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: Mon Mar 17 14:22:37 EDT 2008
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 administrators to safely terminate individual sessions either
30 via an SQL function or SIGTERM
31 <p> Lock table corruption following SIGTERM of an individual backend
32 has been reported in 8.0. A possible cause was fixed in 8.1, but
33 it is unknown whether other problems exist. This item mostly
34 requires additional testing rather than of writing any new code.
36 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php</a>
37 <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00218.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00218.php</a>
39 </li><li>Check for unreferenced table files created by transactions that were
40 in-progress when the server terminated abruptly
41 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00096.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00096.php</a>
43 </li><li>Set proper permissions on non-system schemas during db creation
44 <p> Currently all schemas are owned by the super-user because they are copied
45 from the template1 database. However, since all objects are inherited
46 from the template database, it is not clear that setting schemas to the db
49 </li><li>Add function to report the time of the most recent server reload
50 </li><li>Allow statistics collector information to be pulled from the collector
51 process directly, rather than requiring the collector to write a
52 filesystem file twice a second?
53 </li><li>Allow statistics last vacuum/analyze execution times to be displayed
54 without requiring stats_row_level to be enabled
55 </li><li>Allow log_min_messages to be specified on a per-module basis
56 <p> This would allow administrators to see more detailed information from
57 specific sections of the backend, e.g. checkpoints, autovacuum, etc.
58 Another idea is to allow separate configuration files for each module,
59 or allow arbitrary SET commands to be passed to them.
61 </li><li>Simplify ability to create partitioned tables
62 <p> This would allow creation of partitioned tables without requiring
63 creation of triggers or rules for INSERT/UPDATE/DELETE, and constraints
64 for rapid partition selection. Options could include range and hash
67 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00375.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00375.php</a>
68 <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00151.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00151.php</a>
70 </li><li>Allow auto-selection of partitioned tables for min/max() operations
71 </li><li>Allow more complex user/database default GUC settings
72 <p> Currently ALTER USER and ALTER DATABASE support per-user and
73 per-database defaults. Consider adding per-user-and-database
74 defaults so things like search_path can be defaulted for a
75 specific user connecting to a specific database.
77 </li><li>Allow custom variable classes that can restrict who can set the values
78 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00911.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00911.php</a>
80 </li><li>Implement the SQL standard mechanism whereby REVOKE ROLE revokes only
81 the privilege granted by the invoking role, and not those granted
83 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-05/msg00010.php">http://archives.postgresql.org/pgsql-bugs/2007-05/msg00010.php</a>
85 </li><li>Allow SSL authentication/encryption over unix domain sockets
86 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-12/msg00924.php">http://archives.postgresql.org/pgsql-hackers/2007-12/msg00924.php</a>
88 </li><li>Allow client certificate names to be checked against the client
90 <p> This is already implemented in
91 libpq/fe-secure.c::verify_peer_name_matches_certificate() but the code
94 </li><li>Configuration files
96 <li>Allow pg_hba.conf to specify host names along with IP addresses
97 <p> Host name lookup could occur when the postmaster reads the
98 pg_hba.conf file, or when the backend starts. Another
99 solution would be to reverse lookup the connection IP and
100 check that hostname against the host names in pg_hba.conf.
101 We could also then check that the host name maps to the IP
104 </li><li>%Allow postgresql.conf file values to be changed via an SQL
105 API, perhaps using SET GLOBAL
106 </li><li>Allow the server to be stopped/restarted via an SQL API
107 </li><li>Issue a warning if a change-on-restart-only postgresql.conf value
108 is modified and the server config files are reloaded
109 </li><li>Consider normalizing fractions in postgresql.conf, perhaps
111 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00550.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00550.php</a>
113 </li><li>Allow Kerberos to disable stripping of realms so we can
114 check the username@realm against multiple realms
115 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-11/msg00009.php">http://archives.postgresql.org/pgsql-hackers/2007-11/msg00009.php</a>
120 <li>Allow a database in tablespace t1 with tables created in
121 tablespace t2 to be used as a template for a new database created
122 with default tablespace t2
123 <p> Currently all objects in the default database tablespace must
124 have default tablespace specifications. This is because new
125 databases are created by copying directories. If you mix default
126 tablespace tables and tablespace-specified tables in the same
127 directory, creating a new database from such a mixed directory
128 would create a new database with tables that had incorrect
129 explicit tablespaces. To fix this would require modifying
130 pg_class in the newly copied database, which we don't currently
133 </li><li>Allow reporting of which objects are in which tablespaces
134 <p> This item is difficult because a tablespace can contain objects
135 from multiple databases. There is a server-side function that
136 returns the databases which use a specific tablespace, so this
137 requires a tool that will call that function and connect to each
138 database to find the objects in each database for that tablespace.
140 </li><li>Allow WAL replay of CREATE TABLESPACE to work when the directory
141 structure on the recovery computer is different from the original
142 </li><li>Allow per-tablespace quotas
144 </li><li>Point-In-Time Recovery (PITR)
146 <li>Allow a warm standby system to also allow read-only statements
147 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pitr">pitr</a>]
148 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00050.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00050.php</a>
150 </li><li>%Create dump tool for write-ahead logs for use in determining
151 transaction id for point-in-time recovery
152 <p> This is useful for checking PITR recovery.
154 </li><li>Allow recovery.conf to support the same syntax as
155 postgresql.conf, including quoting
156 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00497.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00497.php</a>
160 <h1><a name="section_3">Data Types</a></h1>
163 <li>Change NUMERIC to enforce the maximum precision
164 </li><li>Reduce storage space for small NUMERICs
165 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg01331.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg01331.php</a>
166 <a href="http://archives.postgresql.org/pgsql-patches/2007-02/msg00505.php">http://archives.postgresql.org/pgsql-patches/2007-02/msg00505.php</a>
167 <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00715.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00715.php</a>
169 </li><li>Fix data types where equality comparison isn't intuitive, e.g. box
170 </li><li>Add support for public SYNONYMs
171 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php</a>
173 </li><li>Fix CREATE CAST on DOMAINs
174 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php">http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php</a>
175 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php</a>
177 </li><li>Allow domains to be cast
178 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2003-06/msg01206.php">http://archives.postgresql.org/pgsql-hackers/2003-06/msg01206.php</a>
179 <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00289.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00289.php</a>
181 </li><li>Add support for SQL-standard GENERATED/IDENTITY columns
182 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php</a>
183 <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00038.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00038.php</a>
184 <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00344.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00344.php</a>
185 <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00076.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00076.php</a>
186 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg00604.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg00604.php</a>
188 </li><li>Improve XML support
189 <p> <a href="http://developer.postgresql.org/index.php/XML_Support">http://developer.postgresql.org/index.php/XML_Support</a>
191 </li><li>Consider placing all sequences in a single table, or create a system
193 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php">http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php</a>
195 </li><li>Allow the UUID type to accept non-standard formats
196 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg01214.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg01214.php</a>
198 </li><li>Allow text search dictionary to filter out only stop words
199 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-11/msg00081.php">http://archives.postgresql.org/pgsql-patches/2007-11/msg00081.php</a>
201 </li><li>Consider a function-based API for '@@' full text searches
202 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-11/msg00511.php">http://archives.postgresql.org/pgsql-hackers/2007-11/msg00511.php</a>
204 </li><li>Consider a special data type for regular expressions
205 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg01067.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg01067.php</a>
207 </li><li>Dates and Times
209 <li>Allow infinite dates and intervals just like infinite timestamps
210 </li><li>Merge hardwired timezone names with the TZ database; allow either
211 kind everywhere a TZ name is currently taken
212 </li><li>Allow TIMESTAMP WITH TIME ZONE to store the original timezone
213 information, either zone name or offset from UTC [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?timezone">timezone</a>]
214 <p> If the TIMESTAMP value is stored with a time zone name, interval
215 computations should adjust based on the time zone rules.
217 </li><li>Fix SELECT '0.01 years'::interval, '0.01 months'::interval
218 </li><li>Add a GUC variable to allow output of interval values in ISO8601
220 </li><li>Have timestamp subtraction not call justify_hours()?
