pgsql/src backend/commands/explain.c backend/n ... 
Author Message
 pgsql/src backend/commands/explain.c backend/n ...

CVSROOT:        /cvsroot
Module name:    pgsql

Modified files:
        src/backend/commands: explain.c
        src/backend/nodes: copyfuncs.c equalfuncs.c outfuncs.c
                           readfuncs.c
        src/backend/optimizer/path: costsize.c joinpath.c joinrels.c
                                    pathkeys.c
        src/backend/optimizer/plan: createplan.c initsplan.c planmain.c
                                    planner.c setrefs.c
        src/backend/optimizer/prep: prepunion.c
        src/backend/optimizer/util: clauses.c relnode.c var.c
        src/backend/parser: analyze.c parse_clause.c parse_expr.c
                            parse_func.c parse_node.c parse_relation.c
                            parse_target.c
        src/backend/rewrite: rewriteHandler.c rewriteManip.c
        src/backend/utils/adt: ruleutils.c
        src/include/catalog: catversion.h
        src/include/nodes: parsenodes.h plannodes.h primnodes.h
                           relation.h
        src/include/optimizer: pathnode.h planmain.h var.h
        src/include/parser: parse_relation.h parsetree.h
        src/include/utils: builtins.h
        src/test/regress/expected: join.out
        src/test/regress/sql: join.sql

Log message:
        Restructure representation of join alias variables.  An explicit JOIN
        now has an RTE of its own, and references to its outputs now are Vars
        referencing the JOIN RTE, rather than CASE-expressions.  This allows
        reverse-listing in ruleutils.c to use the correct alias easily, rather
        than painfully reverse-engineering the alias namespace as it used to do.
        Also, nested FULL JOINs work correctly, because the result of the inner
        joins are simple Vars that the planner can cope with.  This fixes a bug
        reported a couple times now, notably by Tatsuo on 18-Nov-01.  The alias
        Vars are expanded into COALESCE expressions where needed at the very end
        of planning, rather than during parsing.
        Also, beginnings of support for showing plan qualifier expressions in
        EXPLAIN.  There are probably still cases that need work.
        initdb forced due to change of stored-rule representation.

---------------------------(end of broadcast)---------------------------
TIP 5: Have you checked our extensive FAQ?

http://www.***.com/



Sat, 28 Aug 2004 08:52:29 GMT
 
 [ 1 post ] 

 Relevant Pages 

1. pgsql/src backend/commands/explain.c backend/e ...

2. pgsql/src/backend/commands (explain.c command.c copy.c)

3. pgsql/src backend/commands/command.c backend/c ...

4. pgsql/src backend/commands/command.c backend/u ...

5. pgsql/src backend/commands/command.c backend/c ...

6. pgsql/src backend/commands/command.c backend/p ...

7. pgsql/src backend/commands/command.c backend/e ...

8. pgsql/src backend/commands/command.c backend/c ...

9. pgsql/src backend/commands/command.c backend/p ...

10. pgsql/src backend/commands/command.c backend/e ...

11. pgsql/src/backend commands/explain.c utils/mis ...

12. pgsql-server/src backend/commands/explain.c ba ...


 
Powered by phpBB® Forum Software