TODO.html 67.7 KB
Newer Older
1 2 3 4
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
<head>
Bruce Momjian's avatar
Bruce Momjian committed
5
<title>PostgreSQL TODO List</title>
6 7 8
<meta name="generator" content="HTML::TextToHTML v2.25"/>
</head>
<body bgcolor="#FFFFFF" text="#000000" link="#FF0000" vlink="#A00000" alink="#0000FF">
Bruce Momjian's avatar
Bruce Momjian committed
9 10
<h1><a name="section_1">PostgreSQL TODO List</a></h1>
<p>Current maintainer:     Bruce Momjian (<a href="mailto:bruce@momjian.us">bruce@momjian.us</a>)<br/>
Bruce Momjian's avatar
Bruce Momjian committed
11
Last updated:           Wed Sep 20 19:01:42 EDT 2006
Bruce Momjian's avatar
Bruce Momjian committed
12
</p>
Bruce Momjian's avatar
Bruce Momjian committed
13 14
<p>The most recent version of this document can be viewed at<br/>
<a href="http://www.postgresql.org/docs/faqs.TODO.html">http://www.postgresql.org/docs/faqs.TODO.html</a>.
15
</p>
16
<p><strong>A hyphen, "-", marks changes that will appear in the upcoming 8.3 release.</strong><br/>
Bruce Momjian's avatar
Bruce Momjian committed
17
<strong>A percent sign, "%", marks items that are easier to implement.</strong>
18 19 20 21 22 23 24
</p>
<p>Bracketed items, "[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?"></a>]", have more detail.
</p>
<p>This list contains all known PostgreSQL bugs and feature requests. If<br/>
you would like to work on an item, please read the Developer's FAQ<br/>
first.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
25
<h1><a name="section_2">Administration</a></h1>
26 27

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
28
  <li>Allow major upgrades without dump/reload, perhaps using pg_upgrade 
Bruce Momjian's avatar
Bruce Momjian committed
29
  [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pg_upgrade">pg_upgrade</a>]
Bruce Momjian's avatar
Bruce Momjian committed
30
  </li><li>Check for unreferenced table files created by transactions that were
31
  in-progress when the server terminated abruptly
Bruce Momjian's avatar
Bruce Momjian committed
32 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>
</p>
Bruce Momjian's avatar
Bruce Momjian committed
34 35
  </li><li>Allow administrators to safely terminate individual sessions either
  via an SQL function or SIGTERM 
Bruce Momjian's avatar
Bruce Momjian committed
36 37 38 39
<p>  Lock table corruption following SIGTERM of an individual backend
  has been reported in 8.0.  A possible cause was fixed in 8.1, but
  it is unknown whether other problems exist.  This item mostly
  requires additional testing rather than of writing any new code.
40
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00174.php</a>
41
</p>
Bruce Momjian's avatar
Bruce Momjian committed
42
  </li><li>%Set proper permissions on non-system schemas during db creation
43 44 45
<p>  Currently all schemas are owned by the super-user because they are
  copied from the template1 database.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
46 47 48
  </li><li>Support table partitioning that allows a single table to be stored
  in subtables that are partitioned based on the primary key or a WHERE
  clause
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
49
  </li><li>Add function to report the time of the most recent server reload
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
50 51 52
  </li><li>Allow statistics collector information to be pulled from the collector
  process directly, rather than requiring the collector to write a
  filesystem file twice a second?
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
53 54 55
  </li><li>Allow log_min_messages to be specified on a per-module basis
<p>  This would allow administrators to see more detailed information from
  specific sections of the backend, e.g. checkpoints, autovacuum, etc.
56 57
  Another idea is to allow separate configuration files for each module,
  or allow arbitrary SET commands to be passed to them.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
58
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
59 60 61 62 63 64 65
  </li><li>Simplify ability to create partitioned tables
<p>  This would allow creation of partitioned tables without requiring
  creation of rules for INSERT/UPDATE/DELETE, and constraints for
  rapid partition selection.  Options could include range and hash
  partition selection.
</p>
  </li><li>Allow auto-selection of partitioned tables for min/max() operations
66 67 68 69 70 71 72 73 74
  </li><li>Improve replication solutions
  <ul>
    <li>Load balancing
<p>          You can use any of the master/slave replication servers to use a
          standby server for data warehousing. To allow read/write queries to
          multiple servers, you need multi-master replication like pgcluster.
</p>
    </li><li>Allow replication over unreliable or non-persistent links
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
75 76
  </li><li>Configuration files
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
77
    <li>Allow commenting of variables in postgresql.conf to restore them
Bruce Momjian's avatar
Bruce Momjian committed
78
          to defaults
Bruce Momjian's avatar
Bruce Momjian committed
79 80
<p>          Currently, if a variable is commented out, it keeps the
          previous uncommented value until a server restarted.
81
          <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01481.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01481.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
82 83 84 85 86 87 88 89
</p>
    </li><li>Allow pg_hba.conf to specify host names along with IP addresses
<p>          Host name lookup could occur when the postmaster reads the
          pg_hba.conf file, or when the backend starts.  Another
          solution would be to reverse lookup the connection IP and
          check that hostname against the host names in pg_hba.conf.
          We could also then check that the host name maps to the IP
          address.
Bruce Momjian's avatar
Bruce Momjian committed
90
</p>
Bruce Momjian's avatar
Bruce Momjian committed
91
    </li><li>%Allow postgresql.conf file values to be changed via an SQL
Bruce Momjian's avatar
Bruce Momjian committed
92
          API, perhaps using SET GLOBAL
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
93
    </li><li>Allow the server to be stopped/restarted via an SQL API
Bruce Momjian's avatar
Bruce Momjian committed
94
    </li><li>Issue a warning if a change-on-restart-only postgresql.conf value
Bruce Momjian's avatar
Bruce Momjian committed
95
          is modified  and the server config files are reloaded
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
96
    </li><li>Mark change-on-restart-only values in postgresql.conf
Bruce Momjian's avatar
Bruce Momjian committed
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
  </li></ul>
  </li><li>Tablespaces
  <ul>
    <li>Allow a database in tablespace t1 with tables created in
          tablespace t2 to be used as a template for a new database created
          with default tablespace t2
<p>          All objects in the default database tablespace must have default
          tablespace specifications. This is because new databases are
          created by copying directories. If you mix default tablespace
          tables and tablespace-specified tables in the same directory,
          creating a new database from such a mixed directory would create a
          new database with tables that had incorrect explicit tablespaces.
          To fix this would require modifying pg_class in the newly copied
          database, which we don't currently do.
</p>
    </li><li>Allow reporting of which objects are in which tablespaces
<p>          This item is difficult because a tablespace can contain objects
          from multiple databases. There is a server-side function that
          returns the databases which use a specific tablespace, so this
          requires a tool that will call that function and connect to each
          database to find the objects in each database for that tablespace.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
119
    </li><li>%Add a GUC variable to control the tablespace for temporary objects
Bruce Momjian's avatar
Bruce Momjian committed
120 121 122 123
          and sort files
<p>          It could start with a random tablespace from a supplied list and
          cycle through the list.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
124
    </li><li>Allow WAL replay of CREATE TABLESPACE to work when the directory
Bruce Momjian's avatar
Bruce Momjian committed
125
          structure on the recovery computer is different from the original
Bruce Momjian's avatar
Bruce Momjian committed
126
    </li><li>Allow per-tablespace quotas
Bruce Momjian's avatar
Bruce Momjian committed
127
  </li></ul>
128
  </li><li>Point-In-Time Recovery (PITR)
Bruce Momjian's avatar
Bruce Momjian committed
129
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
130
    <li>Add command to archive partially filled write-ahead logs? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pitr">pitr</a>]
Bruce Momjian's avatar
Bruce Momjian committed
131 132
<p>            Currently only full WAL files are archived. This means that the
            most recent transactions aren't available for recovery in case
Bruce Momjian's avatar
Bruce Momjian committed
133
            of a disk failure.
Bruce Momjian's avatar
Bruce Momjian committed
134
</p>
Bruce Momjian's avatar
Bruce Momjian committed
135
    </li><li>%Create dump tool for write-ahead logs for use in determining
Bruce Momjian's avatar
Bruce Momjian committed
136
            transaction id for point-in-time recovery
137
    </li><li>Allow a warm standby system to also allow read-only statements
Bruce Momjian's avatar
Bruce Momjian committed
138 139 140 141 142 143
            [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?pitr">pitr</a>]
<p>            This is useful for checking PITR recovery.
</p>
    </li><li>Allow the PITR process to be debugged and data examined
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
144
<h1><a name="section_3">Monitoring</a></h1>
Bruce Momjian's avatar
Bruce Momjian committed
145 146 147 148 149 150

<ul>
  <li>Allow server log information to be output as INSERT statements
<p>  This would allow server log information to be easily loaded into
  a database for analysis.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
151
  </li><li>%Add ability to monitor the use of temporary sort files
152
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
153
<h1><a name="section_4">Data Types</a></h1>
154 155

