-
Repair sometimes-incorrect computation of StartUpID after a crash
-
Avoid slowness with lots of deferred triggers in one transaction (Stephan)
-
Don't lock referenced row when UPDATE doesn't change foreign key's value (Jan)
-
Use -fPIC not -fpic on Sparc (Tom Callaway)
-
Repair lack of schema-awareness in contrib/reindexdb
-
Fix contrib/intarray error for zero-element result array (Teodor)
-
Ensure createuser script will exit on control-C (Oliver)
-
Fix errors when the type of a dropped column has itself been dropped
-
CHECKPOINT does not cause database panic on failure in noncritical steps
-
Accept 60 in seconds fields of timestamp, time, interval input values
-
Issue notice, not error, if TIMESTAMP, TIME, or INTERVAL precision too large
-
Fix abstime-to-time
cast function (fix is not applied unless you initdb)
-
Fix pg_proc entry for timestampt_izone (fix is not applied unless you initdb)
-
Make EXTRACT(EPOCH FROM timestamp without time zone)
treat input as local time
-
'now'::timestamptz gave wrong answer if timezone changed earlier in transaction
-
HAVE_INT64_TIMESTAMP code for time with timezone overwrote its input
-
Accept GLOBAL TEMP/TEMPORARY as a synonym for TEMPORARY
-
Avoid improper schema-privilege-check failure in foreign-key triggers
-
Fix bugs in foreign-key triggers for SET DEFAULT action
-
Fix incorrect time-qual check in row fetch for UPDATE and DELETE triggers
-
Foreign-key clauses were parsed but ignored in ALTER TABLE ADD COLUMN
-
Fix createlang script breakage for case where handler function already exists
-
Fix misbehavior on zero-column tables in pg_dump, COPY, ANALYZE, other places
-
Fix misbehavior of func_error()
on type names containing '%'
-
Fix misbehavior of replace()
on strings containing '%'
-
Regular-expression patterns containing certain multibyte characters failed
-
Account correctly for NULLs in more cases in join size estimation
-
Avoid conflict with system definition of isblank()
function or macro
-
Fix failure to convert large code point values in EUC_TW conversions (Tatsuo)
-
Fix error recovery for SSL_read
/SSL_write
calls
-
Don't do early constant-folding of type coercion expressions
-
Validate page header fields immediately after reading in any page
-
Repair incorrect check for ungrouped variables in unnamed joins
-
Fix buffer overrun in to_ascii
(Guido Notari)
-
contrib/ltree fixes (Teodor)
-
Fix core dump in deadlock detection on machines where char is unsigned
-
Avoid running out of buffers in many-way indexscan (bug introduced in 7.3)
-
Fix planner's selectivity estimation functions to handle domains properly
-
Fix dbmirror memory-allocation bug (Steven Singer)
-
Prevent infinite loop in ln(numeric)
due to roundoff error
-
GROUP BY got confused if there were multiple equal GROUP BY items
-
Fix bad plan when inherited UPDATE/DELETE references another inherited table
-
Prevent clustering on incomplete (partial or non-NULL-storing) indexes
-
Service shutdown request at proper time if it arrives while still starting up
-
Fix left-links in temporary indexes (could make backwards scans miss entries)
-
Fix incorrect handling of client_encoding setting in postgresql.conf (Tatsuo)
-
Fix failure to respond to pg_ctl stop -m fast after Async_NotifyHandler runs
-
Fix SPI for case where rule contains multiple statements of the same type
-
Fix problem with checking for wrong type of access privilege in rule query
-
Fix problem with EXCEPT in CREATE RULE
-
Prevent problem with dropping temp tables having serial columns
-
Fix replace_vars_with_subplan_refs failure in complex views
-
Fix regexp slowness in single-byte encodings (Tatsuo)
-
Allow qualified type names in CREATE CAST and DROP CAST
-
Accept SETOF type[]
, which formerly had to be written SETOF _type
-
Fix pg_dump core dump in some cases with procedural languages
-
Force ISO datestyle in pg_dump output, for portability (Oliver)
-
pg_dump failed to handle error return from lo_read
(Oleg Drokin)
-
pg_dumpall failed with groups having no members (Nick Eskelinen)
-
pg_dumpall failed to recognize --globals-only switch
-
pg_restore failed to restore blobs if -X disable-triggers is specified
-
Repair intrafunction memory leak in plpgsql
-
pltcl's elog command dumped core if given wrong parameters (Ian Harding)
-
plpython used wrong value of atttypmod (Brad McLean)
-
Fix improper quoting of boolean values in Python interface (D'Arcy)
-
Added addDataType()
method to PGConnection interface for JDBC
-
Fixed various problems with updateable ResultSets for JDBC (Shawn Green)
-
Fixed various problems with DatabaseMetaData for JDBC (Kris Jurka, Peter Royal)
-
Fixed problem with parsing table ACLs in JDBC
-
Better error message for character set conversion problems in JDBC