Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content

Commit cb25bf7

Browse files
bmomjianbmomjian
committed
Re-add release notes for release 8.3.7.
1 parent 32e7174 commit cb25bf7

File tree

1 file changed

+224
-1
lines changed

1 file changed

+224
-1
lines changed

‎doc/src/sgml/release.sgml

+224-1
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
<!-- $PostgreSQL: pgsql/doc/src/sgml/release.sgml,v 1.594 2009/03/26 01:48:27 momjian Exp $ -->
1+
<!-- $PostgreSQL: pgsql/doc/src/sgml/release.sgml,v 1.595 2009/03/26 02:40:37 momjian Exp $ -->
22
<!--
33

44
Typical markup:
@@ -2620,6 +2620,229 @@ do it for earlier branch release files.
26202620
</sect2>
26212621
</sect1>
26222622

2623+
<sect1 id="release-8-3-7">
2624+
<title>Release 8.3.7</title>
2625+
2626+
<note>
2627+
<title>Release date</title>
2628+
<simpara>2009-03-16</simpara>
2629+
</note>
2630+
2631+
<para>
2632+
This release contains a variety of fixes from 8.3.6.
2633+
For information about new features in the 8.3 major release, see
2634+
<xref linkend="release-8-3">.
2635+
</para>
2636+
2637+
<sect2>
2638+
<title>Migration to Version 8.3.7</title>
2639+
2640+
<para>
2641+
A dump/restore is not required for those running 8.3.X.
2642+
However, if you are upgrading from a version earlier than 8.3.5,
2643+
see the release notes for 8.3.5.
2644+
</para>
2645+
2646+
</sect2>
2647+
2648+
<sect2>
2649+
<title>Changes</title>
2650+
2651+
<itemizedlist>
2652+
2653+
<listitem>
2654+
<para>
2655+
Prevent error recursion crashes when encoding conversion fails (Tom)
2656+
</para>
2657+
2658+
<para>
2659+
This change extends fixes made in the last two minor releases for
2660+
related failure scenarios. The previous fixes were narrowly tailored
2661+
for the original problem reports, but we have now recognized that
2662+
<emphasis>any</> error thrown by an encoding conversion function could
2663+
potentially lead to infinite recursion while trying to report the
2664+
error. The solution therefore is to disable translation and encoding
2665+
conversion and report the plain-ASCII form of any error message,
2666+
if we find we have gotten into a recursive error reporting situation.
2667+
(CVE-2009-0922)
2668+
</para>
2669+
</listitem>
2670+
2671+
<listitem>
2672+
<para>
2673+
Disallow <command>CREATE CONVERSION</> with the wrong encodings
2674+
for the specified conversion function (Heikki)
2675+
</para>
2676+
2677+
<para>
2678+
This prevents one possible scenario for encoding conversion failure.
2679+
The previous change is a backstop to guard against other kinds of
2680+
failures in the same area.
2681+
</para>
2682+
</listitem>
2683+
2684+
<listitem>
2685+
<para>
2686+
Fix <function>xpath()</> to not modify the path expression unless
2687+
necessary, and to make a saner attempt at it when necessary (Andrew)
2688+
</para>
2689+
2690+
<para>
2691+
The SQL standard suggests that <function>xpath</> should work on data
2692+
that is a document fragment, but <application>libxml</> doesn't support
2693+
that, and indeed it's not clear that this is sensible according to the
2694+
XPath standard. <function>xpath</> attempted to work around this
2695+
mismatch by modifying both the data and the path expression, but the
2696+
modification was buggy and could cause valid searches to fail. Now,
2697+
<function>xpath</> checks whether the data is in fact a well-formed
2698+
document, and if so invokes <application>libxml</> with no change to the
2699+
data or path expression. Otherwise, a different modification method
2700+
that is somewhat less likely to fail is used.
2701+
</para>
2702+
2703+
<note>
2704+
<para>
2705+
The new modification method is still not 100% satisfactory, and it
2706+
seems likely that no real solution is possible. This patch should
2707+
therefore be viewed as a band-aid to keep from breaking existing
2708+
applications unnecessarily. It is likely that
2709+
<productname>PostgreSQL</> 8.4 will simply reject use of
2710+
<function>xpath</> on data that is not a well-formed document.
2711+
</para>
2712+
</note>
2713+
</listitem>
2714+
2715+
<listitem>
2716+
<para>
2717+
Fix core dump when <function>to_char()</> is given format codes that
2718+
are inappropriate for the type of the data argument (Tom)