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
156 157 158
  <li>Improve the MONEY data type
<p>  Change the MONEY data type to use DECIMAL internally, with special
  locale-aware output formatting.
Bruce Momjian's avatar
Bruce Momjian committed
159
  <a href="http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php">http://archives.postgresql.org/pgsql-general/2005-08/msg01432.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
160
</p>
161
  </li><li>Change NUMERIC to enforce the maximum precision
Bruce Momjian's avatar
Bruce Momjian committed
162 163 164 165 166 167 168 169 170 171 172 173
  </li><li>Add NUMERIC division operator that doesn't round?
<p>  Currently NUMERIC _rounds_ the result to the specified precision.  
  This means division can return a result that multiplied by the 
  divisor is greater than the dividend, e.g. this returns a value &gt; 10:
</p>
</li></ul>
<p>    SELECT (10::numeric(2,0) / 6::numeric(2,0))::numeric(2,0) * 6;
</p>
<p>  The positive modulus result returned by NUMERICs might be considered<br/>
  inaccurate, in one sense.
</p>
<ul>
174
  <li>Fix data types where equality comparison isn't intuitive, e.g. box
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
175 176
  </li><li>Allow user-defined types to specify a type modifier at table creation
  time
Bruce Momjian's avatar
Bruce Momjian committed
177 178 179
  </li><li>Allow user-defined types to accept 'typmod' parameters
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2005-08/msg01142.php">http://archives.postgresql.org/pgsql-hackers/2005-08/msg01142.php</a>
  <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00012.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00012.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
180
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00149.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00149.php</a>
Bruce Momjian's avatar
>  
Bruce Momjian committed
181 182 183
</p>
  </li><li>Add support for public SYNONYMs
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00519.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
184 185 186
</p>
  </li><li>Fix CREATE CAST on DOMAINs
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php">http://archives.postgresql.org/pgsql-hackers/2006-05/msg00072.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
187
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01681.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
188
</p>
Bruce Momjian's avatar
Bruce Momjian committed
189
  </li><li>Add Globally/Universally Unique Identifier (GUID/UUID)
Bruce Momjian's avatar
Bruce Momjian committed
190 191
<p>  <a href="http://archives.postgresql.org/pgsql-patches/2006-09/msg00209.php">http://archives.postgresql.org/pgsql-patches/2006-09/msg00209.php</a>
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
192 193
  </li><li>Add support for SQL-standard GENERATED/IDENTITY columns
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg00543.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
194 195 196
</p>
  </li><li>Support a data type with specific enumerated values (ENUM)
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00979.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00979.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
197 198 199
</p>
  </li><li>Improve XML support
<p>  <a href="http://developer.postgresql.org/index.php/XML_Support">http://developer.postgresql.org/index.php/XML_Support</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
200
</p>
201
  </li><li>Dates and Times
202
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
203
    <li>Allow infinite dates and intervals just like infinite timestamps
204 205 206 207 208
    </li><li>Merge hardwired timezone names with the TZ database; allow either 
          kind everywhere a TZ name is currently taken
    </li><li>Allow TIMESTAMP WITH TIME ZONE to store the original timezone
          information, either zone name or offset from UTC [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?timezone">timezone</a>]
<p>          If the TIMESTAMP value is stored with a time zone name, interval 
Bruce Momjian's avatar
Bruce Momjian committed
209
          computations should adjust based on the time zone rules.
210
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
211
    </li><li>Fix SELECT '0.01 years'::interval, '0.01 months'::interval
Bruce Momjian's avatar
Bruce Momjian committed
212 213 214 215 216 217 218 219 220 221 222 223
    </li><li>Add a GUC variable to allow output of interval values in ISO8601 
          format
    </li><li>Improve timestamptz subtraction to be DST-aware
<p>          Currently, subtracting one date from another that crosses a
          daylight savings time adjustment can return '1 day 1 hour', but
          adding that back to the first date returns a time one hour in
          the future.  This is caused by the adjustment of '25 hours' to
          '1 day 1 hour', and '1 day' is the same time the next day, even
          if daylight savings adjustments are involved.
</p>
    </li><li>Fix interval display to support values exceeding 2^31 hours
    </li><li>Add overflow checking to timestamp and interval arithmetic    
224
    </li><li>Add ISO INTERVAL handling
225 226
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
227 228
<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/>
                  <a href="http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php">http://archives.postgresql.org/pgsql-bugs/2006-04/msg00248.php</a>
229 230
</p>
<ul>
Bruce Momjian's avatar
Bruce Momjian committed
231
  <li>Support ISO INTERVAL syntax if units cannot be determined from
232
                  the string, and are supplied after the string
Bruce Momjian's avatar
Bruce Momjian committed
233 234 235 236 237
<p>                  The SQL standard states that the units after the string
                  specify the units of the string, e.g. INTERVAL '2' MINUTE
                  should return '00:02:00'. The current behavior has the units
                  restrict the interval value to the specified unit or unit
                  range, INTERVAL '70' SECOND returns '00:00:10'.
238 239 240
</p>
<p>                  For syntax that isn't uniquely ISO or PG syntax, like '1' or
                  '1:30', treat as ISO if there is a range specification clause,
Bruce Momjian's avatar
Bruce Momjian committed
241 242 243 244 245 246 247 248 249 250
                  and as PG if there no clause is present, e.g. interpret '1:30'
                  MINUTE TO SECOND as '1 minute 30 seconds', and interpret
                  '1:30' as '1 hour, 30 minutes'.
</p>
<p>                  This makes common cases like SELECT INTERVAL '1' MONTH
                  SQL-standard results. The SQL standard supports a limited
                  number of unit combinations and doesn't support unit names in
                  the string. The PostgreSQL syntax is more flexible in the
                  range of units supported, e.g. PostgreSQL supports '1 year 1
                  hour', while the SQL standard does not.
251
</p>
252 253
  </li><li>Add support for year-month syntax, INTERVAL '50-6' YEAR TO MONTH
  </li><li>Interpret INTERVAL '1 year' MONTH as CAST (INTERVAL '1 year' AS
254
                  INTERVAL MONTH), and this should return '12 months'
255
  </li><li>Round or truncate values to the requested precision, e.g.
256
                  INTERVAL '11 months' AS YEAR should return one or zero
257
  </li><li>Support precision, CREATE TABLE foo (a INTERVAL MONTH(3))
258
  <ul>
259 260
    <li>Arrays
    <ul>
Bruce Momjian's avatar
Bruce Momjian committed
261
      <li>Delay resolution of array expression's data type so assignment
262
          coercion can be performed on empty array expressions
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
263 264
      </li><li>Add support for arrays of domains
      </li><li>Add support for arrays of complex types
265 266 267 268 269 270
    </li></ul>
    </li><li>Binary Data
    <ul>
      <li>Improve vacuum of large objects, like /contrib/vacuumlo?
      </li><li>Add security checking for large objects
      </li><li>Auto-delete large objects when referencing row is deleted
Bruce Momjian's avatar
Bruce Momjian committed
271 272
<p>          /contrib/lo offers this functionality.
</p>
273
      </li><li>Allow read/write into TOAST values like large objects
274
<p>          This requires the TOAST column to be stored EXTERNAL.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
275
</p>
276
      </li><li>Add API for 64-bit large object access
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
277
<p>          <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00781.php</a>
278
</p>
279
    </li></ul>
280 281
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
282
<h1><a name="section_5">Functions</a></h1>
Bruce Momjian's avatar
Bruce Momjian committed
283 284

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
285
  <li>Allow INET subnet tests using non-constants to be indexed
Bruce Momjian's avatar
Bruce Momjian committed
286
  </li><li>%Add pg_get_acldef(), pg_get_typedefault(), pg_get_attrdef(),
Bruce Momjian's avatar
Bruce Momjian committed
287
  pg_get_tabledef(), pg_get_domaindef(), pg_get_functiondef()
288 289
<p>  These would be for application use, not for use by pg_dump.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
290
  </li><li>Allow to_date() and to_timestamp() accept localized month names
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
291 292 293 294 295 296
  </li><li>Add missing parameter handling in to_char()
</li></ul>
<p>        <a href="http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php">http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php</a>
</p>
<ul>
  <li>Allow functions to have a schema search path specified at creation time
Bruce Momjian's avatar
Bruce Momjian committed
297 298 299 300 301 302 303 304 305 306 307 308 309 310
  </li><li>Allow substring/replace() to get/set bit values
  </li><li>Allow to_char() on interval values to accumulate the highest unit
  requested
<p>  Some special format flag would be required to request such
  accumulation.  Such functionality could also be added to EXTRACT. 
  Prevent accumulation that crosses the month/day boundary because of
  the uneven number of days in a month.
</p>
  <ul>
    <li>to_char(INTERVAL '1 hour 5 minutes', 'MI') =&gt; 65
    </li><li>to_char(INTERVAL '43 hours 20 minutes', 'MI' ) =&gt; 2600 
    </li><li>to_char(INTERVAL '43 hours 20 minutes', 'WK:DD:HR:MI') =&gt; 0:1:19:20
    </li><li>to_char(INTERVAL '3 years 5 months','MM') =&gt; 41
  </li></ul>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
311
  </li><li>Add SPI_gettypmod() to return the typemod for a TupleDesc
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
312
  </li><li>Allow inlining of set-returning functions
Bruce Momjian's avatar
Bruce Momjian committed
313
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
314
<h1><a name="section_6">Multi-Language Support</a></h1>
315 316 317 318

<ul>
  <li>Add NCHAR (as distinguished from ordinary varchar),
  </li><li>Allow locale to be set at database creation
Bruce Momjian's avatar
Bruce Momjian committed
319 320 321 322
<p>  Currently locale can only be set during initdb.  No global tables have
  locale-aware columns.  However, the database template used during
  database creation might have locale-aware indexes.  The indexes would
  need to be reindexed to match the new locale.
323
</p>
Bruce Momjian's avatar
Bruce Momjian committed
324
  </li><li>Allow encoding on a per-column basis optionally using the ICU library:
325
<p>  Right now only one encoding is allowed per database.  [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
326
  <a href="http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php">http://archives.postgresql.org/pgsql-hackers/2005-03/msg00932.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
327 328
  <a href="http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php">http://archives.postgresql.org/pgsql-patches/2005-08/msg00309.php</a>
  <a href="http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php">http://archives.postgresql.org/pgsql-patches/2006-03/msg00233.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
329
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg00662.php</a>
330
</p>
331
  </li><li>Add CREATE COLLATE?  [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?locale">locale</a>]
332
  </li><li>Support multiple simultaneous character sets, per SQL92
Bruce Momjian's avatar
Bruce Momjian committed
333
  </li><li>Improve UTF8 combined character handling?
334 335
  </li><li>Add octet_length_server() and octet_length_client()
  </li><li>Make octet_length_client() the same as octet_length()?
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
336
  </li><li>Fix problems with wrong runtime encoding conversion for NLS message files
Bruce Momjian's avatar
Bruce Momjian committed
337 338
  </li><li>Add URL to more complete multi-byte regression tests
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php">http://archives.postgresql.org/pgsql-hackers/2005-07/msg00272.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
339 340 341 342
</p>
  </li><li>Fix ILIKE and regular expressions to handle case insensitivity
  properly in multibyte encodings
<p>  <a href="http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php">http://archives.postgresql.org/pgsql-bugs/2005-10/msg00001.php</a>
343
  <a href="http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php">http://archives.postgresql.org/pgsql-patches/2005-11/msg00173.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
344 345 346 347
</p>
  </li><li>Set client encoding based on the client operating system encoding
<p>  Currently client_encoding is set in postgresql.conf, which
  defaults to the server encoding.
Bruce Momjian's avatar
Bruce Momjian committed
348
</p>
349
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
350
<h1><a name="section_7">Views / Rules</a></h1>
351 352

