== API ==

[WITH [ RECURSIVE ] with_query [, ...]
plannable_statement

where with_query is:
with_query_name [ ( column_name [, ...] ) ] AS ( plannable_statement )

and plannable_statement (see gram.y for the origin of this term) query is:
( select | insert | update | delete )

INSERT, UPDATE and DELETE can have a RETURNING clause. In that case, the results of the RETURNING are stored inside the CTE.

== Implementation ==
Since this patch is quite complicated, I've split it up to a few milestones:

{{TodoSubsection}}
{{TodoItemDone|Add a new executor node, Derived Table scan, to support scanning from tuplestores.|
* None of the existing executor nodes did only this (all of them deal with a subplan of some kind), so adding a new one made seemed to make sense.
* In the future, this node can be used to support other queries in WITHs (COPY, for example).
}}
{{TodoItem|Make all WITH queries work like wCTEs would|
* What this means is that we run all WITH queries to completion and store the results in tuplestores. Then we proceed to execute the main plan.
* This means changes to the portal code, EXPLAIN ANALYZE, SQL function and SPI execution code.
* There's a proof-of-concept code in the git repo that works, but it's still very messy and only changes the portal code. It also doesn't handle snapshots correctly, but it's not yet clear what the code should look like (see NOTE near the bottom of this item). Not all regression tests for WITH are applicable because we don't need to support non-top-level CTEs, so they've been taken out from the git repo. Also a RECURSIVE query must not loop forever, so those tests are gone too. All remaining tests are passed now.
'''NOTE''': There's an unresolved issue that affects this part of the plan, see http://archives.postgresql.org/message-id/4C49E821.8060206@cs.helsinki.fi
}}
{{TodoItem|Allow INSERT, UPDATE and DELETE inside top-level CTEs|
* I think this stage can be completed by just copy-pasting from the patch considered for 9.0.
}}
{{TodoItem|Allow the "main query" to be INSERT, UPDATE or DELETE|
* There were some problems with this for 9.0, see http://archives.postgresql.org/pgsql-hackers/2010-02/msg01065.php
}}
{{TodoEndSubsection}}

== Usage ==

== More Features ==

These are just possibilities. They are not a roadmap. Your measurements may vary.

DCL

DDL

CTAS support

Non-Row-Returning CTEs

== Discussion Logs ==
The execution order of each DML should be syntactically? DML -> SELECT?:
http://archives.postgresql.org/message-id/6537.1255029119@sss.pgh.pa.us

pg_plan_query returns List of PlannedStmt instead of one PlannedStmt:
http://archives.postgresql.org/message-id/1296.1265839071@sss.pgh.pa.us

This should work:
with t1 as (delete from src returning *), t2 as(insert into dest select * from t1 returning *) select * from t2

RECURSIVE query is fine as long as 1) it's SELECT-only and 2) it doesn't loop forever. WITH RETURNING cannot be contained in RECURSIVE. A wCTE can of course refer to the result of the recursive SELECT query with INSERT .. SELECT, UPDATE .. FROM or DELETE .. USING. :
http://archives.postgresql.org/message-id/4C5857DF.4000801@cs.helsinki.fi

WriteableCTEs

2010-08-13T18:30:18Z

Umitanuki: /* Discussion Logs */

== API ==

[WITH [ RECURSIVE ] with_query [, ...]
plannable_statement

where with_query is:
with_query_name [ ( column_name [, ...] ) ] AS ( plannable_statement )

and plannable_statement (see gram.y for the origin of this term) query is:
( select | insert | update | delete )

INSERT, UPDATE and DELETE can have a RETURNING clause. In that case, the results of the RETURNING are stored inside the CTE.