221 <p> <a href="http://archives.postgresql.org/pgsql-sql/2006-10/msg00059.php">http://archives.postgresql.org/pgsql-sql/2006-10/msg00059.php</a>
223 </li><li>Improve timestamptz subtraction to be DST-aware
224 <p> Currently subtracting one date from another that crosses a
225 daylight savings time adjustment can return '1 day 1 hour', but
226 adding that back to the first date returns a time one hour in
227 the future. This is caused by the adjustment of '25 hours' to
228 '1 day 1 hour', and '1 day' is the same time the next day, even
229 if daylight savings adjustments are involved.
231 </li><li>Fix interval display to support values exceeding 2^31 hours
232 </li><li>Add overflow checking to timestamp and interval arithmetic
233 </li><li>Extend timezone code to allow 64-bit values so we can
234 represent years beyond 2038
235 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01363.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01363.php</a>
237 </li><li>Use LC_TIME for localized weekday/month names, rather than
239 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00390.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00390.php</a>
241 </li><li>Add ISO INTERVAL handling
244 <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/>
245 <a href="http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php">http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php</a>
248 <li>Support ISO INTERVAL syntax if units cannot be determined from
249 the string, and are supplied after the string
250 <p> The SQL standard states that the units after the string
251 specify the units of the string, e.g. INTERVAL '2' MINUTE
252 should return '00:02:00'. The current behavior has the units
253 restrict the interval value to the specified unit or unit
254 range, INTERVAL '70' SECOND returns '00:00:10'.
256 <p> For syntax that isn't uniquely ISO or PG syntax, like '1' or
257 '1:30', treat as ISO if there is a range specification clause,
258 and as PG if there no clause is present, e.g. interpret '1:30'
259 MINUTE TO SECOND as '1 minute 30 seconds', and interpret
260 '1:30' as '1 hour, 30 minutes'.
262 <p> This makes common cases like SELECT INTERVAL '1' MONTH
263 SQL-standard results. The SQL standard supports a limited
264 number of unit combinations and doesn't support unit names in
265 the string. The PostgreSQL syntax is more flexible in the
266 range of units supported, e.g. PostgreSQL supports '1 year 1
267 hour', while the SQL standard does not.
269 </li><li>Add support for year-month syntax, INTERVAL '50-6' YEAR
271 </li><li>Interpret INTERVAL '1 year' MONTH as CAST (INTERVAL '1
272 year' AS INTERVAL MONTH), and this should return '12 months'
273 </li><li>Round or truncate values to the requested precision, e.g.
274 INTERVAL '11 months' AS YEAR should return one or zero
275 </li><li>Support precision, CREATE TABLE foo (a INTERVAL MONTH(3))
279 <li>Delay resolution of array expression's data type so assignment
280 coercion can be performed on empty array expressions
281 </li><li>Add support for arrays of domains
282 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00114.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00114.php</a>
284 </li><li>Allow single-byte header storage for array elements
288 <li>Improve vacuum of large objects, like contrib/vacuumlo?
289 </li><li>Add security checking for large objects
290 </li><li>Auto-delete large objects when referencing row is deleted
291 <p> contrib/lo offers this functionality.
293 </li><li>Allow read/write into TOAST values like large objects
294 <p> This requires the TOAST column to be stored EXTERNAL.
296 </li><li>Add API for 64-bit large object access
297 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php</a>
300 </li><li>MONEY data type
302 <li>Add locale-aware MONEY type, and support multiple currencies
303 <p> <a href="http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php">http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php</a>
304 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01181.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01181.php</a>
306 </li><li>MONEY dumps in a locale-specific format making it difficult to
307 restore to a system with a different locale
308 </li><li>Allow MONEY to be easily cast to/from other numeric data types
312 <h1><a name="section_4">Functions</a></h1>
315 <li>Allow INET subnet tests using non-constants to be indexed
316 </li><li>Allow to_date() and to_timestamp() accept localized month names
317 </li><li>Fix to_date()-related functions to consistently issue errors
318 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00915.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00915.php</a>
320 </li><li>Add missing parameter handling in to_char()
321 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php">http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php</a>
323 </li><li>Allow substring/replace() to get/set bit values
324 </li><li>Allow to_char() on interval values to accumulate the highest unit
326 <p> Some special format flag would be required to request such
327 accumulation. Such functionality could also be added to EXTRACT.
328 Prevent accumulation that crosses the month/day boundary because of
329 the uneven number of days in a month.
332 <li>to_char(INTERVAL '1 hour 5 minutes', 'MI') => 65
333 </li><li>to_char(INTERVAL '43 hours 20 minutes', 'MI' ) => 2600
334 </li><li>to_char(INTERVAL '43 hours 20 minutes', 'WK:DD:HR:MI') => 0:1:19:20
335 </li><li>to_char(INTERVAL '3 years 5 months','MM') => 41
337 </li><li>Implement inlining of set-returning functions defined in SQL
338 </li><li>Allow SQL-language functions to return results from RETURNING queries
339 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00665.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00665.php</a>
341 </li><li>Allow SQL-language functions to reference parameters by parameter name
342 <p> Currently SQL-language functions can only refer to dollar parameters,
345 </li><li>Add SPI_gettypmod() to return the typemod for a TupleDesc
346 </li><li>Enforce typmod for function inputs, function results and parameters for
347 spi_prepare'd statements called from PLs
348 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01403.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01403.php</a>
350 </li><li>Allow holdable cursors in SPI
351 </li><li>Tighten function permission checks
352 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00568.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00568.php</a>
354 </li><li>Fix IS OF so it matches the ISO specification, and add documentation
355 <p> <a href="http://archives.postgresql.org/pgsql-patches/2003-08/msg00060.php">http://archives.postgresql.org/pgsql-patches/2003-08/msg00060.php</a>
356 <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00060.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00060.php</a>
358 </li><li>Add missing operators for geometric data types
359 <p> Some geometric types do not have the full suite of geometric operators,
362 </li><li>Implement Boyer-Moore searching in strpos()
363 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-08/msg00012.php">http://archives.postgresql.org/pgsql-patches/2007-08/msg00012.php</a>
365 </li><li>Prevent malicious functions from being executed with the permissions
366 of unsuspecting users
367 <p> Index functions are safe, so VACUUM and ANALYZE are safe too.
368 Triggers, CHECK and DEFAULT expressions, and rules are still vulnerable.
369 <a href="http://archives.postgresql.org/pgsql-hackers/2008-01/msg00268.php">http://archives.postgresql.org/pgsql-hackers/2008-01/msg00268.php</a>
371 </li><li>Reduce memory usage of aggregates in set returning functions
372 <p> <a href="http://archives.postgresql.org/pgsql-performance/2008-01/msg00031.php">http://archives.postgresql.org/pgsql-performance/2008-01/msg00031.php</a>
374 </li><li>Add temporal versions of generate_series()
375 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg01180.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg01180.php</a>
377 </li><li>Add array_accum() and array_to_set() functions for arrays
378 <p> The standards specify array_agg() and UNNEST.
379 <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00464.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00464.php</a>
382 <h1><a name="section_5">Multi-Language Support</a></h1>
385 <li>Add NCHAR (as distinguished from ordinary varchar),
386 </li><li>Allow locale to be set at database creation
387 <p> Currently locale can only be set during initdb. No global tables have
388 locale-aware columns. However, the database template used during
389 database creation might have locale-aware indexes. The indexes would
390 need to be reindexed to match the new locale.
392 </li><li>Allow encoding on a per-column basis optionally using the ICU library:
393 <p> Right now only one encoding is allowed per database. [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
395 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php">http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php</a>
396 <a href="http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php">http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php</a>
397 <a href="http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php">http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php</a>
398 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php</a>
400 </li><li>Add CREATE COLLATE? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
401 </li><li>Support multiple simultaneous character sets, per SQL92
402 </li><li>Improve UTF8 combined character handling?
403 </li><li>Add octet_length_server() and octet_length_client()
404 </li><li>Make octet_length_client() the same as octet_length()?
405 </li><li>Fix problems with wrong runtime encoding conversion for NLS message files
406 </li><li>Add URL to more complete multi-byte regression tests
407 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php">http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php</a>
409 </li><li>Fix ILIKE and regular expressions to handle case insensitivity
410 properly in multibyte encodings
411 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php">http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php</a>
412 <a href="http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php">http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php</a>
414 </li><li>Set client encoding based on the client operating system encoding
415 <p> Currently client_encoding is set in postgresql.conf, which
416 defaults to the server encoding.
417 <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg01696.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg01696.php</a>
419 </li><li>Change memory allocation for multi-byte functions so memory is
420 allocated inside conversion functions
421 <p> Currently we preallocate memory based on worst-case usage.