<ul>
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
353
  <li>Automatically create rules on views so they are updateable, per SQL99
354
<p>  We can only auto-create rules for simple views.  For more complex
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
355
  cases users will still have to write rules manually.
Bruce Momjian's avatar
Bruce Momjian committed
356
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00586.php</a>
357 358 359
</p>
  </li><li>Add the functionality for WITH CHECK OPTION clause of CREATE VIEW
  </li><li>Allow NOTIFY in rules involving conditionals
Bruce Momjian's avatar
Bruce Momjian committed
360
  </li><li>Allow VIEW/RULE recompilation when the underlying tables change
Bruce Momjian's avatar
Bruce Momjian committed
361 362 363 364
<p>  Another issue is whether underlying table changes should be reflected
  in the view, e.g. should SELECT * show additional columns if they
  are added after the view is created.
</p>
365
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
366
<h1><a name="section_8">SQL Commands</a></h1>
367 368

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
369
  <li>Add CORRESPONDING BY to UNION/INTERSECT/EXCEPT
370
  </li><li>Add ROLLUP, CUBE, GROUPING SETS options to GROUP BY
Bruce Momjian's avatar
Bruce Momjian committed
371 372
  </li><li>%Allow SET CONSTRAINTS to be qualified by schema/table name
  </li><li>%Add a separate TRUNCATE permission
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
373 374 375
<p>  Currently only the owner can TRUNCATE a table because triggers are not
  called, and the table is locked in exclusive mode.
</p>
376 377 378 379 380 381 382 383
  </li><li>Allow PREPARE of cursors
  </li><li>Allow finer control over the caching of prepared query plans
<p>  Currently, queries prepared via the libpq API are planned on first
  execute using the supplied parameters --- allow SQL PREPARE to do the
  same.  Also, allow control over replanning prepared queries either
  manually or automatically when statistics for execute parameters
  differ dramatically from those used during planning.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
384 385
  </li><li>Invalidate prepared queries, like INSERT, when the table definition
  is altered
386 387 388 389 390 391 392 393
  </li><li>Allow LISTEN/NOTIFY to store info in memory rather than tables?
<p>  Currently LISTEN/NOTIFY information is stored in pg_listener. Storing
  such information in memory would improve performance.
</p>
  </li><li>Add optional textual message to NOTIFY
<p>  This would allow an informational message to be added to the notify
  message, perhaps indicating the row modified or other custom
  information.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
394
</p>
395
  </li><li>Add a GUC variable to warn about non-standard SQL usage in queries
396
  </li><li>Add SQL-standard MERGE command, typically used to merge two tables
Bruce Momjian's avatar
Bruce Momjian committed
397
  [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
398 399 400 401 402
<p>  This is similar to UPDATE, then for unmatched rows, INSERT.
  Whether concurrent access allows modifications which could cause
  row loss is implementation independent.
</p>
  </li><li>Add REPLACE or UPSERT command that does UPDATE, or on failure, INSERT
Bruce Momjian's avatar
Bruce Momjian committed
403
  [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?merge">merge</a>]
Bruce Momjian's avatar
Bruce Momjian committed
404
<p>  To implement this cleanly requires that the table have a unique index
405 406 407
  so duplicate checking can be easily performed.  It is possible to
  do it without a unique index if we require the user to LOCK the table
  before the MERGE.
Bruce Momjian's avatar
Bruce Momjian committed
408
</p>
409 410
  </li><li>Add NOVICE output level for helpful messages like automatic sequence/index
  creation
411
  </li><li>Add RESET CONNECTION command to reset all session state
412
<p>  This would include resetting of all variables (RESET ALL), dropping of
413 414 415 416
  temporary tables, removing any NOTIFYs, cursors, open transactions,
  prepared queries, currval()s, etc.  This could be used  for connection
  pooling.  We could also change RESET ALL to have this functionality.  
  The difficult of this features is allowing RESET ALL to not affect 
Bruce Momjian's avatar
Bruce Momjian committed
417 418
  changes made by the interface driver for its internal use.  One idea 
  is for this to be a protocol-only feature.  Another approach is to 
419
  notify the protocol when a RESET CONNECTION command is used.
Bruce Momjian's avatar
Bruce Momjian committed
420
  <a href="http://archives.postgresql.org/pgsql-patches/2006-04/msg00192.php">http://archives.postgresql.org/pgsql-patches/2006-04/msg00192.php</a>
421
</p>
422
  </li><li>Add GUC to issue notice about statements that use unjoined tables
423
  </li><li>Allow EXPLAIN to identify tables that were skipped because of 
424
  constraint_exclusion
425
  </li><li>Allow EXPLAIN output to be more easily processed by scripts
Bruce Momjian's avatar
Bruce Momjian committed
426
  </li><li>Enable standard_conforming_strings
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
427 428 429 430 431 432 433
  </li><li>Make standard_conforming_strings the default in 8.3?
<p>  When this is done, backslash-quote should be prohibited in non-E''
  strings because of possible confusion over how such strings treat
  backslashes.  Basically, '' is always safe for a literal single
  quote, while \' might or might not be based on the backslash
  handling rules.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
434
  </li><li>Simplify dropping roles that have objects in several databases
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
435
  </li><li>Allow COMMENT ON to accept an expression rather than just a string
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
436 437
  </li><li>Allow the count returned by SELECT, etc to be to represent as an int64
  to allow a higher range of values
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
438
  </li><li>Make CLUSTER preserve recently-dead tuples per MVCC requirements
Bruce Momjian's avatar
Bruce Momjian committed
439
  </li><li>Add SQL99 WITH clause to SELECT
Bruce Momjian's avatar
Bruce Momjian committed
440
  </li><li>Add SQL:2003 WITH RECURSIVE (hierarchical) queries to SELECT
Bruce Momjian's avatar
Bruce Momjian committed
441 442 443 444
  </li><li>Add DEFAULT .. AS OWNER so permission checks are done as the table
  owner
<p>  This would be useful for SERIAL nextval() calls and CHECK constraints.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
445 446
  </li><li>Add a GUC to control whether BEGIN inside a transcation should abort
  the transaction.
Bruce Momjian's avatar
Bruce Momjian committed
447
  </li><li>Allow DISTINCT to work in multiple-argument aggregate calls
Bruce Momjian's avatar
Bruce Momjian committed
448 449
  </li><li>Add column to pg_stat_activity that shows the progress of long-running
  commands like CREATE INDEX and VACUUM
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
450
  </li><li>Implement SQL:2003 window functions
451 452 453 454 455 456 457 458 459
  </li><li>CREATE
  <ul>
    <li>Allow CREATE TABLE AS to determine column lengths for complex
          expressions like SELECT col1 || col2
    </li><li>Use more reliable method for CREATE DATABASE to get a consistent
          copy of db?
  </li></ul>
  </li><li>UPDATE
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
460
    <li>Allow UPDATE tab SET ROW (col, ...) = (SELECT...)
Bruce Momjian's avatar
Bruce Momjian committed
461
<p>          <a href="http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php">http://archives.postgresql.org/pgsql-hackers/2006-07/msg01306.php</a>
462
</p>
463
  </li></ul>
464 465
  </li><li>ALTER
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
466
    <li>%Have ALTER TABLE RENAME rename SERIAL sequence names
Bruce Momjian's avatar
Bruce Momjian committed
467
    </li><li>Add ALTER DOMAIN to modify the underlying data type
Bruce Momjian's avatar
Bruce Momjian committed
468
    </li><li>%Allow ALTER TABLE ... ALTER CONSTRAINT ... RENAME
469
<p>          <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00168.php</a>
470
</p>
Bruce Momjian's avatar
Bruce Momjian committed
471
    </li><li>%Allow ALTER TABLE to change constraint deferrability and actions
Bruce Momjian's avatar
Bruce Momjian committed
472
    </li><li>Add missing object types for ALTER ... SET SCHEMA
473
    </li><li>Allow ALTER TABLESPACE to move to different directories
474
    </li><li>Allow databases to be moved to different tablespaces
475 476
    </li><li>Allow moving system tables to other tablespaces, where possible
<p>          Currently non-global system tables must be in the default database
477
          tablespace. Global system tables can never be moved.
478
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
479
    </li><li>Prevent parent tables from altering or dropping constraints
480 481 482 483
          like CHECK that are inherited by child tables unless CASCADE
          is used
    </li><li>%Prevent child tables from altering or dropping constraints
          like CHECK that were inherited from the parent table
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
484
    </li><li>Have ALTER INDEX update the name of a constraint using that index
Bruce Momjian's avatar
Bruce Momjian committed
485
    </li><li>Add ALTER TABLE RENAME CONSTRAINT, update index name also