== Implementation ==
Since this patch is quite complicated, I've split it up to a few milestones:

{{TodoSubsection}}
{{TodoItemDone|Add a new executor node, Derived Table scan, to support scanning from tuplestores.|
* None of the existing executor nodes did only this (all of them deal with a subplan of some kind), so adding a new one made seemed to make sense.
* In the future, this node can be used to support other queries in WITHs (COPY, for example).
}}
{{TodoItem|Make all WITH queries work like wCTEs would|
* What this means is that we run all WITH queries to completion and store the results in tuplestores. Then we proceed to execute the main plan.
* This means changes to the portal code, EXPLAIN ANALYZE, SQL function and SPI execution code.
* There's a proof-of-concept code in the git repo that works, but it's still very messy and only changes the portal code. It also doesn't handle snapshots correctly, but it's not yet clear what the code should look like (see NOTE near the bottom of this item). Not all regression tests for WITH are applicable because we don't need to support non-top-level CTEs, so they've been taken out from the git repo. Also a RECURSIVE query must not loop forever, so those tests are gone too. All remaining tests are passed now.
'''NOTE''': There's an unresolved issue that affects this part of the plan, see http://archives.postgresql.org/message-id/4C49E821.8060206@cs.helsinki.fi
}}
{{TodoItem|Allow INSERT, UPDATE and DELETE inside top-level CTEs|
* I think this stage can be completed by just copy-pasting from the patch considered for 9.0.
}}
{{TodoItem|Allow the "main query" to be INSERT, UPDATE or DELETE|
* There were some problems with this for 9.0, see http://archives.postgresql.org/pgsql-hackers/2010-02/msg01065.php
}}
{{TodoEndSubsection}}

== Usage ==

== More Features ==

These are just possibilities. They are not a roadmap. Your measurements may vary.

DCL

DDL

CTAS support

Non-Row-Returning CTEs

== Discussion Logs ==
The execution order of each DML should be syntactically? DML -> SELECT?:
http://archives.postgresql.org/message-id/6537.1255029119@sss.pgh.pa.us

pg_plan_query returns List of PlannedStmt instead of one PlannedStmt:
http://archives.postgresql.org/message-id/1296.1265839071@sss.pgh.pa.us

This should work:
with t1 as (delete from src returning *), t2 as(insert into dest select * from t1 returning *) select * from t2

RECURSIVE query is fine as long as 1) it's SELECT-only and 2) it doesn't loop forever. A wCTE can of course refer to the result of the recursive SELECT query with INSERT .. SELECT, UPDATE .. FROM or DELETE .. USING. :
http://archives.postgresql.org/message-id/4C5857DF.4000801@cs.helsinki.fi

WriteableCTEs

2010-08-13T18:10:57Z

Umitanuki: /* Discussion Logs */

WriteableCTEs

2010-08-13T18:08:56Z

Umitanuki:

== API ==

[WITH [ RECURSIVE ] with_query [, ...]
plannable_statement

where with_query is:
with_query_name [ ( column_name [, ...] ) ] AS ( plannable_statement )

and plannable_statement (see gram.y for the origin of this term) query is:
( select | insert | update | delete )

INSERT, UPDATE and DELETE can have a RETURNING clause. In that case, the results of the RETURNING are stored inside the CTE.

== Implementation ==
Since this patch is quite complicated, I've split it up to a few milestones:

{{TodoSubsection}}
{{TodoItemDone|Add a new executor node, Derived Table scan, to support scanning from tuplestores.|
* None of the existing executor nodes did only this (all of them deal with a subplan of some kind), so adding a new one made seemed to make sense.
* In the future, this node can be used to support other queries in WITHs (COPY, for example).
}}
{{TodoItem|Make all WITH queries work like wCTEs would|
* What this means is that we run all WITH queries to completion and store the results in tuplestores. Then we proceed to execute the main plan.
* This means changes to the portal code, EXPLAIN ANALYZE, SQL function and SPI execution code.
* There's a proof-of-concept code in the git repo that works, but it's still very messy and only changes the portal code. It also doesn't handle snapshots correctly, but it's not yet clear what the code should look like (see NOTE near the bottom of this item). Not all regression tests for WITH are applicable because we don't need to support non-top-level CTEs, so they've been taken out from the git repo. Also a RECURSIVE query must not loop forever, so those tests are gone too. All remaining tests are passed now.
'''NOTE''': There's an unresolved issue that affects this part of the plan, see http://archives.postgresql.org/message-id/4C49E821.8060206@cs.helsinki.fi
}}
{{TodoItem|Allow INSERT, UPDATE and DELETE inside top-level CTEs|
* I think this stage can be completed by just copy-pasting from the patch considered for 9.0.
}}
{{TodoItem|Allow the "main query" to be INSERT, UPDATE or DELETE|
* There were some problems with this for 9.0, see http://archives.postgresql.org/pgsql-hackers/2010-02/msg01065.php
}}
{{TodoEndSubsection}}

== Usage ==

== More Features ==

These are just possibilities. They are not a roadmap. Your measurements may vary.