424 <h1><a name="section_6">Views / Rules</a></h1>
427 <li>Automatically create rules on views so they are updateable, per SQL99
428 <p> We can only auto-create rules for simple views. For more complex
429 cases users will still have to write rules manually.
431 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php</a>
432 <a href="http://archives.postgresql.org/pgsql-patches/2006-08/msg00255.php">http://archives.postgresql.org/pgsql-patches/2006-08/msg00255.php</a>
434 </li><li>Add the functionality for WITH CHECK OPTION clause of CREATE VIEW
435 </li><li>Allow NOTIFY in rules involving conditionals
436 </li><li>Allow VIEW/RULE recompilation when the underlying tables change
437 <p> Another issue is whether underlying table changes should be reflected
438 in the view, e.g. should SELECT * show additional columns if they
439 are added after the view is created.
441 </li><li>Make it possible to use RETURNING together with conditional DO INSTEAD
442 rules, such as for partitioning setups
443 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-09/msg00577.php">http://archives.postgresql.org/pgsql-hackers/2007-09/msg00577.php</a>
445 </li><li>Add the ability to automatically create materialized views
446 <p> Right now materialized views require the user to create triggers on the
447 main table to keep the summary table current. SQL syntax should be able
448 to manager the triggers and summary table automatically. A more
449 sophisticated implementation would automatically retrieve from the
450 summary table when the main table is referenced, if possible.
453 <h1><a name="section_7">SQL Commands</a></h1>
456 <li>Add CORRESPONDING BY to UNION/INTERSECT/EXCEPT
457 </li><li>Add ROLLUP, CUBE, GROUPING SETS options to GROUP BY
458 </li><li>%Allow SET CONSTRAINTS to be qualified by schema/table name
459 </li><li>%Add a separate TRUNCATE permission
460 <p> Currently only the owner can TRUNCATE a table because triggers are not
461 called, and the table is locked in exclusive mode.
463 </li><li>Allow PREPARE of cursors
464 </li><li>Allow finer control over the caching of prepared query plans
465 <p> Currently queries prepared via the libpq API are planned on first
466 execute using the supplied parameters --- allow SQL PREPARE to do the
467 same. Also, allow control over replanning prepared queries either
468 manually or automatically when statistics for execute parameters
469 differ dramatically from those used during planning.
471 </li><li>Improve logging of prepared transactions recovered during startup
472 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00092.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00092.php</a>
474 </li><li>Allow LISTEN/NOTIFY to store info in memory rather than tables?
475 <p> Currently LISTEN/NOTIFY information is stored in pg_listener. Storing
476 such information in memory would improve performance.
478 </li><li>Add optional textual message to NOTIFY
479 <p> This would allow an informational message to be added to the notify
480 message, perhaps indicating the row modified or other custom
483 </li><li>Allow multiple identical NOTIFY events to always be communicated to the
484 client, rather than sent as a single notification to the listener
485 <p> <a href="http://archives.postgresql.org/pgsql-general/2008-01/msg00057.php">http://archives.postgresql.org/pgsql-general/2008-01/msg00057.php</a>
487 </li><li>Add a GUC variable to warn about non-standard SQL usage in queries
488 </li><li>Add SQL-standard MERGE command, typically used to merge two tables
489 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
490 <p> This is similar to UPDATE, then for unmatched rows, INSERT.
491 Whether concurrent access allows modifications which could cause
492 row loss is implementation independent.
494 </li><li>Add REPLACE or UPSERT command that does UPDATE, or on failure, INSERT
495 [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
496 <p> To implement this cleanly requires that the table have a unique index
497 so duplicate checking can be easily performed. It is possible to
498 do it without a unique index if we require the user to LOCK the table
501 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00501.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00501.php</a>
502 <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00536.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00536.php</a>
504 </li><li>Add NOVICE output level for helpful messages like automatic sequence/index
506 </li><li>Add GUC to issue notice about statements that use unjoined tables
507 </li><li>Allow EXPLAIN to identify tables that were skipped because of
509 </li><li>Allow EXPLAIN output to be more easily processed by scripts, perhaps XML
510 </li><li>Enable standard_conforming_strings
511 </li><li>Make standard_conforming_strings the default in 8.5?
512 <p> When this is done, backslash-quote should be prohibited in non-E''
513 strings because of possible confusion over how such strings treat
514 backslashes. Basically, '' is always safe for a literal single
515 quote, while \' might or might not be based on the backslash
518 </li><li>Simplify dropping roles that have objects in several databases
519 </li><li>Allow COMMENT ON to accept an expression rather than just a string
520 </li><li>Allow the count returned by SELECT, etc to be represented as an int64
521 to allow a higher range of values
522 </li><li>Add SQL99 WITH clause to SELECT
523 </li><li>Add SQL:2003 WITH RECURSIVE (hierarchical) queries to SELECT
524 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01375.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01375.php</a>
525 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg00642.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg00642.php</a>
526 <a href="http://archives.postgresql.org/pgsql-patches/2007-03/msg00139.php">http://archives.postgresql.org/pgsql-patches/2007-03/msg00139.php</a>
528 </li><li>Add DEFAULT .. AS OWNER so permission checks are done as the table
530 <p> This would be useful for SERIAL nextval() calls and CHECK constraints.
532 </li><li>Allow DISTINCT to work in multiple-argument aggregate calls
533 </li><li>Add column to pg_stat_activity that shows the progress of long-running
534 commands like CREATE INDEX and VACUUM
535 </li><li>Implement SQL:2003 window functions
536 </li><li>Improve failure message when DROP DATABASE is used on a database that
537 has prepared transactions
538 </li><li>Allow INSERT/UPDATE ... RETURNING inside a SELECT 'FROM' clause
539 <p> <a href="http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php">http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php</a>
540 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php</a>
542 </li><li>Increase locking when DROPing objects so dependent objects cannot
543 get dropped while the DROP operation is happening
544 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00937.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00937.php</a>
546 </li><li>-<em>Allow AS in "SELECT col AS label" to be optional in certain cases</em>
547 </li><li>Allow INSERT ... DELETE ... RETURNING, namely allow the DELETE ...
548 RETURNING to supply values to the INSERT
549 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/thrd2.php#00979">http://archives.postgresql.org/pgsql-hackers/2008-02/thrd2.php#00979</a>
550 </li><li>Add comments on system tables/columns using the information in
552 <p> Ideally the information would be pulled from the SGML file
557 <li>Allow CREATE TABLE AS to determine column lengths for complex
558 expressions like SELECT col1 || col2
559 </li><li>Have WITH CONSTRAINTS also create constraint indexes
560 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-04/msg00149.php">http://archives.postgresql.org/pgsql-patches/2007-04/msg00149.php</a>
562 </li><li>Have CONSTRAINT cname NOT NULL record the contraint name
563 <p> Right now pg_attribute.attnotnull records the NOT NULL status
564 of the column, but does not record the contraint name
569 <li>Allow UPDATE tab SET ROW (col, ...) = (SELECT...)
570 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php</a>
571 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00865.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00865.php</a>
573 </li><li>Research self-referential UPDATEs that see inconsistent row versions
574 in read-committed mode
575 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00507.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00507.php</a>
576 <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00016.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00016.php</a>
578 </li><li>Allow GLOBAL temporary tables to exist as empty by default in
580 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-07/msg00006.php">http://archives.postgresql.org/pgsql-hackers/2007-07/msg00006.php</a>
585 <li>Have ALTER TABLE RENAME rename SERIAL sequence names
586 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php">http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php</a>
588 </li><li>Have ALTER SEQUENCE RENAME rename the sequence name stored
589 in the sequence table
590 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-09/msg00092.php">http://archives.postgresql.org/pgsql-bugs/2007-09/msg00092.php</a>
591 <a href="http://archives.postgresql.org/pgsql-bugs/2007-10/msg00007.php">http://archives.postgresql.org/pgsql-bugs/2007-10/msg00007.php</a>
592 <a href="http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php">http://archives.postgresql.org/pgsql-hackers/2008-03/msg00008.php</a>
594 </li><li>Add ALTER DOMAIN to modify the underlying data type
595 </li><li>%Allow ALTER TABLE ... ALTER CONSTRAINT ... RENAME
596 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php</a>
598 </li><li>%Allow ALTER TABLE to change constraint deferrability and actions
599 </li><li>Add missing object types for ALTER ... SET SCHEMA
600 </li><li>Allow ALTER TABLESPACE to move to different directories
601 </li><li>Allow databases to be moved to different tablespaces
602 </li><li>Allow moving system tables to other tablespaces, where possible
603 <p> Currently non-global system tables must be in the default database
604 tablespace. Global system tables can never be moved.