486 487 488 489 490 491
  </li></ul>
  </li><li>CLUSTER
  <ul>
    <li>Automatically maintain clustering on a table
<p>          This might require some background daemon to maintain clustering
          during periods of low usage. It might also require tables to be only
Bruce Momjian's avatar
Bruce Momjian committed
492
          partially filled for easier reorganization.  Another idea would
493 494 495 496
          be to create a merged heap/index data file so an index lookup would
          automatically access the heap data too.  A third idea would be to
          store heap rows in hashed groups, perhaps using a user-supplied
          hash function.
Bruce Momjian's avatar
Bruce Momjian committed
497
          <a href="http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php">http://archives.postgresql.org/pgsql-performance/2004-08/msg00349.php</a>
498
</p>
Bruce Momjian's avatar
Bruce Momjian committed
499
    </li><li>%Add default clustering to system tables
500 501 502 503 504 505 506 507 508 509
<p>          To do this, determine the ideal cluster index for each system
          table and set the cluster setting during initdb.
</p>
  </li></ul>
  </li><li>COPY
  <ul>
    <li>Allow COPY to report error lines and continue
<p>          This requires the use of a savepoint before each COPY line is
          processed, with ROLLBACK on COPY failure.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
510 511
    </li><li>Allow COPY on a newly-created table to skip WAL logging
<p>          On crash recovery, the table involved in the COPY would
Bruce Momjian's avatar
Bruce Momjian committed
512
          be removed or have its heap and index files truncated.  One
Bruce Momjian's avatar
Bruce Momjian committed
513 514 515
          issue is that no other backend should be able to add to 
          the table at the same time, which is something that is 
          currently allowed.
Bruce Momjian's avatar
Bruce Momjian committed
516
</p>
517
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
518 519 520
  </li><li>GRANT/REVOKE
  <ul>
    <li>Allow column-level privileges
Bruce Momjian's avatar
Bruce Momjian committed
521
    </li><li>%Allow GRANT/REVOKE permissions to be applied to all schema objects
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
522
          with one command
Bruce Momjian's avatar
Bruce Momjian committed
523 524 525 526
<p>          The proposed syntax is:
</p><p>                GRANT SELECT ON ALL TABLES IN public TO phpuser;
                GRANT SELECT ON NEW TABLES IN public TO phpuser;
</p>
Bruce Momjian's avatar
Bruce Momjian committed
527
    </li><li>Allow GRANT/REVOKE permissions to be inherited by objects based on
Bruce Momjian's avatar
Bruce Momjian committed
528
          schema permissions
Bruce Momjian's avatar
Bruce Momjian committed
529
    </li><li>Allow SERIAL sequences to inherit permissions from the base table?
Bruce Momjian's avatar
Bruce Momjian committed
530
  </li></ul>
531 532 533 534 535 536 537 538 539
  </li><li>CURSOR
  <ul>
    <li>Allow UPDATE/DELETE WHERE CURRENT OF cursor
<p>          This requires using the row ctid to map cursor rows back to the
          original heap row. This become more complicated if WITH HOLD cursors
          are to be supported because WITH HOLD cursors have a copy of the row
          and no FOR UPDATE lock.
</p>
    </li><li>Prevent DROP TABLE from dropping a row referenced by its own open
Bruce Momjian's avatar
Bruce Momjian committed
540
          cursor?
541 542 543 544
  </li></ul>
  </li><li>INSERT
  <ul>
    <li>Allow INSERT/UPDATE of the system-generated oid value for a row
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
545
    </li><li>In rules, allow VALUES() to contain a mixture of 'old' and 'new'
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
546
          references
547
  </li></ul>
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
548
  </li><li>SHOW/SET
549
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
550
    <li>Add SET PERFORMANCE_TIPS option to suggest INDEX, VACUUM, VACUUM
551
          ANALYZE, and CLUSTER
Bruce Momjian's avatar
Bruce Momjian committed
552
    </li><li>Add SET PATH for schemas?
553 554 555
<p>          This is basically the same as SET search_path.
</p>
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
556 557 558 559 560 561
  </li><li>Referential Integrity
  <ul>
    <li>Add MATCH PARTIAL referential integrity
    </li><li>Change foreign key constraint for array -&gt; element to mean element
          in array?
    </li><li>Enforce referential integrity for system tables
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
562 563 564 565
    </li><li>Fix problem when cascading referential triggers make changes on
          cascaded tables, seeing the tables in an intermediate state
<p>          <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
          <a href="http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php">http://archives.postgresql.org/pgsql-hackers/2005-09/msg00174.php</a>
566 567 568 569 570 571
</p>
    </li><li>Allow DEFERRABLE and end-of-statement UNIQUE constraints?
<p>          This would allow UPDATE tab SET col = col + 1 to work if col has
          a unique index.  Currently, uniqueness checks are done while the
          command is being executed, rather than at the end of the statement
          or transaction.
Bruce Momjian's avatar
Bruce Momjian committed
572
          <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>
573
          <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01458.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
574
</p>
Bruce Momjian's avatar
Bruce Momjian committed
575
  </li></ul>
576
  </li><li>Server-Side Languages
577
  <ul>
578 579 580 581
    <li>PL/pgSQL
    <ul>
      <li>Fix RENAME to work on variables other than OLD/NEW
      </li><li>Allow function parameters to be passed by name,
Bruce Momjian's avatar
Fix:  
Bruce Momjian committed
582
                  get_employee_salary(12345 AS emp_id, 2001 AS tax_year)
Bruce Momjian's avatar
Bruce Momjian committed
583
      </li><li>Add Oracle-style packages  (Pavel)
Bruce Momjian's avatar
Bruce Momjian committed
584
<p>                  A package would be a schema with session-local variables,
Bruce Momjian's avatar
Bruce Momjian committed
585
                  public/private functions, and initialization functions.  It
Bruce Momjian's avatar
Bruce Momjian committed
586 587 588
                  is also possible to implement these capabilities
                  in all schemas and not use a separate "packages"
                  syntax at all.
Bruce Momjian's avatar
Bruce Momjian committed
589
                  <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00384.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
590
</p>
591 592 593 594 595 596 597 598 599 600
      </li><li>Allow handling of %TYPE arrays, e.g. tab.col%TYPE[<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?"></a>]
      </li><li>Allow listing of record column names, and access to
                  record columns via variables, e.g. columns := r.(*),
                  tval2 := r.(colname)
<p>                  <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00458.php</a>
                  <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00302.php</a>
                  <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00031.php</a>
</p>
      </li><li>Add MOVE
      </li><li>Add single-step debugging of functions
Bruce Momjian's avatar
Bruce Momjian committed
601 602 603 604
      </li><li>Add support for WITH HOLD and SCROLL cursors
<p>                  PL/pgSQL cursors should support the same syntax as
                  backend cursors.
</p>
605 606
      </li><li>Allow PL/RETURN to return row or record functions
<p>                  <a href="http://archives.postgresql.org/pgsql-patches/2005-11/msg00045.php">http://archives.postgresql.org/pgsql-patches/2005-11/msg00045.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
607 608 609
</p>
      </li><li>Fix memory leak from exceptions
<p>                  <a href="http://archives.postgresql.org/pgsql-performance/2006-06/msg00305.php">http://archives.postgresql.org/pgsql-performance/2006-06/msg00305.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
610 611 612 613
</p>
      </li><li>Fix problems with RETURN NEXT on tables with
                  dropped/added columns after function creation
<p>                  <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00165.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00165.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
614
</p>
615 616 617 618 619 620 621 622 623
    </li></ul>
    </li><li>Other
    <ul>
      <li>Add table function support to pltcl, plpython
      </li><li>Add support for polymorphic arguments and return types to
                  languages other than PL/PgSQL
      </li><li>Add capability to create and call PROCEDURES
      </li><li>Add support for OUT and INOUT parameters to languages other 
                  than PL/PgSQL
Bruce Momjian's avatar
Bruce Momjian committed
624 625
      </li><li>Add PL/Python tracebacks
<p>                  <a href="http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php">http://archives.postgresql.org/pgsql-patches/2006-02/msg00288.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
626
</p>
627
    </li></ul>
628 629
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
630
<h1><a name="section_9">Clients</a></h1>
631 632