DCL

DDL

CTAS support

Non-Row-Returning CTEs

== Discussion Logs ==
http://archives.postgresql.org/message-id/6537.1255029119@sss.pgh.pa.us

pg_plan_query returns List of PlannedStmt instead of one PlannedStmt:
http://archives.postgresql.org/message-id/1296.1265839071@sss.pgh.pa.us

TestFest For 9 JP

2010-04-03T15:32:24Z

Umitanuki: /* その他 */

== 概要 ==
PostgreSQL 9のベータリリースに合わせて、Hackathon形式で集中テストを行います。同日にサンフランシスコで行われる[[SFPUG_Beta_Test_Day]]と連携します。

'''以下は現在時点での情報です。変更になる可能性がありますのでご注意ください。'''

Twitterハッシュタグ : [http://twitter.com/#search?q=%23TestFest #TestFest]

== 内容 ==

基本的にはSFPUGのテスト内容に準じます。

* コンパイルテスト
** 各種オプション付けて
** contribモジュール
** 外部モジュール
** 回帰テスト
* pgBenchテスト
** 8.4 vs. 9.0
* 新機能テスト（[http://lets.postgresql.jp/documents/technical/9.0/1 新機能紹介ページ]）
** ホットスタンバイ
** ストリーミングレプリケーション
** 除外制約
** LISTEN/NOTIFY
** DO文（各種言語で）
** 追加されたウィンドウ関数フレーム
** その他募集中
* データベース移行テストデータの用意もお願いします！
* アプリケーション移行テスト
** MediaWiki
** Slony
** Bucardo
** pgsnmpd
** pgpool
** textsearch_senna
** その他募集中
* ドライバ
** Java
** Python
*** psycopg2
** Perl
*** DBD::Pg
** R
** Ruby
** その他募集中
* ORMツール
** SQLAlchemy
** Django
** Hibernate
** その他募集中

9を先取りして勉強することができます。記事等観てもぱっとわからない機能も隣の人が教えてくれます。また、PostgreSQLの今後の開発や普段疑問に思っていることなどについて自由にディスカッションできます。たぶん飲みながらやるのでぐだぐだになりそうです。

== 場所 ==
フォルシア株式会社会議室

東京都新宿区新宿4丁目3-25 オリックス新宿ビル9階
[http://maps.google.co.jp/maps?f=q&hl=ja&geocode=&time=&date=&ttype=&q=%93%8C%8B%9E%93s%90V%8Fh%8B%E6%90V%8Fh%82S%92%9A%96%DA%82R-%82Q%82T&ie=UTF8&ll=35.689321,139.704277&spn=0.010386,0.013776&z=14&iwloc=addr&om=1&source=embed Google Map]

新宿駅南口徒歩3分
東京メトロ新宿3丁目駅徒歩1分

== 時間 ==
4月4日午前3時～午前10時（日本時間）
※サンフランシスコ時間4月3日午前11時～午後6時で行われるTestに合わせています。

時間が時間なので前日からのスタンバイあり

== 食事 ==
軽食を用意します。持ち込み（飲食）歓迎。徒歩圏内でも調達可能。

== 持ち物 ==
ラップトップ持参。インターネット環境（DHCP）は用意します。Wi-fiに対応できるかは調整中。各自テスト環境を外部サーバに用意できれば尚可。

ラップトップは若干台数貸し出し可能です。お気軽にご連絡ください。

{| border="1"
|+
'''準備機材'''
| '''所有者''' || '''物品''' || '''備考'''
|-
| 永安 || AirMac Express（無線LAN AP) ||
|-
| 永安 || VAIO X || 作業用
|-
| 永安 || MacBook || テスト用or作業用
|-
| 永安 || HP 2133 mini || 誰か必要な人がいれば貸し出し用
|}