606 </li><li>Prevent parent tables from altering or dropping constraints
607 like CHECK that are inherited by child tables unless CASCADE
609 </li><li>%Prevent child tables from altering or dropping constraints
610 like CHECK that were inherited from the parent table
611 </li><li>Have ALTER INDEX update the name of a constraint using that index
612 </li><li>Add ALTER TABLE RENAME CONSTRAINT, update index name also
613 </li><li>Allow column display reordering by recording a display,
614 storage, and permanent id for every column?
615 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00782.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00782.php</a>
620 <li>Automatically maintain clustering on a table
621 <p> This might require some background daemon to maintain clustering
622 during periods of low usage. It might also require tables to be only
623 partially filled for easier reorganization. Another idea would
624 be to create a merged heap/index data file so an index lookup would
625 automatically access the heap data too. A third idea would be to
626 store heap rows in hashed groups, perhaps using a user-supplied
628 <a href="http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php">http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php</a>
630 </li><li>%Add default clustering to system tables
631 <p> To do this, determine the ideal cluster index for each system
632 table and set the cluster setting during initdb.
634 </li><li>%Add VERBOSE option to report tables as they are processed,
639 <li>Allow COPY to report error lines and continue
640 <p> This requires the use of a savepoint before each COPY line is
641 processed, with ROLLBACK on COPY failure.
643 </li><li>Allow COPY on a newly-created table to skip WAL logging
644 <p> On crash recovery, the table involved in the COPY would
645 be removed or have its heap and index files truncated. One
646 issue is that no other backend should be able to add to
647 the table at the same time, which is something that is
648 currently allowed. This currently is done if the table is
649 created inside the same transaction block as the COPY because
650 no other backends can see the table.
652 </li><li>Consider using a ring buffer for COPY FROM
653 <p> <a href="http://archives.postgresql.org/pgsql-patches/2008-02/msg00140.php">http://archives.postgresql.org/pgsql-patches/2008-02/msg00140.php</a>
654 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg01080.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg01080.php</a>
656 </li><li>Allow COPY FROM to create index entries in bulk
657 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg00811.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg00811.php</a>
659 </li><li>Allow COPY in CSV mode to control whether a quoted zero-length
660 string is treated as NULL
661 <p> Currently this is always treated as a zero-length string,
662 which generates an error when loading into an integer column
663 <a href="http://archives.postgresql.org/pgsql-hackers/2007-07/msg00905.php">http://archives.postgresql.org/pgsql-hackers/2007-07/msg00905.php</a>
666 </li><li>GRANT/REVOKE
668 <li>Allow column-level privileges
669 </li><li>%Allow GRANT/REVOKE permissions to be applied to all schema objects
671 <p> The proposed syntax is:
672 </p><p> GRANT SELECT ON ALL TABLES IN public TO phpuser;
673 GRANT SELECT ON NEW TABLES IN public TO phpuser;
675 </li><li>Allow GRANT/REVOKE permissions to be inherited by objects based on
677 </li><li>Allow SERIAL sequences to inherit permissions from the base table?
681 <li>Prevent DROP TABLE from dropping a row referenced by its own open
686 <li>Allow INSERT/UPDATE of the system-generated oid value for a row
687 </li><li>In rules, allow VALUES() to contain a mixture of 'old' and 'new'
692 <li>Add SET PERFORMANCE_TIPS option to suggest INDEX, VACUUM, VACUUM
696 <h1><a name="section_8">Referential Integrity</a></h1>
699 <li>Add MATCH PARTIAL referential integrity
700 </li><li>Change foreign key constraint for array -> element to mean element
702 </li><li>Fix problem when cascading referential triggers make changes on
703 cascaded tables, seeing the tables in an intermediate state
704 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
705 <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
707 </li><li>Allow DEFERRABLE and end-of-statement UNIQUE constraints?
708 <p> This would allow UPDATE tab SET col = col + 1 to work if col has
709 a unique index. Currently, uniqueness checks are done while the
710 command is being executed, rather than at the end of the statement
712 <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>
713 <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php</a>
715 </li><li>Optimize referential integrity checks
716 <p> <a href="http://archives.postgresql.org/pgsql-performance/2005-10/msg00458.php">http://archives.postgresql.org/pgsql-performance/2005-10/msg00458.php</a>
717 <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00744.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00744.php</a>
720 <h1><a name="section_9">Server-Side Languages</a></h1>
725 <li>Fix RENAME to work on variables other than OLD/NEW
726 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2002-03/msg00591.php">http://archives.postgresql.org/pgsql-hackers/2002-03/msg00591.php</a>
727 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01615.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01615.php</a>
728 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg01587.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg01587.php</a>
730 </li><li>Allow function parameters to be passed by name,
731 get_employee_salary(12345 AS emp_id, 2001 AS tax_year)
732 </li><li>Allow handling of %TYPE arrays, e.g. tab.col%TYPE[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?"></a>]
733 </li><li>Allow listing of record column names, and access to
734 record columns via variables, e.g. columns := r.(*),
736 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php</a>
737 <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php</a>
738 <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php</a>
740 </li><li>Add support for SCROLL cursors
741 </li><li>Add support for WITH HOLD cursors
742 </li><li>Allow row and record variables to be set to NULL constants,
743 and allow NULL tests on such variables
744 <p> Because a row is not scalar, do not allow assignment
745 from NULL-valued scalars.
747 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00070.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00070.php</a>
749 </li><li>Review handling of MOVE and FETCH
750 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-04/msg00527.php">http://archives.postgresql.org/pgsql-patches/2007-04/msg00527.php</a>
752 </li><li>Improve logic of determining if an identifier is a a
753 variable or column name
754 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-07/msg00436.php">http://archives.postgresql.org/pgsql-hackers/2007-07/msg00436.php</a>
759 <li>Add table function support to pltcl, plpythonu
760 </li><li>Add support for polymorphic arguments and return types to
761 languages other than PL/PgSQL
762 </li><li>Add capability to create and call PROCEDURES
763 </li><li>Add support for OUT and INOUT parameters to languages other
765 </li><li>Add PL/PythonU tracebacks
766 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php</a>
768 </li><li>Allow data to be passed in native language formats, rather
770 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00289.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00289.php</a>
774 <h1><a name="section_10">Clients</a></h1>
777 <li>Have pg_ctl look at PGHOST in case it is a socket directory?
778 </li><li>Allow pg_ctl to work properly with configuration files located outside
780 <p> pg_ctl can not read the pid file because it isn't located in the
781 config directory but in the PGDATA directory. The solution is to
782 allow pg_ctl to read and understand postgresql.conf to find the
783 data_directory value.
785 </li><li>Add a function like pg_get_indexdef() that report more detailed index
787 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-12/msg00166.php">http://archives.postgresql.org/pgsql-bugs/2007-12/msg00166.php</a>
789 </li><li>Prefix command-line utilities like createuser with 'pg_'
790 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00025.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00025.php</a>
794 <li>Have psql show current values for a sequence
795 </li><li>Move psql backslash database information into the backend, use
796 mnemonic commands? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?psql">psql</a>]
797 <p> This would allow non-psql clients to pull the same information out
798 of the database as psql.
800 </li><li>Fix psql's \d commands more consistent
801 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
802 <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
804 </li><li>Consistently display privilege information for all objects in psql
805 </li><li>Add auto-expanded mode so expanded output is used if the row
806 length is wider than the screen width.
807 <p> Consider using auto-expanded mode for backslash commands like \df+.
809 </li><li>Prevent tab completion of SET TRANSACTION from querying the
810 database and therefore preventing the transaction isolation
811 level from being set.
812 <p> Currently SET <tab> causes a database lookup to check all
813 supported session variables. This query causes problems
814 because setting the transaction isolation level must be the
815 first statement of a transaction.
817 </li><li>Add a \set variable to control whether \s displays line numbers
818 <p> Another option is to add \# which lists line numbers, and
819 allows command execution.