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
633
  <li>Have pg_ctl look at PGHOST in case it is a socket directory?
634 635 636 637 638 639
  </li><li>Allow pg_ctl to work properly with configuration files located outside
  the PGDATA directory
<p>  pg_ctl can not read the pid file because it isn't located in the
  config directory but in the PGDATA directory.  The solution is to
  allow pg_ctl to read and understand postgresql.conf to find the
  data_directory value.
640
</p>
641 642 643 644 645 646 647 648
  </li><li>psql
  <ul>
    <li>Have psql show current values for a sequence
    </li><li>Move psql backslash database information into the backend, use
          mnemonic commands? [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?psql">psql</a>]
<p>          This would allow non-psql clients to pull the same information out
          of the database as psql.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
649
    </li><li>Fix psql's \d commands more consistent
Bruce Momjian's avatar
Bruce Momjian committed
650
<p>          <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
651
          <a href="http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php">http://archives.postgresql.org/pgsql-hackers/2004-11/msg00014.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
652
</p>
653 654 655
    </li><li>Allow psql \pset boolean variables to set to fixed values, rather
          than toggle
    </li><li>Consistently display privilege information for all objects in psql
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
656 657 658
    </li><li>Add auto-expanded mode so expanded output is used if the row
          length is wider than the screen width.  
<p>          Consider using auto-expanded mode for backslash commands like \df+.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
659 660 661 662 663 664 665 666
</p>
    </li><li>Prevent tab completion of SET TRANSACTION from querying the
          database and therefore preventing the transaction isolation
          level from being set.
<p>          Currently, SET &lt;tab&gt; causes a database lookup to check all
          supported session variables.  This query causes problems
          because setting the transaction isolation level must be the
          first statement of a transaction.
Bruce Momjian's avatar
Bruce Momjian committed
667
</p>
668
  </li></ul>
669 670
  </li><li>pg_dump
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
671
    <li>%Add dumping of comments on index columns and composite type columns
Bruce Momjian's avatar
Bruce Momjian committed
672
    </li><li>%Add full object name to the tag field.  eg. for operators we need
673
          '=(integer, integer)', instead of just '='.
Bruce Momjian's avatar
Bruce Momjian committed
674 675 676
    </li><li>Add pg_dumpall custom format dumps?
    </li><li>Remove unnecessary function pointer abstractions in pg_dump source
          code
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
677 678 679 680
    </li><li>Allow selection of individual object(s) of all types, not just
          tables
    </li><li>In a selective dump, allow dumping of an object and all its 
          dependencies
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
681
    </li><li>Add options like pg_restore -l and -L to pg_dump
Bruce Momjian's avatar
Bruce Momjian committed
682
    </li><li>Stop dumping CASCADE on DROP TYPE commands in clean mode
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
683 684
    </li><li>Allow pg_dump --clean to drop roles that own objects or have
          privileges
Bruce Momjian's avatar
Bruce Momjian committed
685
    </li><li>Add -f to pg_dumpall
686
  </li></ul>
687
  </li><li>ecpg
688 689 690 691 692
  <ul>
    <li>Docs
<p>          Document differences between ecpg and the SQL standard and
          information about the Informix-compatibility module.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
693
    </li><li>Solve cardinality &gt; 1 for input descriptors / variables?
694 695 696 697 698
    </li><li>Add a semantic check level, e.g. check if a table really exists
    </li><li>fix handling of DB attributes that are arrays
    </li><li>Use backend PREPARE/EXECUTE facility for ecpg where possible
    </li><li>Implement SQLDA
    </li><li>Fix nested C comments
Bruce Momjian's avatar
Bruce Momjian committed
699
    </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
700 701
    </li><li>Make SET CONNECTION thread-aware, non-standard?
    </li><li>Allow multidimensional arrays
Bruce Momjian's avatar
Bruce Momjian committed
702
    </li><li>Add internationalized message strings
Bruce Momjian's avatar
Bruce Momjian committed
703
    </li><li>Implement COPY FROM STDIN
704
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
705 706
  </li><li>libpq
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
707
    <li>Add PQescapeIdentifierConn()
Bruce Momjian's avatar
Bruce Momjian committed
708
    </li><li>Prevent PQfnumber() from lowercasing unquoted the column name
709 710 711
<p>          PQfnumber() should never have been doing lowercasing, but 
          historically it has so we need a way to prevent it
</p>
Bruce Momjian's avatar
Bruce Momjian committed
712
    </li><li>Allow statement results to be automatically batched to the client
Bruce Momjian's avatar
Bruce Momjian committed
713
<p>          Currently, all statement results are transferred to the libpq
714 715
          client before libpq makes the results available to the 
          application.  This feature would allow the application to make
Bruce Momjian's avatar
Bruce Momjian committed
716
          use of the first result rows while the rest are transferred, or
717
          held on the server waiting for them to be requested by libpq.
718
          One complexity is that a statement like SELECT 1/col could error
719 720
          out mid-way through the result set.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
721
  </li></ul>
722
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
723
<h1><a name="section_10">Triggers</a></h1>
724
<ul>
Bruce Momjian's avatar
Bruce Momjian committed
725
  <li>Add deferred trigger queue file
726 727 728
<p>  Right now all deferred trigger information is stored in backend
  memory.  This could exhaust memory for very large trigger queues.
  This item involves dumping large queues into files.
Bruce Momjian's avatar
Bruce Momjian committed
729
</p>
730 731 732 733
  </li><li>Allow triggers to be disabled in only the current session.
<p>  This is currently possible by starting a multi-statement transaction,
  modifying the system tables, performing the desired SQL, restoring the
  system tables, and committing the transaction.  ALTER TABLE ...
Bruce Momjian's avatar
Bruce Momjian committed
734
  TRIGGER requires a table lock so it is not ideal for this usage.
735 736 737 738 739 740
</p>
  </li><li>With disabled triggers, allow pg_dump to use ALTER TABLE ADD FOREIGN KEY
<p>  If the dump is known to be valid, allow foreign keys to be added
  without revalidating the data.
</p>
  </li><li>Allow statement-level triggers to access modified rows
Bruce Momjian's avatar
Bruce Momjian committed
741
  </li><li>Support triggers on columns
742
<p>  <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00107.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
743
</p>
744 745 746 747 748 749 750
  </li><li>Allow AFTER triggers on system tables
<p>  System tables are modified in many places in the backend without going
  through the executor and therefore not causing triggers to fire. To
  complete this item, the functions that modify system tables will have
  to fire triggers.
</p>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
751
<h1><a name="section_11">Dependency Checking</a></h1>
752 753

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
754 755 756
  <li>Flush cached query plans when the dependent objects change,
  when the cardinality of parameters changes dramatically, or
  when new ANALYZE statistics are available
Bruce Momjian's avatar
Bruce Momjian committed
757 758 759
<p>  A more complex solution would be to save multiple plans for different
  cardinality and use the appropriate plan based on the EXECUTE values.
</p>
760
  </li><li>Track dependencies in function bodies and recompile/invalidate
Bruce Momjian's avatar
Bruce Momjian committed
761 762
<p>  This is particularly important for references to temporary tables
  in PL/PgSQL because PL/PgSQL caches query plans.  The only workaround
763 764 765
  in PL/PgSQL is to use EXECUTE.  One complexity is that a function
  might itself drop and recreate dependent tables, causing it to
  invalidate its own query plan.
Bruce Momjian's avatar
Bruce Momjian committed
766
</p>
767
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
768
<h1><a name="section_12">Exotic Features</a></h1>
769 770

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
771
  <li>Add pre-parsing phase that converts non-ISO syntax to supported
772 773 774 775 776 777 778
  syntax
<p>  This could allow SQL written for other databases to run without
  modification.
</p>
  </li><li>Allow plug-in modules to emulate features from other databases
  </li><li>SQL*Net listener that makes PostgreSQL appear as an Oracle database
  to clients
779
  </li><li>Allow statements across databases or servers with transaction
780
  semantics
781
<p>  This can be done using dblink and two-phase commit.
782
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
783 784
  </li><li>Add the features of packages
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
785
    <li>Make private objects accessible only to objects in the same schema
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
786 787 788 789
    </li><li>Allow current_schema.objname to access current schema objects
    </li><li>Add session variables
    </li><li>Allow nested schemas
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
790 791
  </li><li>Consider allowing control of upper/lower case folding of unquoted
  identifiers
792
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php">http://archives.postgresql.org/pgsql-hackers/2004-04/msg00818.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
793
</p>
794
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
795
<h1><a name="section_13">Indexes</a></h1>
796

Bruce Momjian's avatar
Bruce Momjian committed
797 798 799 800 801 802 803 804 805
<ul>
  <li>Allow inherited tables to inherit index, UNIQUE constraint, and primary
  key, foreign key
  </li><li>UNIQUE INDEX on base column not honored on INSERTs/UPDATEs from
  inherited table:  INSERT INTO inherit_table (unique_index_col) VALUES
  (dup) should fail
<p>  The main difficulty with this item is the problem of creating an index
  that can span more than one table.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
806
  </li><li>Allow SELECT ... FOR UPDATE on inherited tables
Bruce Momjian's avatar
Bruce Momjian committed
807 808 809 810 811 812 813 814 815 816
  </li><li>Add UNIQUE capability to non-btree indexes
  </li><li>Prevent index uniqueness checks when UPDATE does not modify the column
<p>  Uniqueness (index) checks are done when updating a column even if the
  column is not modified by the UPDATE.
</p>
  </li><li>Allow the creation of on-disk bitmap indexes which can be quickly
  combined with other bitmap indexes
<p>  Such indexes could be more compact if there are only a few distinct values.
  Such indexes can also be compressed.  Keeping such indexes updated can be
  costly.
Bruce Momjian's avatar
Bruce Momjian committed
817
  <a href="http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php">http://archives.postgresql.org/pgsql-patches/2005-07/msg00512.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
818 819 820 821 822 823
</p>
  </li><li>Allow use of indexes to search for NULLs
<p>  One solution is to create a partial index on an IS NULL expression.
</p>
  </li><li>Allow accurate statistics to be collected on indexes with more than
  one column or expression indexes, perhaps using per-index statistics
Bruce Momjian's avatar
Bruce Momjian committed
824 825 826 827 828 829 830
  </li><li>Allow the creation of indexes with mixed ascending/descending
  specifiers
<p>  This is possible now by creating an operator class with reversed sort
  operators.  One complexity is that NULLs would then appear at the start
  of the result set, and this might affect certain sort types, like
  merge join.
</p>
831
  </li><li>Allow constraint_exclusion to work for UNIONs like it does for
832 833
  inheritance, allow it to work for UPDATE and DELETE statements, and allow
  it to be used for all statements with little performance impact
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
834 835
  </li><li>Allow CREATE INDEX to take an additional parameter for use with
  special index types
Bruce Momjian's avatar
Bruce Momjian committed
836
  </li><li>Consider compressing indexes by storing key values duplicated in
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
837
  several rows as a single index entry
Bruce Momjian's avatar
Bruce Momjian committed
838 839
<p>  This is difficult because it requires datatype-specific knowledge.
</p>
840 841 842 843 844 845
  </li><li>GIST
  <ul>
    <li>Add more GIST index support for geometric data types
    </li><li>Allow GIST indexes to create certain complex index types, like
          digital trees (see Aoki)
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
846 847 848
  </li><li>Hash
  <ul>
    <li>Pack hash index buckets onto disk pages more efficiently
Bruce Momjian's avatar
Bruce Momjian committed
849
<p>          Currently only one hash bucket can be stored on a page. Ideally
Bruce Momjian's avatar
Bruce Momjian committed
850 851 852 853 854 855 856
          several hash buckets could be stored on a single page and greater
          granularity used for the hash algorithm.
</p>
    </li><li>Consider sorting hash buckets so entries can be found using a
          binary search, rather than a linear scan
    </li><li>In hash indexes, consider storing the hash value with or instead
          of the key itself
Bruce Momjian's avatar
Bruce Momjian committed
857 858
    </li><li>Add WAL logging for crash recovery
    </li><li>Allow multi-column hash indexes
Bruce Momjian's avatar
Bruce Momjian committed
859 860
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
861
<h1><a name="section_14">Fsync</a></h1>
862 863 864 865