2719+
</para>
2720+
</listitem>
2721+
2722+
<listitem>
2723+
<para>
2724+
Fix possible failure in text search when C locale is used with
2725+
a multi-byte encoding (Teodor)
2726+
</para>
2727+
2728+
<para>
2729+
Crashes were possible on platforms where <type>wchar_t</> is narrower
2730+
than <type>int</>; Windows in particular.
2731+
</para>
2732+
</listitem>
2733+
2734+
<listitem>
2735+
<para>
2736+
Fix extreme inefficiency in text search parser's handling of an
2737+
email-like string containing multiple <literal>@</> characters (Heikki)
2738+
</para>
2739+
</listitem>
2740+
2741+
<listitem>
2742+
<para>
2743+
Fix planner problem with sub-<command>SELECT</> in the output list
2744+
of a larger subquery (Tom)
2745+
</para>
2746+
2747+
<para>
2748+
The known symptom of this bug is a <quote>failed to locate grouping
2749+
columns</> error that is dependent on the datatype involved;
2750+
but there could be other issues as well.
2751+
</para>
2752+
</listitem>
2753+
2754+
<listitem>
2755+
<para>
2756+
Fix decompilation of <literal>CASE WHEN</> with an implicit coercion
2757+
(Tom)
2758+
</para>
2759+
2760+
<para>
2761+
This mistake could lead to Assert failures in an Assert-enabled build,
2762+
or an <quote>unexpected CASE WHEN clause</> error message in other
2763+
cases, when trying to examine or dump a view.
2764+
</para>
2765+
</listitem>
2766+
2767+
<listitem>
2768+
<para>
2769+
Fix possible misassignment of the owner of a TOAST table's rowtype (Tom)
2770+
</para>
2771+
2772+
<para>
2773+
If <command>CLUSTER</> or a rewriting variant of <command>ALTER TABLE</>
2774+
were executed by someone other than the table owner, the
2775+
<structname>pg_type</> entry for the table's TOAST table would end up
2776+
marked as owned by that someone. This caused no immediate problems,
2777+
since the permissions on the TOAST rowtype aren't examined by any
2778+
ordinary database operation. However, it could lead to unexpected
2779+
failures if one later tried to drop the role that issued the command
2780+
(in 8.1 or 8.2), or <quote>owner of data type appears to be invalid</>
2781+
warnings from <application>pg_dump</> after having done so (in 8.3).
2782+
</para>
2783+
</listitem>
2784+
2785+
<listitem>
2786+
<para>
2787+
Change <command>UNLISTEN</> to exit quickly if the current session has
2788+
never executed any <command>LISTEN</> command (Tom)
2789+
</para>
2790+
2791+
<para>
2792+
Most of the time this is not a particularly useful optimization, but
2793+
since <command>DISCARD ALL</> invokes <command>UNLISTEN</>, the previous
2794+
coding caused a substantial performance problem for applications that
2795+
made heavy use of <command>DISCARD ALL</>.
2796+
</para>
2797+
</listitem>
2798+
2799+
<listitem>
2800+
<para>
2801+
Fix PL/pgSQL to not treat <literal>INTO</> after <command>INSERT</> as
2802+
an INTO-variables clause anywhere in the string, not only at the start;
2803+
in particular, don't fail for <command>INSERT INTO</> within
2804+
<command>CREATE RULE</> (Tom)
2805+
</para>
2806+
</listitem>
2807+
2808+
<listitem>
2809+
<para>
2810+
Clean up PL/pgSQL error status variables fully at block exit
2811+
(Ashesh Vashi and Dave Page)
2812+
</para>
2813+
2814+
<para>
2815+
This is not a problem for PL/pgSQL itself, but the omission could cause
2816+
the PL/pgSQL Debugger to crash while examining the state of a function.
2817+
</para>
2818+
</listitem>
2819+
2820+
<listitem>
2821+
<para>
2822+
Retry failed calls to <function>CallNamedPipe()</> on Windows
2823+
(Steve Marshall, Magnus)
2824+
</para>
2825+
2826+
<para>
2827+
It appears that this function can sometimes fail transiently;
2828+
we previously treated any failure as a hard error, which could
2829+
confuse <command>LISTEN</>/<command>NOTIFY</> as well as other
2830+
operations.
2831+
</para>
2832+
</listitem>
2833+
2834+
<listitem>
2835+
<para>
2836+
Add <literal>MUST</> (Mauritius Island Summer Time) to the default list
2837+
of known timezone abbreviations (Xavier Bugaud)
2838+
</para>
2839+
</listitem>
2840+
2841+
</itemizedlist>
2842+
2843+
</sect2>
2844+
</sect1>
2845+
26232846
<sect1 id="release-8-3-6">
26242847
<title>Release 8.3.6</title>
26252848

0 commit comments

Comments
 (0)