Add index information to /contrib/pgstattuple:

This is an extension of pgstattuple to query information from indexes.
It supports btree, hash and gist. Gin is not supported. It scans only
index pages and does not read corresponding heap tuples. Therefore,
'dead_tuple' means the number of tuples with LP_DELETE flag.

Also, I added an experimental feature for btree indexes. It checks
fragmentation factor of indexes. If an leaf has the right link on the
next adjacent page in the file, it is assumed to be continuous (not
fragmented). It will help us to decide when to REINDEX.

ITAGAKI Takahiro

contrib/pgstattuple/README.pgstattuple

@@ -2,8 +2,8 @@ pgstattuple README			2002/08/29 Tatsuo Ishii
 
 1. What is pgstattuple?
 
-   pgstattuple returns the table length, percentage of the "dead"
-   tuples of a table and other info. This may help users to determine
+   pgstattuple returns the relation length, percentage of the "dead"
+   tuples of a relation and other info. This may help users to determine
    whether vacuum is necessary or not. Here is an example session:
 
 test=# \x
@@ -23,7 +23,7 @@ free_percent       | 1.95
 
 Here are explanations for each column:
 
-table_len		-- physical table length in bytes
+table_len		-- physical relation length in bytes
 tuple_count		-- number of live tuples
 tuple_len		-- total tuples length in bytes
 tuple_percent		-- live tuples in %
@@ -40,7 +40,7 @@ free_percent		-- free space in %
 
 3. Using pgstattuple
 
-   pgstattuple may be called as a table function and is
+   pgstattuple may be called as a relation function and is
    defined as follows:
 
    CREATE OR REPLACE FUNCTION pgstattuple(text) RETURNS pgstattuple_type
@@ -51,14 +51,20 @@ free_percent		-- free space in %
      AS 'MODULE_PATHNAME', 'pgstattuplebyid'
      LANGUAGE C STRICT;
 
-   The argument is the table name (optionally it may be qualified)
-   or the OID of the table.  Note that pgstattuple only returns
+   The argument is the relation name (optionally it may be qualified)
+   or the OID of the relation.  Note that pgstattuple only returns
    one row.
 
 4. Notes
 
-   pgstattuple acquires only a read lock on the table. So concurrent
+   pgstattuple acquires only a read lock on the relation. So concurrent
    update may affect the result.
 
    pgstattuple judges a tuple is "dead" if HeapTupleSatisfiesNow()
    returns false.
+
+5. History
+
+   2006/06/28
+
+	Extended to work against indexes.

contrib/pgstattuple/README.pgstattuple.euc_jp

-$PostgreSQL: pgsql/contrib/pgstattuple/README.pgstattuple.euc_jp,v 1.6 2006/02/27 16:09:49 petere Exp $
+$PostgreSQL: pgsql/contrib/pgstattuple/README.pgstattuple.euc_jp,v 1.7 2006/07/06 02:18:07 momjian Exp $
 
 pgstattuple README		2002/08/22 石井達夫
 
 1. pgstattupleとは
 
-   pgstattupleは，UPDATEやDELETEで作られたテーブルのゴミ領域の大きさを，
-   テーブル自体の物理的な大きさに対するパーセンテージで返却します．つ
+   pgstattupleは，UPDATEやDELETEで作られたリレーションのゴミ領域の大きさを，
+   リレーション自体の物理的な大きさに対するパーセンテージで返却します．つ
    まり，返却値が大きければ，それだけゴミも多いので，vacuumをかける必
    要があるという判断の助けになるわけです．これ以外にもいろいろな情報
    が返ります．
@@ -26,7 +26,7 @@ free_percent       | 1.95
 
 各項目の説明です．
 
-table_len		-- テーブルの物理的な大きさ(バイト)
+table_len		-- リレーションの物理的な大きさ(バイト)
 tuple_count		-- タプル数
 tuple_len		-- タプル長の合計(バイト)
 tuple_percent		-- タプルの割合．table_lenに対するtuple_lenの比率．
@@ -56,12 +56,12 @@ free_percent		--
      AS 'MODULE_PATHNAME', 'pgstattuple'
      LANGUAGE C STRICT;
 
-   第一引数: テーブル名
+   第一引数: リレーション名
 
    関数の戻りはpgstattuple_type型です．
 
-   pgstattupleはテーブルにAccessShareLockしかかけないので，
-   pgstattuple を実行中に該当テーブルに更新や削除が発生すると，正しく
+   pgstattupleはリレーションにAccessShareLockしかかけないので，
+   pgstattuple を実行中に該当リレーションに更新や削除が発生すると，正しく
    ない結果を返す可能性があります．
 
    pgstattupleがタプルを「ゴミ」と判断する基準は，
@@ -75,6 +75,10 @@ free_percent		--
 
 5. 改訂履歴
 
+   2006/06/28
+
+	インデックスに対しても動作するように拡張。
+
    2002/09/04
 
 	SRF変更に伴い，Tom Lane	が修正インターフェイスの修正を行った．