821 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00255.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00255.php</a>
823 </li><li>Prevent escape string warnings when object names have
825 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-01/msg00227.php">http://archives.postgresql.org/pgsql-hackers/2008-01/msg00227.php</a>
827 </li><li>Have \d show foreign keys that reference a table's primary key
828 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00424.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00424.php</a>
830 </li><li>Have \d show child tables that inherit from the specified parent
831 </li><li>Have \l+ show database size, if permissions allow
832 <p> Ideally it will not generate an error for invalid permissions
834 </li><li>Include the symbolic SQLSTATE name in verbose error reports
835 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-09/msg00438.php">http://archives.postgresql.org/pgsql-general/2007-09/msg00438.php</a>
838 </li><li>pg_dump / pg_restore
840 <li>%Add dumping of comments on index columns and composite type columns
841 </li><li>%Add full object name to the tag field. eg. for operators we need
842 '=(integer, integer)', instead of just '='.
843 </li><li>Add pg_dumpall custom format dumps?
844 </li><li>Allow selection of individual object(s) of all types, not just
846 </li><li>In a selective dump, allow dumping of an object and all its
848 </li><li>Add options like pg_restore -l and -L to pg_dump
849 </li><li>Stop dumping CASCADE on DROP TYPE commands in clean mode
850 </li><li>Allow pg_dump --clean to drop roles that own objects or have
852 </li><li>Change pg_dump so that a comment on the dumped database is
853 applied to the loaded database, even if the database has a
854 different name. This will require new backend syntax, perhaps
855 COMMENT ON CURRENT DATABASE.
856 </li><li>Remove unnecessary function pointer abstractions in pg_dump source
858 </li><li>Allow pg_dump to utilize multiple CPUs and I/O channels by dumping
859 multiple objects simultaneously
860 <p> The difficulty with this is getting multiple dump processes to
861 produce a single dump output file. It also would require
862 several sessions to share the same snapshot.
863 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg00205.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg00205.php</a>
865 </li><li>Allow pg_restore to utilize multiple CPUs and I/O channels by
866 restoring multiple objects simultaneously
867 <p> This might require a pg_restore flag to indicate how many
868 simultaneous operations should be performed. Only pg_dump's
869 -Fc format has the necessary dependency information.
870 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg00963.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg00963.php</a>
872 </li><li>To better utilize resources, allow pg_restore to check foreign
873 keys simultaneously, where possible
874 </li><li>Allow pg_restore to create all indexes of a table
875 concurrently, via a single heap scan
876 <p> This requires a pg_dump -Fc file because that format contains
877 the required dependency information.
878 <a href="http://archives.postgresql.org/pgsql-general/2007-05/msg01274.php">http://archives.postgresql.org/pgsql-general/2007-05/msg01274.php</a>
880 </li><li>Allow pg_restore to load different parts of the COPY data
882 </li><li>Prevent pg_dump/pg_restore from being affected by
884 <p> Using psql to restore a pg_dump dump is also affected.
890 <p> Document differences between ecpg and the SQL standard and
891 information about the Informix-compatibility module.
893 </li><li>Solve cardinality > 1 for input descriptors / variables?
894 </li><li>Add a semantic check level, e.g. check if a table really exists
895 </li><li>fix handling of DB attributes that are arrays
896 </li><li>Use backend PREPARE/EXECUTE facility for ecpg where possible
897 </li><li>Implement SQLDA
898 </li><li>Fix nested C comments
899 </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
900 </li><li>Make SET CONNECTION thread-aware, non-standard?
901 </li><li>Allow multidimensional arrays
902 </li><li>Add internationalized message strings
903 </li><li>Implement COPY FROM STDIN
907 <li>Add PQescapeIdentifierConn()
908 </li><li>Prevent PQfnumber() from lowercasing unquoted the column name
909 <p> PQfnumber() should never have been doing lowercasing, but
910 historically it has so we need a way to prevent it
912 </li><li>Allow statement results to be automatically batched to the client
913 <p> Currently all statement results are transferred to the libpq
914 client before libpq makes the results available to the
915 application. This feature would allow the application to make
916 use of the first result rows while the rest are transferred, or
917 held on the server waiting for them to be requested by libpq.
918 One complexity is that a statement like SELECT 1/col could error
919 out mid-way through the result set.
921 </li><li>Consider disallowing multiple queries in PQexec() as an
922 additional barrier to SQL injection attacks
923 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00184.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00184.php</a>
925 </li><li>Add PQexecf() that allows complex parameter substitution
926 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01803.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01803.php</a>
928 </li><li>Add SQLSTATE severity to PGconn return status
929 <p> <a href="http://archives.postgresql.org/pgsql-interfaces/2007-11/msg00015.php">http://archives.postgresql.org/pgsql-interfaces/2007-11/msg00015.php</a>
933 <h1><a name="section_11">Triggers</a></h1>
936 <li>Add deferred trigger queue file
937 <p> Right now all deferred trigger information is stored in backend
938 memory. This could exhaust memory for very large trigger queues.
939 This item involves dumping large queues into files.
941 </li><li>Allow triggers to be disabled in only the current session.
942 <p> This is currently possible by starting a multi-statement transaction,
943 modifying the system tables, performing the desired SQL, restoring the
944 system tables, and committing the transaction. ALTER TABLE ...
945 TRIGGER requires a table lock so it is not ideal for this usage.
947 </li><li>With disabled triggers, allow pg_dump to use ALTER TABLE ADD FOREIGN KEY
948 <p> If the dump is known to be valid, allow foreign keys to be added
949 without revalidating the data.
951 </li><li>Allow statement-level triggers to access modified rows
952 </li><li>Support triggers on columns
953 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php</a>
955 </li><li>Allow AFTER triggers on system tables
956 <p> System tables are modified in many places in the backend without going
957 through the executor and therefore not causing triggers to fire. To
958 complete this item, the functions that modify system tables will have
961 </li><li>Tighten trigger permission checks
962 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00564.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00564.php</a>
964 </li><li>Allow BEFORE INSERT triggers on views
965 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-02/msg01466.php">http://archives.postgresql.org/pgsql-general/2007-02/msg01466.php</a>
967 </li><li>Add ability to trigger on TRUNCATE
968 <p> <a href="http://archives.postgresql.org/pgsql-sql/2008-01/msg00050.php">http://archives.postgresql.org/pgsql-sql/2008-01/msg00050.php</a>
971 <h1><a name="section_12">Indexes</a></h1>
974 <li>Add UNIQUE capability to non-btree indexes
975 </li><li>Prevent index uniqueness checks when UPDATE does not modify the column
976 <p> Uniqueness (index) checks are done when updating a column even if the
977 column is not modified by the UPDATE.
979 </li><li>Allow the creation of on-disk bitmap indexes which can be quickly
980 combined with other bitmap indexes
981 <p> Such indexes could be more compact if there are only a few distinct values.
982 Such indexes can also be compressed. Keeping such indexes updated can be
985 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php</a>
986 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01107.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01107.php</a>
988 </li><li>Allow accurate statistics to be collected on indexes with more than
989 one column or expression indexes, perhaps using per-index statistics
990 <p> <a href="http://archives.postgresql.org/pgsql-performance/2006-10/msg00222.php">http://archives.postgresql.org/pgsql-performance/2006-10/msg00222.php</a>
991 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01131.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01131.php</a>
993 </li><li>Consider increasing the number of default statistics target, and
994 reduce statistics target overhead
995 <p> Also consider having a larger statistics target for indexed columns
996 and expression indexes
997 <a href="http://archives.postgresql.org/pgsql-general/2007-05/msg01228.php">http://archives.postgresql.org/pgsql-general/2007-05/msg01228.php</a>
998 <a href="http://archives.postgresql.org/pgsql-general/2007-06/msg00542.php">http://archives.postgresql.org/pgsql-general/2007-06/msg00542.php</a>
1000 </li><li>Consider compressing indexes by storing key values duplicated in
1001 several rows as a single index entry
1002 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00341.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00341.php</a>
1003 <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg01264.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg01264.php</a>
1004 <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00465.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00465.php</a>
1005 <a href="http://archives.postgresql.org/pgsql-patches/2007-03/msg00163.php">http://archives.postgresql.org/pgsql-patches/2007-03/msg00163.php</a>
1007 </li><li>Add REINDEX CONCURRENTLY, like CREATE INDEX CONCURRENTLY
1008 <p> This is difficult because you must upgrade to an exclusive table lock
1009 to replace the existing index file. CREATE INDEX CONCURRENTLY does not
1010 have this complication. This would allow index compaction without
1012 <a href="http://archives.postgresql.org/pgsql-performance/2007-08/msg00289.php">http://archives.postgresql.org/pgsql-performance/2007-08/msg00289.php</a>
1014 </li><li>Allow multiple indexes to be created concurrently, ideally via a
1015 single heap scan, and have pg_restore use it
1016 </li><li>Inheritance
1018 <li>Allow inherited tables to inherit indexes, UNIQUE constraints,
1019 and primary/foreign keys
1020 </li><li>Honor UNIQUE INDEX on base column in INSERTs/UPDATEs
1021 on inherited table, e.g. INSERT INTO inherit_table
1022 (unique_index_col) VALUES (dup) should fail
1023 <p> The main difficulty with this item is the problem of
1024 creating an index that can span multiple tables.
