to use readline once non-US code pages work with
1298 </li><li>Re-enable timezone output on log_line_prefix '%t' when a
1299 shorter timezone string is available
1300 </li><li>Fix problem with shared memory on the Win32 Terminal Server
1301 </li><li>Improve signal handling
1302 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php">http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php</a>
1305 </li><li>Wire Protocol Changes
1307 <li>Allow dynamic character set handling
1308 </li><li>Add decoded type, length, precision
1309 </li><li>Use compression?
1310 </li><li>Update clients to use data types, typmod, schema.table.column names
1311 of result sets using new statement protocol
1314 <h1><a name="section_20">Exotic Features</a></h1>
1317 <li>Add pre-parsing phase that converts non-ISO syntax to supported
1319 <p> This could allow SQL written for other databases to run without
1322 </li><li>Allow plug-in modules to emulate features from other databases
1323 </li><li>Allow statements across databases or servers with transaction
1325 <p> This can be done using dblink and two-phase commit.
1327 </li><li>Add the features of packages
1329 <li>Make private objects accessible only to objects in the same schema
1330 </li><li>Allow current_schema.objname to access current schema objects
1331 </li><li>Add session variables
1332 </li><li>Allow nested schemas
1334 </li><li>Consider allowing control of upper/lower case folding of unquoted
1336 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php">http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php</a>
1337 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg01527.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg01527.php</a>
1340 <h1><a name="section_21">Features We Do <u>Not</u> Want</a></h1>
1343 <li>All backends running as threads in a single process (not wanted)
1344 <p> This eliminates the process protection we get from the current setup.
1345 Thread creation is usually the same overhead as process creation on
1346 modern systems, so it seems unwise to use a pure threaded model.
1348 </li><li>Optimizer hints (not wanted)
1349 <p> Optimizer hints are used to work around problems in the optimizer. We
1350 would rather have the problems reported and fixed.
1352 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php</a>
1353 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00517.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00517.php</a>
1354 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00663.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00663.php</a>
1356 </li><li>Allow AS in "SELECT col AS label" to be optional (not wanted)
1357 <p> Because we support postfix operators, it isn't possible to make AS
1358 optional and continue to use bison.
1360 <p> <a href="http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php">http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php</a>
1362 </li><li>Embedded server (not wanted)
1363 <p> While PostgreSQL clients runs fine in limited-resource environments, the
1364 server requires multiple processes and a stable pool of resources to
1365 run reliabily and efficiently. Stripping down the PostgreSQL server
1366 to run in the same process address space as the client application
1367 would add too much complexity and failure cases.</p>
1369 </li></ul></li></ul>