{| border="1"
|+
'''テスト環境'''
| '''所有者''' || '''アーキテクチャ''' || '''OS''' || '''IPアドレス'''
|-
| 永安 || x86_32 (VM) || Red Hat Enterprise Linux 5.4 || （外部リモート）
|-
| 永安 || x86_32 (VM) || CentOS 5.3 || （外部リモート）
|-
| 永安 || x86_32 (VM) || FreeBSD 6.4 || （外部リモート）
|-
| 永安 || x86_32 || MacOS X 10.5.8 || x.x.x.x
|-
| 永安 || x86 || 未定 || （Amazon EC2）
|-
| 原田 || x64 || Windows || Amazon EC2
|-
| 原田 || x86,x64 || Linux(some distros) x N || Amazon EC2
|}

== 参加者一覧 ==

ATNDを参照ください。
http://atnd.org/events/3785

== その他 ==
* TwitterでSFPUGと連携します。
* SFPUGのUstream（http://www.ustream.tv/channel/postgresql---sfpug） [http://twitter.com/#search?q=%23sfpug #sfpug]
* Ustream実施予定。
* Streaming Replication Q&A
* SE-PostgreSQL Q&A
* Foreign Data Wrapper Q&A
* Global Replication Test

テストの方法については、[[HowToBetaTest]]のフォーマットに従うとよいと思います。翻訳求む。

2008-10-31T14:41:38Z

Umitanuki:

__TOC__

This is the page for the CommitFest starting '''2008 November'''.

{{CommitFestOpen}}

== Pending patches ==
{{CommitFestSectionNew}}

{{patch|4863B3DA.8030209@kaigai.gr.jp|SE-PostgreSQL patches|Kaigai Kohei|reviewers=Peter Eisentraut, Abhijit Menon-Sen}}
{{comment|Peter|checking with Solaris engineers about compatibility with Solaris TX; will continue review throughout August}}
{{comment|KaiGai|{{messageLink|488F0C02.4020708@ak.jp.nec.com|latest patch versions}}}}
{{comment|KaiGai|{{messageLink|48BA12F4.1090201@kaigai.gr.jp|latest patch versions}}}}
{{comment|Robert Haas|KaiGai's latest version is {{messageLink|48F46606.4080207@ak.jp.nec.com|r1120}}}}
{{comment|Robert Haas|{{messageLink|48F71E36.9010203@ak.jp.nec.com|latest version}}, now with 6 patches}}
{{comment|KaiGai|[[SEPostgreSQL]] can be a comprehensive documentation (now in progress)}}
{{comment|KaiGai|{{messageLink|490AD10A.6020902@ak.jp.nec.com|latest patches (r1168)}}}}

{{patch|1220213074.4371.173.camel@ebony.2ndQuadrant|Infrastructure changes for recovery|status=Pending Review|Simon Riggs|reviewers=Tom Lane, Heikki Linnakangas}}
{{comment|sriggs| important patch for other work}}
{{comment|tgl|some comments {{messageLink|1905.1220895241@sss.pgh.pa.us|here}}}}
{{comment|tgl|updated patch {{messageLink|1221080797.3913.768.camel@ebony.2ndQuadrant|here}}, still being tested}}
{{comment|tgl|Simon found {{messageLink|1221725145.3913.2315.camel@ebony.2ndQuadrant|some issues}}}}
{{comment|tgl|{{messageLink|1222184541.4445.400.camel@ebony.2ndQuadrant|patch v7}} here}}
{{comment|tgl|it's still got {{messageLink|28973.1222381719@sss.pgh.pa.us|issues}}}}
{{comment|Robert Haas|{{messageLink|1222815151.4445.1397.camel@ebony.2ndQuadrant|patch v8}} here}}
{{comment|Robert Haas|{{messageLink|1223472898.4747.310.camel@ebony.2ndQuadrant|patch v9}} here, in response to {{messageLink|48EC9CC5.6060803@enterprisedb.com|Heikki's review}}}}