1026 </li><li>Allow SELECT ... FOR UPDATE on inherited tables
1027 </li><li>Require all CHECK constraints to be inherited
1028 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-04/msg00026.php">http://archives.postgresql.org/pgsql-bugs/2007-04/msg00026.php</a>
1033 <li>Add more GIST index support for geometric data types
1034 </li><li>Allow GIST indexes to create certain complex index types, like
1035 digital trees (see Aoki)
1039 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-09/msg00051.php">http://archives.postgresql.org/pgsql-hackers/2007-09/msg00051.php</a>
1042 <li>Pack hash index buckets onto disk pages more efficiently
1043 <p> Currently only one hash bucket can be stored on a page. Ideally
1044 several hash buckets could be stored on a single page and greater
1045 granularity used for the hash algorithm.
1047 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-06/msg00168.php">http://archives.postgresql.org/pgsql-hackers/2004-06/msg00168.php</a>
1049 </li><li>Consider sorting hash buckets so entries can be found using a
1050 binary search, rather than a linear scan
1051 </li><li>In hash indexes, consider storing the hash value with or instead
1053 </li><li>Add WAL logging for crash recovery
1054 </li><li>Allow multi-column hash indexes
1055 </li><li>-<em>During index creation, pre-sort the tuples to improve build speed</em>
1057 <h1><a name="section_13">Fsync</a></h1>
1060 <li>Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options
1061 <p> Ideally this requires a separate test program that can be run
1062 at initdb time or optionally later. Consider O_SYNC when
1065 </li><li>Add program to test if fsync has a delay compared to non-fsync
1066 </li><li>Consider sorting writes during checkpoint
1067 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg00541.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg00541.php</a>
1070 <h1><a name="section_14">Cache Usage</a></h1>
1073 <li>Speed up COUNT(*)
1074 <p> We could use a fixed row count and a +/- count to follow MVCC
1075 visibility rules, or a single cached value could be used and
1076 invalidated if anyone modifies the table. Another idea is to
1077 get a count directly from a unique index, but for this to be
1078 faster than a sequential scan it must avoid access to the heap
1079 to obtain tuple visibility information.
1081 </li><li>Provide a way to calculate an "estimated COUNT(*)"
1082 <p> Perhaps by using the optimizer's cardinality estimates or random
1085 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php">http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php</a>
1087 </li><li>Allow data to be pulled directly from indexes
1088 <p> Currently indexes do not have enough tuple visibility information
1089 to allow data to be pulled from the index without also accessing
1090 the heap. One way to allow this is to set a bit on index tuples
1091 to indicate if a tuple is currently visible to all transactions
1092 when the first valid heap lookup happens. This bit would have to
1093 be cleared when a heap tuple is expired.
1095 <p> Another idea is to maintain a bitmap of heap pages where all rows
1096 are visible to all backends, and allow index lookups to reference
1097 that bitmap to avoid heap lookups, perhaps the same bitmap we might
1098 add someday to determine which heap pages need vacuuming. Frequently
1099 accessed bitmaps would have to be stored in shared memory. One 8k
1100 page of bitmaps could track 512MB of heap pages.
1102 <p> A third idea would be for a heap scan to check if all rows are visible
1103 and if so set a per-table flag which can be checked by index scans.
1104 Any change to the table would have to clear the flag. To detect
1105 changes during the heap scan a counter could be set at the start and
1106 checked at the end --- if it is the same, the table has not been
1107 modified --- any table change would increment the counter.
1109 </li><li>Consider automatic caching of statements at various levels:
1111 <li>Parsed query tree
1112 </li><li>Query execute plan
1113 </li><li>Query results
1115 </li><li>Consider increasing internal areas when shared buffers is increased
1116 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php">http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php</a>
1118 </li><li>Consider decreasing the amount of memory used by PrivateRefCount
1119 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00797.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00797.php</a>
1120 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00752.php</a>
1122 </li><li>Consider allowing higher priority queries to have referenced buffer
1123 cache pages stay in memory longer
1124 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-11/msg00562.php">http://archives.postgresql.org/pgsql-hackers/2007-11/msg00562.php</a>
1127 <h1><a name="section_15">Vacuum</a></h1>
1130 <li>Improve speed with indexes
1131 <p> For large table adjustments during VACUUM FULL, it is faster to cluster
1132 or reindex rather than update the index. Also, index updates can bloat
1135 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg00024.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg00024.php</a>
1136 <a href="http://archives.postgresql.org/pgsql-performance/2007-05/msg00296.php">http://archives.postgresql.org/pgsql-performance/2007-05/msg00296.php</a>
1137 <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00307.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00307.php</a>
1139 </li><li>Auto-fill the free space map by scanning the buffer cache or by
1140 checking pages written by the background writer
1141 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php">http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php</a>
1142 <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php</a>
1144 </li><li>Create a bitmap of pages that need vacuuming
1145 <p> Instead of sequentially scanning the entire table, have the background
1146 writer or some other process record pages that have expired rows, then
1147 VACUUM can look at just those pages rather than the entire table. In
1148 the event of a system crash, the bitmap would probably be invalidated.
1149 One complexity is that index entries still have to be vacuumed, and
1150 doing this without an index scan (by using the heap values to find the
1151 index entry) might be slow and unreliable, especially for user-defined
1154 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01188.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01188.php</a>
1155 <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00121.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00121.php</a>
1157 </li><li>Allow FSM to return free space toward the beginning of the heap file,
1158 in hopes that empty pages at the end can be truncated by VACUUM
1159 </li><li>Allow FSM page return free space based on table clustering, to assist
1160 in maintaining clustering?
1161 </li><li>Improve dead row detection during multi-statement transactions usage
1162 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-03/msg00358.php">http://archives.postgresql.org/pgsql-patches/2007-03/msg00358.php</a>
1164 </li><li>Consider a more compact data representation for dead tuples
1165 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00143.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00143.php</a>
1167 </li><li>Auto-vacuum
1169 <li>%Issue log message to suggest VACUUM FULL if a table is nearly
1171 </li><li>Improve control of auto-vacuum
1172 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00876.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00876.php</a>
1174 </li><li>Prevent long-lived temporary tables from causing frozen-xid
1175 advancement starvation
1176 <p> The problem is that autovacuum cannot vacuum them to set frozen xids;
1177 only the session that created them can do that.
1178 <a href="http://archives.postgresql.org/pgsql-general/2007-06/msg01645.php">http://archives.postgresql.org/pgsql-general/2007-06/msg01645.php</a>
1180 </li><li>Store per-table autovacuum settings in pg_class.reloptions.