<ul>
  <li>Improve commit_delay handling to reduce fsync()
  </li><li>Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options
Bruce Momjian's avatar
Bruce Momjian committed
866
<p>  Ideally this requires a separate test program that can be run
Bruce Momjian's avatar
Bruce Momjian committed
867 868
  at initdb time or optionally later.  Consider O_SYNC when
  O_DIRECT exists.
Bruce Momjian's avatar
Bruce Momjian committed
869
</p>
Bruce Momjian's avatar
Bruce Momjian committed
870
  </li><li>%Add an option to sync() before fsync()'ing checkpoint files
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
871
  </li><li>Add program to test if fsync has a delay compared to non-fsync
872
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
873
<h1><a name="section_15">Cache Usage</a></h1>
Bruce Momjian's avatar
Bruce Momjian committed
874

875 876 877 878 879 880 881 882 883 884 885 886 887 888 889
<ul>
  <li>Allow free-behind capability for large sequential scans, perhaps using
  posix_fadvise()
<p>  Posix_fadvise() can control both sequential/random file caching and
  free-behind behavior, but it is unclear how the setting affects other
  backends that also have the file open, and the feature is not supported
  on all operating systems.
</p>
  </li><li>Speed up COUNT(*)
<p>  We could use a fixed row count and a +/- count to follow MVCC
  visibility rules, or a single cached value could be used and
  invalidated if anyone modifies the table.  Another idea is to
  get a count directly from a unique index, but for this to be
  faster than a sequential scan it must avoid access to the heap
  to obtain tuple visibility information.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
890 891
</p>
  </li><li>Add estimated_count(*) to return an estimate of COUNT(*)
Bruce Momjian's avatar
Bruce Momjian committed
892
<p>  This would use the planner ANALYZE statistics to return an estimated
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
893
  count.
894 895
</p>
  </li><li>Allow data to be pulled directly from indexes
Bruce Momjian's avatar
Bruce Momjian committed
896 897
<p>  Currently indexes do not have enough tuple visibility information 
  to allow data to be pulled from the index without also accessing 
898
  the heap.  One way to allow this is to set a bit on index tuples 
Bruce Momjian's avatar
Bruce Momjian committed
899 900
  to indicate if a tuple is currently visible to all transactions 
  when the first valid heap lookup happens.  This bit would have to 
901 902 903 904 905 906 907 908
  be cleared when a heap tuple is expired.  
</p>
<p>  Another idea is to maintain a bitmap of heap pages where all rows
  are visible to all backends, and allow index lookups to reference 
  that bitmap to avoid heap lookups, perhaps the same bitmap we might
  add someday to determine which heap pages need vacuuming.  Frequently
  accessed bitmaps would have to be stored in shared memory.  One 8k
  page of bitmaps could track 512MB of heap pages.
909
</p>
910
  </li><li>Consider automatic caching of statements at various levels:
911 912 913 914 915 916
  <ul>
    <li>Parsed query tree
    </li><li>Query execute plan
    </li><li>Query results
  </li></ul>
  </li><li>Allow sequential scans to take advantage of other concurrent
Bruce Momjian's avatar
Bruce Momjian committed
917
  sequential scans, also called "Synchronised Scanning"
918 919 920 921
<p>  One possible implementation is to start sequential scans from the lowest
  numbered buffer in the shared cache, and when reaching the end wrap
  around to the beginning, rather than always starting sequential scans
  at the start of the table.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
922 923 924
</p>
  </li><li>Consider increasing internal areas when shared buffers is increased
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php">http://archives.postgresql.org/pgsql-hackers/2005-10/msg01419.php</a>
925
</p>
926
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
927
<h1><a name="section_16">Vacuum</a></h1>
928 929 930

<ul>
  <li>Improve speed with indexes
Bruce Momjian's avatar
Bruce Momjian committed
931
<p>  For large table adjustments during VACUUM FULL, it is faster to 
Bruce Momjian's avatar
Bruce Momjian committed
932
  reindex rather than update the index.
933
</p>
Bruce Momjian's avatar
Bruce Momjian committed
934 935
  </li><li>Reduce lock time during VACUUM FULL by moving tuples with read lock,
  then write lock and truncate table
936 937 938
<p>  Moved tuples are invisible to other backends so they don't require a
  write lock. However, the read lock promotion to write lock could lead
  to deadlock situations.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
939
</p>
Bruce Momjian's avatar
Bruce Momjian committed
940 941
  </li><li>Auto-fill the free space map by scanning the buffer cache or by
  checking pages written by the background writer
942
<p>  <a href="http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php">http://archives.postgresql.org/pgsql-hackers/2006-02/msg01125.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
943
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php">http://archives.postgresql.org/pgsql-hackers/2006-03/msg00011.php</a>
944
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
945 946 947 948 949
  </li><li>Create a bitmap of pages that need vacuuming
<p>  Instead of sequentially scanning the entire table, have the background
  writer or some other process record pages that have expired rows, then
  VACUUM can look at just those pages rather than the entire table.  In
  the event of a system crash, the bitmap would probably be invalidated.
Bruce Momjian's avatar
Bruce Momjian committed
950 951 952 953
  One complexity is that index entries still have to be vacuumed, and
  doing this without an index scan (by using the heap values to find the
  index entry) might be slow and unreliable, especially for user-defined
  index functions.
954
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
955 956
  </li><li>Allow FSM to return free space toward the beginning of the heap file,
  in hopes that empty pages at the end can be truncated by VACUUM
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
957 958
  </li><li>Allow FSM page return free space based on table clustering, to assist
  in maintaining clustering?
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
959 960
  </li><li>Consider shrinking expired tuples to just their headers
<p>  <a href="http://archives.postgresql.org/pgsql-patches/2006-03/msg00142.php">http://archives.postgresql.org/pgsql-patches/2006-03/msg00142.php</a>
961
</p>
Bruce Momjian's avatar
Bruce Momjian committed
962 963 964 965 966 967
  </li><li>Allow heap reuse of UPDATEd rows if no indexed columns are changed,
  and old and new versions are on the same heap page?
<p>  While vacuum handles DELETEs fine, updating of non-indexed columns, like
  counters, are difficult for VACUUM to handle efficiently.  This method
  is possible for same-page updates because a single index row can be
  used to point to both old and new values.
968
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-06/msg01305.php">http://archives.postgresql.org/pgsql-hackers/2006-06/msg01305.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
969
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-06/msg01534.php">http://archives.postgresql.org/pgsql-hackers/2006-06/msg01534.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
970
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
971 972
  </li><li>Reuse index tuples that point to heap tuples that are not visible to 
  anyone?
973 974
  </li><li>Auto-vacuum
  <ul>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
975
    <li>Use free-space map information to guide refilling
Bruce Momjian's avatar
Bruce Momjian committed
976 977
    </li><li>%Issue log message to suggest VACUUM FULL if a table is nearly
          empty?
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
978 979
    </li><li>Improve xid wraparound detection by recording per-table rather
          than per-database
Bruce Momjian's avatar
Bruce Momjian committed
980
    </li><li>Consider logging activity either to the logs or a system view
Bruce Momjian's avatar
Bruce Momjian committed
981
    </li><li>Turn on by default
982 983
<p>          <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg01852.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg01852.php</a>
</p>
984 985
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
986
<h1><a name="section_17">Locking</a></h1>
987 988

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
989
  <li>Fix priority ordering of read and write light-weight locks (Neil)
990
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
991
<h1><a name="section_18">Startup Time Improvements</a></h1>
992 993

<ul>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
994
  <li>Experiment with multi-threaded backend for backend creation [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?thread">thread</a>]
995 996
<p>  This would prevent the overhead associated with process creation. Most
  operating systems have trivial process creation time compared to
Bruce Momjian's avatar
Bruce Momjian committed
997
  database startup overhead, but a few operating systems (Win32,
Bruce Momjian's avatar
Bruce Momjian committed
998
  Solaris) might benefit from threading.  Also explore the idea of
999
  a single session using multiple threads to execute a statement faster.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1000 1001 1002
</p>
  </li><li>Experiment with multi-threaded backend better resource utilization
<p>  This would allow a single query to make use of multiple CPU's or
Bruce Momjian's avatar
Bruce Momjian committed
1003 1004 1005 1006
  multiple I/O channels simultaneously.  One idea is to create a
  background reader that can pre-fetch sequential and index scan
  pages needed by other backends.  This could be expanded to allow
  concurrent reads from multiple devices in a partitioned table.
1007 1008 1009 1010 1011 1012 1013
</p>
  </li><li>Add connection pooling
<p>  It is unclear if this should be done inside the backend code or done
  by something external like pgpool. The passing of file descriptors to
  existing backends is one of the difficulties with a backend approach.
</p>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1014
<h1><a name="section_19">Write-Ahead Log</a></h1>
1015 1016 1017