{{patch|1220174867.4371.107.camel@ebony.2ndQuadrant|rmgr hooks and contrib/rmgr_hook|status=Waiting on author|Simon Riggs}}
{{comment|sriggs| deferrable, if required}}
{{comment|tgl|I think the plan is for this to wait till "infrastructure" patch goes in}}

{{patch|37ed240d0809030015u73b65b66v70fd76741317c6f5@mail.gmail.com|pg_typeof()|Brendan Jurd}}

{{patch|48C181D4.8030807@gmail.com|reducing statistics write overhead|Martin Pihlak}}
{{comment|Martin Pihlak|Latest patch {{messageLink|48C594D8.6050504@gmail.com|here}}}}

{{patch|BLU110-W38CE0A22D467A8C3D9736CFF540@phx.gbl|Support PLUGINS lines in Makefiles, similar to MODULES|Asif Naeem}}
{{comment|Dave Page|This doesn't work with the edb-debugger plugin, which is the only such plugin around AFAIK. It needs to ignore comments on the PLUGINS line, and handle multiple targets (plugin_debugger, pldbgapi, targetinfo etc). Not sure if we want that complexity though.}}
{{comment|Heikki|Unix-makefile version: {{messageLink|BLU110-W57823CEB7DC113354471EFF3B0@phx.gbl|here}}}}

{{patch|48D15471.6080305@cheapcomplexdevices.com|SQL Standard Interval output and IntervalStyle GUC|Ron Mayer}}
{{comment|Ron Mayer|
An update {{messageLink|48D52E48.406@cheapcomplexdevices.com|here}} fixed
some bugs in GUC setting.
Another update {{messageLink|48E49BDC.4090402@cheapcomplexdevices.com|here}},
brought up-to-date with HEAD}}
{{comment|Ron Mayer|Brought up-to-date with head [http://0ape.com/postgres_interval_patches/ here]. Now allows setting IntervalStyle through an environment variable because pg_regress had set interval styles (as a side-effect of setting DateStyles) through the environment, some minor cleanup.}}

{{patch|48E4A30C.3000503@cheapcomplexdevices.com|ISO 8601 interval literal input and output|Ron Mayer}}
{{comment|Ron Mayer|Note that this patch doesn't apply directly to HEAD, but depends on the {{messageLink|48E49BDC.4090402@cheapcomplexdevices.com|IntervalStyle GUC patch}} to be applied first.}}
{{comment|Ron Mayer|The initial (2003) version of this patch along with a thread which explains the feature a bit more can be found [http://archives.postgresql.org/pgsql-patches/2003-09/msg00103.php here].}}
{{comment|Ron Mayer|An update [http://0ape.com/postgres_interval_patches/ here]. Cleaned up code and added regression tests.}}

{{patch|48CEDE68.7090501@cheapcomplexdevices.com|Interval rounding consistency|Ron Mayer}}
{{comment|Ron Mayer|Patch 3 [http://0ape.com/postgres_interval_patches/ here] refactors the interval code to remove much of the copy&paste in DecodeInterval and EncodeInterval with the side effect of making interval rounding more consistent between styles. This patch applies on top of the IntervalStyle patch and the ISO 8601 Interval patch.}}

{{patch|03c301c91737$be5a2a30$0e01a8c0@IBMC9A0F63B40D|Solve a problem of LC_TIME of windows|Hiroshi Saito}}

{{patch|20080901063855.GJ16005@tamriel.snowman.net|Column-level Permissions|Stephen Frost|status=WIP|reviewers=Markus Wanner}}
{{comment|Stephen Frost|Updated patch {{messageLink|20080902221823.GL16005@tamriel.snowman.net|here}}}}
{{review|48DB48AC.5010400@bluegap.ch|Markus Wanner|Awaiting updated patch.}}

{{patch|0BD2E4E9-AF5A-4B4F-B546-027424EE8FAD@kineticode.com|Tests citext casts|David Wheeler}}