1181 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg01440.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg01440.php</a>
1182 <a href="http://archives.postgresql.org/pgsql-hackers/2008-01/msg00724.php">http://archives.postgresql.org/pgsql-hackers/2008-01/msg00724.php</a>
1184 </li><li>Prevent autovacuum from running if an old transaction is still
1185 running from the last vacuum
1186 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-11/msg00899.php">http://archives.postgresql.org/pgsql-hackers/2007-11/msg00899.php</a>
1190 <h1><a name="section_16">Locking</a></h1>
1193 <li>Fix priority ordering of read and write light-weight locks (Neil)
1194 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00893.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00893.php</a>
1195 <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00905.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00905.php</a>
1197 </li><li>Fix problem when multiple subtransactions of the same outer transaction
1198 hold different types of locks, and one subtransaction aborts
1199 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg01011.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg01011.php</a>
1200 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg00001.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg00001.php</a>
1201 <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00435.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00435.php</a>
1202 <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00773.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00773.php</a>
1204 </li><li>Allow UPDATEs on only non-referential integrity columns not to conflict
1205 with referential integrity locks
1206 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00073.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00073.php</a>
1208 </li><li>Add idle_in_transaction_timeout GUC so locks are not held for long
1210 </li><li>Improve deadlock detection when a page cleaning lock conflicts
1211 with a shared buffer that is pinned
1212 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2008-01/msg00138.php">http://archives.postgresql.org/pgsql-bugs/2008-01/msg00138.php</a>
1213 <a href="http://archives.postgresql.org/pgsql-hackers/2008-01/msg00873.php">http://archives.postgresql.org/pgsql-hackers/2008-01/msg00873.php</a>
1214 <a href="http://archives.postgresql.org/pgsql-committers/2008-01/msg00365.php">http://archives.postgresql.org/pgsql-committers/2008-01/msg00365.php</a>
1217 <h1><a name="section_17">Startup Time Improvements</a></h1>
1220 <li>Experiment with multi-threaded backend for backend creation [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?thread">thread</a>]
1221 <p> This would prevent the overhead associated with process creation. Most
1222 operating systems have trivial process creation time compared to
1223 database startup overhead, but a few operating systems (Win32,
1224 Solaris) might benefit from threading. Also explore the idea of
1225 a single session using multiple threads to execute a statement faster.
1228 <h1><a name="section_18">Write-Ahead Log</a></h1>
1231 <li>Eliminate need to write full pages to WAL before page modification [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?wal">wal</a>]
1232 <p> Currently, to protect against partial disk page writes, we write
1233 full page images to WAL before they are modified so we can correct any
1234 partial page writes during recovery. These pages can also be
1235 eliminated from point-in-time archive files.
1238 <li>When off, write CRC to WAL and check file system blocks
1240 <p> If CRC check fails during recovery, remember the page in case
1241 a later CRC for that page properly matches.
1243 </li><li>Write full pages during file system write and not when
1244 the page is modified in the buffer cache
1245 <p> This allows most full page writes to happen in the background
1246 writer. It might cause problems for applying WAL on recovery
1247 into a partially-written page, but later the full page will be
1251 </li><li>Allow WAL traffic to be streamed to another server for stand-by
1253 </li><li>Reduce WAL traffic so only modified values are written rather than
1255 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-03/msg01589.php">http://archives.postgresql.org/pgsql-hackers/2007-03/msg01589.php</a>
1257 </li><li>Allow WAL information to recover corrupted pg_controldata
1258 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php</a>
1260 </li><li>Find a way to reduce rotational delay when repeatedly writing
1262 <p> Currently fsync of WAL requires the disk platter to perform a full
1263 rotation to fsync again. One idea is to write the WAL to different
1264 offsets that might reduce the rotational delay.
1266 </li><li>Allow WAL logging to be turned off for a table, but the table
1267 might be dropped or truncated during crash recovery [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
1268 <p> Allow tables to bypass WAL writes and just fsync() dirty pages on
1269 commit. This should be implemented using ALTER TABLE, e.g. ALTER
1270 TABLE PERSISTENCE [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo? DROP | TRUNCATE | DEFAULT "> DROP | TRUNCATE | DEFAULT </a>]. Tables using
1271 non-default logging should not use referential integrity with
1272 default-logging tables. A table without dirty buffers during a
1273 crash could perhaps avoid the drop/truncate.
1275 </li><li>Allow WAL logging to be turned off for a table, but the table would
1276 avoid being truncated/dropped [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
1277 <p> To do this, only a single writer can modify the table, and writes
1278 must happen only on new pages so the new pages can be removed during
1279 crash recovery. Readers can continue accessing the table. Such
1280 tables probably cannot have indexes. One complexity is the handling
1281 of indexes on TOAST tables.
1283 </li><li>Speed WAL recovery by allowing more than one page to be prefetched
1284 <p> This involves having a separate process that can be told which pages
1285 the recovery process will need in the near future.
1286 <a href="http://archives.postgresql.org/pgsql-hackers/2008-02/msg01279.php">http://archives.postgresql.org/pgsql-hackers/2008-02/msg01279.php</a>
1289 <h1><a name="section_19">Optimizer / Executor</a></h1>
1292 <li>Improve selectivity functions for geometric operators
1293 </li><li>Precompile SQL functions to avoid overhead
1294 </li><li>Create utility to compute accurate random_page_cost value
1295 </li><li>Improve ability to display optimizer analysis using OPTIMIZER_DEBUG
1296 </li><li>Have EXPLAIN ANALYZE issue NOTICE messages when the estimated and
1297 actual row counts differ by a specified percentage
1298 </li><li>Consider using hash buckets to do DISTINCT, rather than sorting
1299 <p> This would be beneficial when there are few distinct values. This is
1300 already used by GROUP BY.
1302 </li><li>Log statements where the optimizer row estimates were dramatically
1303 different from the number of rows actually found?
1304 </li><li>Consider compressed annealing to search for query plans
1305 <p> This might replace GEQO, <a href="http://sixdemonbag.org/Djinni">http://sixdemonbag.org/Djinni</a>.
1307 </li><li>Improve merge join performance by allowing mark/restore of
1309 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-01/msg00096.php">http://archives.postgresql.org/pgsql-hackers/2007-01/msg00096.php</a>
1311 </li><li>Consider using a hash for joining to a large IN (VALUES ...) list
1312 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-05/msg00450.php">http://archives.postgresql.org/pgsql-hackers/2007-05/msg00450.php</a>
1315 <h1><a name="section_20">Background Writer</a></h1>
1318 <li>Consider having the background writer update the transaction status
1319 hint bits before writing out the page
1320 <p> Implementing this requires the background writer to have access to system
1321 catalogs and the transaction status log.
1323 </li><li>Consider adding buffers the background writer finds reusable to the
1325 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00781.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00781.php</a>
1327 </li><li>Automatically tune bgwriter_delay based on activity rather then using a
1329 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-04/msg00781.php">http://archives.postgresql.org/pgsql-hackers/2007-04/msg00781.php</a>
1331 </li><li>Consider wither increasing BM_MAX_USAGE_COUNT improves performance
1332 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-06/msg01007.php">http://archives.postgresql.org/pgsql-hackers/2007-06/msg01007.php</a>
1334 </li><li>Test to see if calling PreallocXlogFiles() from the background writer
1335 will help with WAL segment creation latency
1336 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-06/msg00340.php">http://archives.postgresql.org/pgsql-patches/2007-06/msg00340.php</a>
1339 <h1><a name="section_21">Miscellaneous Performance</a></h1>
1342 <li>Do async I/O for faster random read-ahead of data
1343 <p> Async I/O allows multiple I/O requests to be sent to the disk with
1344 results coming back asynchronously.
1346 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00820.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00820.php</a>
1347 <a href="http://archives.postgresql.org/pgsql-performance/2007-09/msg00255.php">http://archives.postgresql.org/pgsql-performance/2007-09/msg00255.php</a>
1348 <a href="http://archives.postgresql.org/pgsql-hackers/2007-12/msg00027.php">http://archives.postgresql.org/pgsql-hackers/2007-12/msg00027.php</a>
1350 </li><li>Use mmap() rather than SYSV shared memory or to write WAL files?
1351 <p> This would remove the requirement for SYSV SHM but would introduce
1352 portability issues. Anonymous mmap (or mmap to /dev/zero) is required
1353 to prevent I/O overhead.
1355 </li><li>Consider mmap()'ing files into a backend?
1356 <p> Doing I/O to large tables would consume a lot of address space or
1357 require frequent mapping/unmapping. Extending the file also causes
1358 mapping problems that might require mapping only individual pages,
1359 leading to thousands of mappings. Another problem is that there is no
1360 way to <u>prevent</u> I/O to disk from the dirty shared buffers so changes
1361 could hit disk before WAL is written.
1363 </li><li>Add a script to ask system configuration questions and tune postgresql.conf
1364 </li><li>Consider ways of storing rows more compactly on disk
1366 <li>Reduce the row header size?