<ul>
  <li>Eliminate need to write full pages to WAL before page modification [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?wal">wal</a>]
Bruce Momjian's avatar
Bruce Momjian committed
1018
<p>  Currently, to protect against partial disk page writes, we write
1019 1020 1021 1022
  full page images to WAL before they are modified so we can correct any
  partial page writes during recovery.  These pages can also be
  eliminated from point-in-time archive files.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1023
  <ul>
Bruce Momjian's avatar
Bruce Momjian committed
1024
    <li>When off, write CRC to WAL and check file system blocks
Bruce Momjian's avatar
Bruce Momjian committed
1025
           on recovery
Bruce Momjian's avatar
Bruce Momjian committed
1026 1027 1028
<p>           If CRC check fails during recovery, remember the page in case
           a later CRC for that page properly matches.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1029 1030 1031
    </li><li>Write full pages during file system write and not when
           the page is modified in the buffer cache
<p>           This allows most full page writes to happen in the background
Bruce Momjian's avatar
Bruce Momjian committed
1032 1033 1034
           writer.  It might cause problems for applying WAL on recovery
           into a partially-written page, but later the full page will be
           replaced from WAL.
Bruce Momjian's avatar
Bruce Momjian committed
1035 1036
</p>
  </li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1037
  </li><li>Allow WAL traffic to be streamed to another server for stand-by
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1038
  replication
1039
  </li><li>Reduce WAL traffic so only modified values are written rather than
Bruce Momjian's avatar
Bruce Momjian committed
1040
  entire rows?
1041 1042
  </li><li>Allow the pg_xlog directory location to be specified during initdb
  with a symlink back to the /data location
Bruce Momjian's avatar
Bruce Momjian committed
1043
  </li><li>Allow WAL information to recover corrupted pg_controldata
1044
<p>  <a href="http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php">http://archives.postgresql.org/pgsql-patches/2006-06/msg00025.php</a>
1045
</p>
1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
  </li><li>Find a way to reduce rotational delay when repeatedly writing
  last WAL page
<p>  Currently fsync of WAL requires the disk platter to perform a full
  rotation to fsync again. One idea is to write the WAL to different
  offsets that might reduce the rotational delay.
</p>
  </li><li>Allow buffered WAL writes and fsync
<p>  Instead of guaranteeing recovery of all committed transactions, this
  would provide improved performance by delaying WAL writes and fsync
  so an abrupt operating system restart might lose a few seconds of
  committed transactions but still be consistent.  We could perhaps
  remove the 'fsync' parameter (which results in an an inconsistent
  database) in favor of this capability.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1059
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1060 1061
  </li><li>Allow WAL logging to be turned off for a table, but the table
  might be dropped or truncated during crash recovery [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1062
<p>  Allow tables to bypass WAL writes and just fsync() dirty pages on
Bruce Momjian's avatar
Bruce Momjian committed
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075
  commit.  This should be implemented using ALTER TABLE, e.g. ALTER 
  TABLE PERSISTENCE [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo? DROP | TRUNCATE | DEFAULT "> DROP | TRUNCATE | DEFAULT </a>].  Tables using 
  non-default logging should not use referential integrity with 
  default-logging tables.  A table without dirty buffers during a
  crash could perhaps avoid the drop/truncate.
</p>
  </li><li>Allow WAL logging to be turned off for a table, but the table would
  avoid being truncated/dropped [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?walcontrol">walcontrol</a>]
<p>  To do this, only a single writer can modify the table, and writes 
  must happen only on new pages so the new pages can be removed during
  crash recovery.  Readers can continue accessing the table.  Such 
  tables probably cannot have indexes.  One complexity is the handling 
  of indexes on TOAST tables.
1076 1077
</p>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1078
<h1><a name="section_20">Optimizer / Executor</a></h1>
1079 1080

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
1081
  <li>Improve selectivity functions for geometric operators
Bruce Momjian's avatar
Bruce Momjian committed
1082
  </li><li>Allow ORDER BY ... LIMIT # to select high/low value without sort or
1083
  index using a sequential scan for highest/lowest values
Bruce Momjian's avatar
Bruce Momjian committed
1084
<p>  Right now, if no index exists, ORDER BY ... LIMIT # requires we sort
1085 1086
  all values to return the high/low value.  Instead The idea is to do a 
  sequential scan to find the high/low value, thus avoiding the sort.
Bruce Momjian's avatar
Bruce Momjian committed
1087
  MIN/MAX already does this, but not for LIMIT &gt; 1.
1088
</p>
1089 1090 1091 1092 1093
  </li><li>Precompile SQL functions to avoid overhead
  </li><li>Create utility to compute accurate random_page_cost value
  </li><li>Improve ability to display optimizer analysis using OPTIMIZER_DEBUG
  </li><li>Have EXPLAIN ANALYZE highlight poor optimizer estimates
  </li><li>Consider using hash buckets to do DISTINCT, rather than sorting
Bruce Momjian's avatar
Bruce Momjian committed
1094 1095
<p>  This would be beneficial when there are few distinct values.  This is
  already used by GROUP BY.
1096
</p>
1097
  </li><li>Log statements where the optimizer row estimates were dramatically
Bruce Momjian's avatar
Bruce Momjian committed
1098
  different from the number of rows actually found?
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1099 1100 1101
  </li><li>Consider compressed annealing to search for query plans
<p>  This might replace GEQO, <a href="http://sixdemonbag.org/Djinni">http://sixdemonbag.org/Djinni</a>.
</p>
1102
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1103
<h1><a name="section_21">Miscellaneous Performance</a></h1>
1104 1105 1106 1107 1108 1109

<ul>
  <li>Do async I/O for faster random read-ahead of data
<p>  Async I/O allows multiple I/O requests to be sent to the disk with
  results coming back asynchronously.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1110
  </li><li>Use mmap() rather than SYSV shared memory or to write WAL files?
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123
<p>  This would remove the requirement for SYSV SHM but would introduce
  portability issues. Anonymous mmap (or mmap to /dev/zero) is required
  to prevent I/O overhead.
</p>
  </li><li>Consider mmap()'ing files into a backend?
<p>  Doing I/O to large tables would consume a lot of address space or
  require frequent mapping/unmapping.  Extending the file also causes
  mapping problems that might require mapping only individual pages,
  leading to thousands of mappings.  Another problem is that there is no
  way to _prevent_ I/O to disk from the dirty shared buffers so changes
  could hit disk before WAL is written.
</p>
  </li><li>Add a script to ask system configuration questions and tune postgresql.conf
Bruce Momjian's avatar
Bruce Momjian committed
1124 1125 1126 1127 1128 1129 1130 1131
  </li><li>Merge xmin/xmax/cmin/cmax back into three header fields
<p>  Before subtransactions, there used to be only three fields needed to
  store these four values. This was possible because only the current
  transaction looks at the cmin/cmax values. If the current transaction
  created and expired the row the fields stored where xmin (same as
  xmax), cmin, cmax, and if the transaction was expiring a row from a
  another transaction, the fields stored were xmin (cmin was not
  needed), xmax, and cmax. Such a system worked because a transaction
Bruce Momjian's avatar
Bruce Momjian committed
1132
  could only see rows from another completed transaction. However,
Bruce Momjian's avatar
Bruce Momjian committed
1133 1134 1135 1136 1137
  subtransactions can see rows from outer transactions, and once the
  subtransaction completes, the outer transaction continues, requiring
  the storage of all four fields. With subtransactions, an outer
  transaction can create a row, a subtransaction expire it, and when the
  subtransaction completes, the outer transaction still has to have
Bruce Momjian's avatar
Bruce Momjian committed
1138
  proper visibility of the row's cmin, for example, for cursors.
Bruce Momjian's avatar
Bruce Momjian committed
1139 1140
</p>
<p>  One possible solution is to create a phantom cid which represents a
Bruce Momjian's avatar
Bruce Momjian committed
1141 1142
  cmin/cmax pair and is stored in local memory.  Another idea is to
  store both cmin and cmax only in local memory.
Bruce Momjian's avatar
Bruce Momjian committed
1143
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1144 1145
  </li><li>Consider ways of storing rows more compactly on disk
  <ul>
1146 1147 1148 1149 1150 1151 1152
    <li>Support a smaller header for short variable-length fields?
<p>          One idea is to create zero-or-one-byte-header versions 
          of varlena data types.  In involves setting the high-bit and 
          0-127 length in the single-byte header, or clear the high bit
          and store the 7-bit ASCII value in the rest of the byte.
          The small-header versions have no alignment requirements.
          <a href="http://archives.postgresql.org/pgsql-hackers/2006-09/msg01372.php">http://archives.postgresql.org/pgsql-hackers/2006-09/msg01372.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
1153
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1154 1155
    </li><li>Reduce the row header size?
  </li></ul>
1156
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1157
<h1><a name="section_22">Source Code</a></h1>
1158 1159 1160 1161 1162