{{patch|48E22CF5.4050503@sun.com|PageGetTempPage cleanup |Zdenek Kotala}}
{{comment|Zdenek Kotala|Original discussion {{messageLink|4938.1217862947@sss.pgh.pa.us|here}}}}

{{patch|603c8f070808070503jd7be083kcce3e16345affb08@mail.gmail.com|add columns via CREATE OR REPLACE VIEW|Robert Haas}}
{{comment|Robert Haas|some further justification of the proposed design {{messageLink|603c8f070808070956t1ab98dcdr933575eb096e4c28@mail.gmail.com|here}}}}
{{comment|Robert Haas|questions about how to move forward {{messageLink|603c8f070810031839y7ce9a852o86effbab9fd6dd9b@mail.gmail.com|here}}}}
{{comment|Robert Haas|a {{messageLink|482096EC.3000805@dunslane.net|similar feature request}} from Andrew Dunstan}}

{{patch|1223379173.4747.145.camel@ebony.2ndQuadrant|Reducing some DDL Locks to ShareLock|status=Pending Review|Simon Riggs}}
{{comment|Robert Haas|{{messageLink|1722552592.7.1224882093461.JavaMail.root@spotone|ddl_lock_reduce.v4.patch}}}}

{{patch|Pine.BSO.4.64.0810081931240.11647@leary.csoft.net|Fixes for psql describeOneTableDetails|Kris Jurka}}

{{patch|1223472186.4747.301.camel@ebony.2ndQuadrant|pg_stop_backup wait bug fix|Simon Riggs}}
{{comment|Robert Haas|sriggs extracted this from recovery infrastructure patch per {{messageLink|48EC9CC5.6060803@enterprisedb.com|Heikki's suggestion}}}}

{{patch|20081009165157.9BE4.52131E4D@oss.ntt.co.jp|contrib/auto_explain|Takahiro Itagaki|}}

{{patch|603c8f070810092009s1312d30bxc854cdfb5d9fed7e@mail.gmail.com|Allow the UUID type to accept non-standard formats|Robert Haas}}

{{patch|603c8f070810101937n776c1e7cvd6a12345b0bbda28@mail.gmail.com|array_ndims|Robert Haas}}

{{patch|48F04B7C.3080303@benedekl.tvnetwork.hu|pg_dump roles support|Benedek Laszlo}}

{{patch|20081011172450.3402.52131E4D@oss.ntt.co.jp|contrib/pg_stat_statements|Takahiro Itagaki}}
{{comment|Itagaki|{{messageLink|20081027171917.ADD6.52131E4D@oss.ntt.co.jp|latest patch versions}}}}

{{patch|49078031.9030002@gmail.com|contrib/pg_stat_staments querydesc|Martin Pihlak}}
{{comment|Martin Pihlak|Updated patch {{messageLink|490A00A8.7050708@gmail.com|here}}}}

{{patch|c2ee6dbd0810100553gd328275ue2eb6e14bee70a8@mail.gmail.com|adding VERBOSE option to CLUSTER|Jim Cox}}

{{patch|e08cc0400810270912u49a6ec83vc23984c01f368f76@mail.gmail.com|Window Functions|Hitoshi Harada|reviewers=Heikki Linnakangas}}
{{comment|Harada|updated patch is {{messageLink|e08cc0400810310721l22a6bb7kb6a300ce66443806@mail.gmail.com|here}}}}

{{patch|48F1FC7E.1060406@sun.com|Extending pg_class info + more flexible TOAST chunk size|Zdenek Kotala}}

{{patch|48EFA13C.2060407@frogthinker.org|Transactions and temp tables|Emmanuel Cecchet|reviewers=Heikki Linnakangas}}

{{patch|20952F54-6730-49D8-99A3-1834697790B5@decibel.org|array_length|Jim C. Nasby}}

{{patch|603c8f070810151933p445978b3td481a5422a2aebf4@mail.gmail.com|array_agg|Robert Haas}}
{{comment|Robert Haas|{{messageLink|27bbfebe0810151633q1aac3ea8v864bea997f3cc5ec@mail.gmail.com|another version}} from Ian Caulfield and {{messageLink|1225045937.4434.21.camel@localhost.localdomain|yet another}} from Jeff Davis}}