1367 </li><li>Consider reducing on-disk varlena length from four bytes to
1368 two because a heap row cannot be more than 64k in length
1370 </li><li>Consider increasing NUM_CLOG_BUFFERS
1371 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00030.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00030.php</a>
1372 <a href="http://archives.postgresql.org/pgsql-performance/2007-08/msg00024.php">http://archives.postgresql.org/pgsql-performance/2007-08/msg00024.php</a>
1374 </li><li>Allow user configuration of TOAST thresholds
1375 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-02/msg00213.php">http://archives.postgresql.org/pgsql-hackers/2007-02/msg00213.php</a>
1376 <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00082.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00082.php</a>
1378 </li><li>Allow configuration of backend priorities via the operating system
1379 <p> Though backend priorities make priority inversion during lock
1380 waits possible, research shows that this is not a huge problem.
1382 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-02/msg00493.php">http://archives.postgresql.org/pgsql-general/2007-02/msg00493.php</a>
1384 </li><li>Experiment with multi-threaded backend better resource utilization
1385 <p> This would allow a single query to make use of multiple CPU's or
1386 multiple I/O channels simultaneously. One idea is to create a
1387 background reader that can pre-fetch sequential and index scan
1388 pages needed by other backends. This could be expanded to allow
1389 concurrent reads from multiple devices in a partitioned table.
1391 </li><li>Consider increasing the minimum allowed number of shared buffers
1392 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2008-02/msg00157.php">http://archives.postgresql.org/pgsql-bugs/2008-02/msg00157.php</a>
1394 </li><li>Expire published xmin for read-only and idle transactions
1395 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-09/msg00343.php">http://archives.postgresql.org/pgsql-hackers/2007-09/msg00343.php</a>
1398 <h1><a name="section_22">Source Code</a></h1>
1401 <li>Add use of 'const' for variables in source tree
1402 </li><li>Move some things from contrib into main tree
1403 </li><li>%Remove warnings created by -Wcast-align
1404 </li><li>Move platform-specific ps status display info from ps_status.c to ports
1405 </li><li>Add optional CRC checksum to heap and index pages
1406 </li><li>Improve documentation to build only interfaces (Marc)
1407 </li><li>Remove or relicense modules that are not under the BSD license, if possible
1408 </li><li>Acquire lock on a relation before building a relcache entry for it
1409 </li><li>Allow cross-compiling by generating the zic database on the target system
1410 </li><li>Improve NLS maintenance of libpgport messages linked onto applications
1411 </li><li>Clean up casting in contrib/isn
1412 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-11/msg00245.php">http://archives.postgresql.org/pgsql-hackers/2006-11/msg00245.php</a>
1414 </li><li>Use UTF8 encoding for NLS messages so all server encodings can
1416 </li><li>Update Bonjour to work with newer cross-platform SDK
1417 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg02238.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg02238.php</a>
1418 <a href="http://archives.postgresql.org/pgsql-patches/2006-10/msg00048.php">http://archives.postgresql.org/pgsql-patches/2006-10/msg00048.php</a>
1420 </li><li>Consider detoasting keys before sorting
1421 </li><li>Consider GnuTLS if OpenSSL license becomes a problem
1422 <p> <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php</a>
1423 <a href="http://archives.postgresql.org/pgsql-hackers/2006-12/msg01213.php">http://archives.postgresql.org/pgsql-hackers/2006-12/msg01213.php</a>
1425 </li><li>Consider changing documentation format from SGML to XML
1426 <p> <a href="http://archives.postgresql.org/pgsql-docs/2006-12/msg00152.php">http://archives.postgresql.org/pgsql-docs/2006-12/msg00152.php</a>
1428 </li><li>Consider making NAMEDATALEN more configurable in future releases
1429 </li><li>Update our code to handle 64-bit timezone files to match the zic
1430 source code, which now uses them
1431 </li><li>Have configure choose integer datetimes by default
1432 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-05/msg00046.php">http://archives.postgresql.org/pgsql-patches/2007-05/msg00046.php</a>
1434 </li><li>Support scoped IPv6 addresses
1435 <p> <a href="http://archives.postgresql.org/pgsql-bugs/2007-05/msg00111.php">http://archives.postgresql.org/pgsql-bugs/2007-05/msg00111.php</a>
1437 </li><li>Consider allowing 64-bit integers and floats to be passed by value on
1439 <p> Also change 32-bit floats (float4) to be passed by value at the same
1442 </li><li>Research use of signals and sleep wake ups
1443 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-07/msg00003.php">http://archives.postgresql.org/pgsql-hackers/2007-07/msg00003.php</a>
1445 </li><li>Add automated check for invalid C++ source code constructs
1446 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-07/msg00056.php">http://archives.postgresql.org/pgsql-patches/2007-07/msg00056.php</a>
1448 </li><li>Consider simplifying how memory context resets handle child contexts
1449 <p> <a href="http://archives.postgresql.org/pgsql-patches/2007-08/msg00067.php">http://archives.postgresql.org/pgsql-patches/2007-08/msg00067.php</a>
1451 </li><li>Remove use of MAKE_PTR and MAKE_OFFSET macros
1452 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-08/msg01510.php">http://archives.postgresql.org/pgsql-general/2007-08/msg01510.php</a>
1456 <li>Remove configure.in check for link failure when cause is found
1457 </li><li>Remove readdir() errno patch when runtime/mingwex/dirent.c rev
1459 </li><li>Remove psql newline patch when we find out why mingw outputs an
1461 </li><li>Allow psql to use readline once non-US code pages work with
1463 </li><li>Fix problem with shared memory on the Win32 Terminal Server
1464 </li><li>Diagnose problem where shared memory can sometimes not be
1465 attached by postmaster children
1466 <p> <a href="http://archives.postgresql.org/pgsql-general/2007-08/msg01377.php">http://archives.postgresql.org/pgsql-general/2007-08/msg01377.php</a>
1468 </li><li>Improve signal handling
1469 <p> <a href="http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php">http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php</a>
1471 </li><li>Convert MSVC build system to remove most batch files
1472 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2007-08/msg00961.php">http://archives.postgresql.org/pgsql-hackers/2007-08/msg00961.php</a>
1475 </li><li>Wire Protocol Changes
1477 <li>Allow dynamic character set handling
1478 </li><li>Add decoded type, length, precision
1479 </li><li>Use compression?
1480 </li><li>Update clients to use data types, typmod, schema.table.column names
1481 of result sets using new statement protocol
1484 <h1><a name="section_23">Exotic Features</a></h1>
1487 <li>Add pre-parsing phase that converts non-ISO syntax to supported
1489 <p> This could allow SQL written for other databases to run without
1492 </li><li>Allow plug-in modules to emulate features from other databases
1493 </li><li>Add features of Oracle-style packages (Pavel)
1494 <p> A package would be a schema with session-local variables,
1495 public/private functions, and initialization functions. It
1496 is also possible to implement these capabilities
1497 in any schema and not use a separate "packages"
1500 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php</a>
1502 </li><li>Consider allowing control of upper/lower case folding of unquoted
1504 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php">http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php</a>
1505 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg01527.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg01527.php</a>
1507 </li><li>Add autonomous transactions
1508 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2008-01/msg00893.php">http://archives.postgresql.org/pgsql-hackers/2008-01/msg00893.php</a>
1511 <h1><a name="section_24">Features We Do <u>Not</u> Want</a></h1>
1514 <li>All backends running as threads in a single process (not wanted)
1515 <p> This eliminates the process protection we get from the current setup.
1516 Thread creation is usually the same overhead as process creation on
1517 modern systems, so it seems unwise to use a pure threaded model.
1519 </li><li>Optimizer hints (not wanted)
1520 <p> Optimizer hints are used to work around problems in the optimizer. We
1521 would rather have the problems reported and fixed.
1523 <p> <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php</a>
1524 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00517.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00517.php</a>
1525 <a href="http://archives.postgresql.org/pgsql-hackers/2006-10/msg00663.php">http://archives.postgresql.org/pgsql-hackers/2006-10/msg00663.php</a>
1527 <p> Because we support postfix operators, it isn't possible to make AS
1528 optional and continue to use bison.
1529 <a href="http://archives.postgresql.org/pgsql-hackers/2003-04/msg00436.php">http://archives.postgresql.org/pgsql-hackers/2003-04/msg00436.php</a>
1531 <p> <a href="http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php">http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php</a>
1533 </li><li>Embedded server (not wanted)
1534 <p> While PostgreSQL clients runs fine in limited-resource environments, the
1535 server requires multiple processes and a stable pool of resources to
1536 run reliabily and efficiently. Stripping down the PostgreSQL server
1537 to run in the same process address space as the client application
1538 would add too much complexity and failure cases.</p>