<ul>
  <li>Add use of 'const' for variables in source tree
  </li><li>Move some things from /contrib into main tree
  </li><li>Move some /contrib modules out to their own project sites
1163 1164 1165
<p>  Particularly, move GPL-licensed /contrib/userlock and 
  /contrib/dbmirror/clean_pending.pl.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1166
  </li><li>%Remove warnings created by -Wcast-align
1167 1168 1169 1170
  </li><li>Move platform-specific ps status display info from ps_status.c to ports
  </li><li>Add optional CRC checksum to heap and index pages
  </li><li>Improve documentation to build only interfaces (Marc)
  </li><li>Remove or relicense modules that are not under the BSD license, if possible
Bruce Momjian's avatar
Bruce Momjian committed
1171
  </li><li>%Remove memory/file descriptor freeing before ereport(ERROR)
1172
  </li><li>Acquire lock on a relation before building a relcache entry for it
Bruce Momjian's avatar
Bruce Momjian committed
1173 1174
  </li><li>%Promote debug_query_string into a server-side function current_query()
  </li><li>%Allow the identifier length to be increased via a configure option
1175
  </li><li>Allow cross-compiling by generating the zic database on the target system
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
1176
  </li><li>Improve NLS maintenance of libpgport messages linked onto applications
1177 1178 1179 1180 1181 1182
  </li><li>Allow ecpg to work with MSVC and BCC
  </li><li>Add xpath_array() to /contrib/xml2 to return results as an array
  </li><li>Allow building in directories containing spaces
<p>  This is probably not possible because 'gmake' and other compiler tools
  do not fully support quoting of paths with spaces.
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1183
  </li><li>Fix sgmltools so PDFs can be generated with bookmarks
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1184 1185
  </li><li>Use UTF8 encoding for NLS messages so all server encodings can
  read them properly
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1186
  </li><li>Update Bonjour to work with newer cross-platform SDK
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1187 1188
  </li><li>Split out libpq pgpass and environment documentation sections to make
  it easier for non-developers to find
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1189 1190 1191
  </li><li>Improve port/qsort() to handle sorts with 50% unique and 50% duplicate
  value [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?qsort">qsort</a>]
<p>  This involves choosing better pivot points for the quicksort.
Bruce Momjian's avatar
Bruce Momjian committed
1192
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1193
  </li><li>Consider detoasting keys before sorting
Bruce Momjian's avatar
Bruce Momjian committed
1194
  </li><li>Consider GnuTLS if OpenSSL license becomes a problem
1195
<p>  <a href="http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php">http://archives.postgresql.org/pgsql-patches/2006-05/msg00040.php</a>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1196
</p>
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208
  </li><li>Win32
  <ul>
    <li>Remove configure.in check for link failure when cause is found
    </li><li>Remove readdir() errno patch when runtime/mingwex/dirent.c rev
          1.4 is released
    </li><li>Remove psql newline patch when we find out why mingw outputs an
          extra newline
    </li><li>Allow psql to use readline once non-US code pages work with
          backslashes
    </li><li>Re-enable timezone output on log_line_prefix '%t' when a
          shorter timezone string is available
    </li><li>Fix problem with shared memory on the Win32 Terminal Server
1209 1210 1211
    </li><li>Improve signal handling
<p>          <a href="http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php">http://archives.postgresql.org/pgsql-patches/2005-06/msg00027.php</a>
</p>
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1212 1213 1214
    </li><li>Add long file support for binary pg_dump output
<p>          While Win32 supports 64-bit files, the MinGW API does not,
          meaning we have to build an fseeko replacement on top of the
Bruce Momjian's avatar
Bruce Momjian committed
1215 1216 1217
          Win32 API, and we have to make sure MinGW handles it.  Another
          option is to wait for the MinGW project to fix it, or use the
          code from the LibGW32C project as a guide.
Bruce Momjian's avatar
Add:  
Bruce Momjian committed
1218
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1219
    </li><li>Check WSACancelBlockingCall() for interrupts [<a href="http://momjian.postgresql.org/cgi-bin/pgtodo?win32intr">win32intr</a>]
1220 1221 1222 1223 1224 1225 1226
  </li></ul>
  </li><li>Wire Protocol Changes
  <ul>
    <li>Allow dynamic character set handling
    </li><li>Add decoded type, length, precision
    </li><li>Use compression?
    </li><li>Update clients to use data types, typmod, <a href="http://schema.table.column">schema.table.column</a>/ names
1227
          of result sets using new statement protocol
1228 1229
  </li></ul>
</li></ul>
Bruce Momjian's avatar
Bruce Momjian committed
1230
<h1><a name="section_23">Features We Do _Not_ Want</a></h1>
1231 1232

<ul>
Bruce Momjian's avatar
Bruce Momjian committed
1233
  <li>All backends running as threads in a single process (not wanted)
1234 1235 1236 1237
<p>  This eliminates the process protection we get from the current setup.
  Thread creation is usually the same overhead as process creation on
  modern systems, so it seems unwise to use a pure threaded model.
</p>
Bruce Momjian's avatar
Bruce Momjian committed
1238
  </li><li>Optimizer hints (not wanted)
1239 1240 1241
<p>  Optimizer hints are used to work around problems in the optimizer.  We
  would rather have the problems reported and fixed.
  <a href="http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php">http://archives.postgresql.org/pgsql-hackers/2006-08/msg00506.php</a>
Bruce Momjian's avatar
Bruce Momjian committed
1242 1243 1244 1245 1246
</p>
  </li><li>Allow AS in "SELECT col AS label" to be optional (not wanted)
<p>  Because we support postfix operators, it isn't possible to make AS
  optional and continue to use bison.
  <a href="http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php">http://archives.postgresql.org/pgsql-sql/2006-08/msg00164.php</a>
1247 1248
</p>
</li></ul>
1249 1250
<hr/>

Bruce Momjian's avatar
Bruce Momjian committed
1251
<h2><a name="section_23_1">Developers who have claimed items are:</a></h2>
1252 1253 1254
<ul>
  <li>Alvaro is Alvaro Herrera &lt;<a href="mailto:alvherre@dcc.uchile.cl">alvherre@dcc.uchile.cl</a>&gt;
  </li><li>Andrew is Andrew Dunstan &lt;<a href="mailto:andrew@dunslane.net">andrew@dunslane.net</a>&gt;
Bruce Momjian's avatar
Bruce Momjian committed
1255
  </li><li>Bruce is Bruce Momjian &lt;<a href="mailto:bruce@momjian.us">bruce@momjian.us</a>&gt; of EnterpriseDB
1256 1257 1258
  </li><li>Christopher is Christopher Kings-Lynne &lt;<a href="mailto:chriskl@familyhealth.com.au">chriskl@familyhealth.com.au</a>&gt; of
    Family Health Network
  </li><li>D'Arcy is D'Arcy J.M. Cain &lt;<a href="mailto:darcy@druid.net">darcy@druid.net</a>&gt; of The Cain Gang Ltd.
Bruce Momjian's avatar
Bruce Momjian committed
1259
  </li><li>David is David Fetter &lt;<a href="mailto:david@fetter.org">david@fetter.org</a>&gt;
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
  </li><li>Fabien is Fabien Coelho &lt;<a href="mailto:coelho@cri.ensmp.fr">coelho@cri.ensmp.fr</a>&gt;
  </li><li>Gavin is Gavin Sherry &lt;<a href="mailto:swm@linuxworld.com.au">swm@linuxworld.com.au</a>&gt; of Alcove Systems Engineering
  </li><li>Greg is Greg Sabino Mullane &lt;<a href="mailto:greg@turnstep.com">greg@turnstep.com</a>&gt;
  </li><li>Jan is Jan Wieck &lt;<a href="mailto:JanWieck@Yahoo.com">JanWieck@Yahoo.com</a>&gt; of Afilias, Inc.
  </li><li>Joe is Joe Conway &lt;<a href="mailto:mail@joeconway.com">mail@joeconway.com</a>&gt;
  </li><li>Karel is Karel Zak &lt;<a href="mailto:zakkr@zf.jcu.cz">zakkr@zf.jcu.cz</a>&gt;
  </li><li>Magnus is Magnus Hagander &lt;<a href="mailto:mha@sollentuna.net">mha@sollentuna.net</a>&gt;
  </li><li>Marc is Marc Fournier &lt;<a href="mailto:scrappy@hub.org">scrappy@hub.org</a>&gt; of PostgreSQL, Inc.
  </li><li>Matthew T. O'Connor &lt;<a href="mailto:matthew@zeut.net">matthew@zeut.net</a>&gt;
  </li><li>Michael is Michael Meskes &lt;<a href="mailto:meskes@postgresql.org">meskes@postgresql.org</a>&gt; of Credativ
  </li><li>Neil is Neil Conway &lt;<a href="mailto:neilc@samurai.com">neilc@samurai.com</a>&gt;
  </li><li>Oleg is Oleg Bartunov &lt;<a href="mailto:oleg@sai.msu.su">oleg@sai.msu.su</a>&gt;
Bruce Momjian's avatar
Bruce Momjian committed
1272
  </li><li>Pavel is Pavel Stehule &lt;<a href="mailto:pavel.stehule@hotmail.com">pavel.stehule@hotmail.com</a>&gt;
1273 1274 1275 1276 1277
  </li><li>Peter is Peter Eisentraut &lt;<a href="mailto:peter_e@gmx.net">peter_e@gmx.net</a>&gt;
  </li><li>Philip is Philip Warner &lt;<a href="mailto:pjw@rhyme.com.au">pjw@rhyme.com.au</a>&gt; of Albatross Consulting Pty. Ltd.
  </li><li>Rod is Rod Taylor &lt;<a href="mailto:pg@rbt.ca">pg@rbt.ca</a>&gt;
  </li><li>Simon is Simon Riggs &lt;<a href="mailto:simon@2ndquadrant.com">simon@2ndquadrant.com</a>&gt;
  </li><li>Stephan is Stephan Szabo &lt;<a href="mailto:sszabo@megazone23.bigpanda.com">sszabo@megazone23.bigpanda.com</a>&gt;
Bruce Momjian's avatar
Bruce Momjian committed
1278
  </li><li>Tatsuo is Tatsuo Ishii &lt;<a href="mailto:ishii@sraoss.co.jp">ishii@sraoss.co.jp</a>&gt; of SRA OSS, Inc. Japan
Bruce Momjian's avatar
Bruce Momjian committed
1279
  </li><li>Teodor is Teodor Sigaev &lt;<a href="mailto:teodor@sigaev.ru">teodor@sigaev.ru</a>&gt;
1280 1281
  </li><li>Tom is Tom Lane &lt;<a href="mailto:tgl@sss.pgh.pa.us">tgl@sss.pgh.pa.us</a>&gt; of Red Hat
</li></ul>
Bruce Momjian's avatar
Done:  
Bruce Momjian committed
1282

1283 1284
</body>
</html>