{{patch|490878AC.1@dunslane.net|parallel restore|status=WIP|Andrew Dunstan}}

{{patch|48F53EC0.3020004@timbira.com|autovacuum and reloption|status=WIP|Euler Taveira de Oliveira|reviewers=Alvaro Herrera}}

{{patch|162867790810170316l4eeecb0bq321dd771f8f4e661@mail.gmail.com|grouping sets|status=WIP|Pavel Stehule}}

{{patch|6EEA43D22289484890D119821101B1DF2C1683@exchange20.mercury.ad.ubc.ca|Improve Performance of Multi-Batch Hash Join for Skewed Data Sets|Ramon Lawrence/Bryce Cutt}}

{{patch|48FC7E84.3080702@hagander.net|SSL cleanups/hostname verification|Magnus Hagander}}

{{patch|49009D67.4040202@hagander.net|clientcert option for pg_hba|Magnus Hagander}}

{{patch|e4ccc24e0810222010p12bae2f4xa3a11cb2bc51bd89@mail.gmail.com|pg_standby support for compressed segments|Charles Duffy}}

{{patch|162867790810260428p56d16352qa1ec5c4c5330f25c@mail.gmail.com|default values for function's parameters|Pavel Stehule}}
{{comment|Pavel Stehule| fix known bugs, currently only doc should be finished}}

{{patch|490394B7.6090503@lelarge.info|ALTER DATABASE WITH TABLESPACE Statement|Guillaume Lelarge}}

{{patch|Pine.BSO.4.64.0810271955030.28764@leary.csoft.net|use new heap_(form/deform/modify)_tuple API|Kris Jurka|reviewers=Zdenek Kotala|status=Ready for commit}}
{{comment|Zdenek Kotala| It seems OK. Needs apply also {{messageLink|Pine.BSO.4.64.0810281721500.6833@leary.csoft.net|pfree patch.}}}}

{{patch|49076F39.1080502@hagander.net|regexp support in usermaps|Magnus Hagander}}

{{patch|Pine.BSO.4.64.0810281622240.9851@leary.csoft.net|don't use MAKE_PTR/OFFSET for shmem pointers|Kris Jurka}}

{{patch|603c8f070810282045g7fa6bdf9p53f03114d49643d7@mail.gmail.com|bulk inserts - keep most recent page pinned|Robert Haas}}
{{comment|Robert Haas|based on work by Simon Riggs: [http://archives.postgresql.org/pgsql-hackers/2008-02/msg01080.php original idea],[http://archives.postgresql.org/pgsql-patches/2008-02/msg00140.php original patch], [http://archives.postgresql.org/pgsql-patches/2008-02/msg00154.php Tom Lane's comments]}}
{{comment|Robert Haas|Tom's comments on my [http://archives.postgresql.org/pgsql-hackers/2008-10/msg01276.php design proposal] are {{messageLink|17996.1225049097@sss.pgh.pa.us|here}}}}

{{patch|49086E4C.9030602@sun.com|htup and bufpage API clean up|Zdenek Kotala}}
{{comment|Zdenek Kotala|New version is {{messageLink|490875FA.4020709@sun.com|here}}}}

{{patch|49097338.2070700@gmail.com|SQL/MED compatible connection manager|Martin Pihlak}}

{{patch|20081029164000.GL27466@fetter.org|pre-MED|David Fetter}}

{{patch|4905AE17.7090305@enterprisedb.com|Visibility map, partial vacuums|status=WIP|Heikki Linnakangas}}

{{patch|49085327.5010608@enterprisedb.com|Updating FSM on recovery|Heikki Linnakangas}}

{{patch|1225010283.21241.33.camel@localhost.localdomain|new correlation metric|Jeff Davis}}

{{patch|20081029183248.GE4331@alvh.no-ip.org|Block-level CRC checks|Alvaro Herrera|status=WIP}}

{{patch|4909DFCE.1000702@sun.com|HeapTuple version extension + code cleanup|Zdenek Kotala}}

{{patch|4909B326.507@enterprisedb.com|Optimizing COPY with memchr()|Heikki Linnakangas}}

{{patch|87ljw5c0lx.fsf@oxford.xeocode.com|posix_fadvise|Gregory Stark}}

{{patch|4909726F.8000800@gmx.net|TABLE command|Peter Eisentraut}}

{{patch|200810271936.m9RJaGW09544@momjian.us|libpq callback unregistration|Bruce Momjian}}

{{patch|f205bb120810280833x340324d3m19a4f8aa65d822b@mail.gmail.com|FAQ_Solaris 1.28 to spanish|Emanuel CALVO FRANCO}}

{{patch|3f0b79eb0810310436w360f0afdy76ff1499b177ce0d@mail.gmail.com|Synchronous log-shipping replication|Masao Fujii|reviewers=Heikki Linnakangas}}

{{patch|490B07BA.6030109@sigaev.ru|GIN fast insert|Teodor Sigaev, Oleg Bartunov}}

{{CommitFestEndSection}}

== Committed patches ==
{{CommitFestSectionNew}}

{{patch|20080803.114711.73374027.t-ishii@sraoss.co.jp|Common Table Expressions|Yoshiyuki Asaba|status=Committed 2008-10-04|reviewers=Jeff Davis}}
{{comment|David Fetter|README is {{messageLink|20080818.163852.105172778.t-ishii@sraoss.co.jp|here.}}}}
{{comment|David Fetter|Git repository is [http://git.postgresql.org/git/~davidfetter/cte/.git here.]}}
{{comment|David Fetter|[[CTEReadme]]}}
{{comment|David Fetter|[[CTESQLStandard| Fair Use from the draft SQL:2008 standard (WIP)]]}}
{{review|13449.1221589943@sss.pgh.pa.us|tgl|needs a good bit of work yet}}

{{patch|Pine.GSO.4.64.0809152349440.19910@westnet.com|Add default_val to pg_settings|status=Committed 2008-10-06|Greg Smith|reviewers=Simon Riggs,Magnus Hagander}}

{{patch|20081016102603.897D.52131E4D@oss.ntt.co.jp|Noisy _dosmaperror|status=Committed 2008-10-16|Takahiro Itagaki|reviewers=Tom Lane}}

{{patch|b4e5ce320810151918g2acf69ebh9f80f4f2e1c203a0@mail.gmail.com|Memory leak on hashed agg rescan|status=Committed 2008-10-16|Neil Conway|reviewers=Tom Lane}}
{{review|10455.1224160007@sss.pgh.pa.us|Tom Lane|move some logic to separate function}}

{{patch|1223383838.4747.154.camel@ebony.2ndQuadrant|Atomic subtransaction commit|status=Committed 2008-10-20|Simon Riggs|reviewers=Alvaro Herrera}}

{{patch|3327.1224535092@sss.pgh.pa.us|add placeholder variables to planner|status=Committed 2008-10-22|Tom Lane}}

{{patch|48E271C5.7010907@hagander.net|pg_hba options parsing|status=Committed 2008-10-23|Magnus Hagander|reviewers=Bruce Momjian}}
{{comment|Robert Haas|{{messageLink|48F0DE8B.8030004@hagander.net|updated patch}}}}

{{patch|48FC3994.8040809@hagander.net|libpq ssl -> clear fallback looses error messages|status=Committed 2008-10-27|Magnus Hagander}}

{{patch|4905D22D.9040001@hagander.net|better hba parsing error messages|status=Committed 2008-10-27|Magnus Hagander}}

{{patch|4905A1DE.5030102@hagander.net|remove crypt authentication|status=Committed 2008-10-28|Magnus Hagander}}

{{patch|4905F515.3020208@gmx.net|Unicode escapes in literals|status=Committed 2008-10-29|Peter Eisentraut}}

{{patch|490A138E.80100@enterprisedb.com|User defined I/O conversion casts|status=Committed 2008-10-31|Heikki Linnakangas}}

{{CommitFestEndSection}}

== Returned with Feedback ==
{{CommitFestSectionNew}}

{{CommitFestEndSection}}