|
1 | 1 | <!-- doc/src/sgml/release-8.2.sgml -->
|
2 | 2 | <!-- See header comment in release.sgml about typical markup -->
|
3 | 3 |
|
| 4 | + <sect1 id="release-8-2-23"> |
| 5 | + <title>Release 8.2.23</title> |
| 6 | + |
| 7 | + <note> |
| 8 | + <title>Release Date</title> |
| 9 | + <simpara>2011-12-05</simpara> |
| 10 | + </note> |
| 11 | + |
| 12 | + <para> |
| 13 | + This release contains a variety of fixes from 8.2.22. |
| 14 | + For information about new features in the 8.2 major release, see |
| 15 | + <xref linkend="release-8-2">. |
| 16 | + </para> |
| 17 | + |
| 18 | + <para> |
| 19 | + This is expected to be the last <productname>PostgreSQL</> release |
| 20 | + in the 8.2.X series. Users are encouraged to update to a newer |
| 21 | + release branch soon. |
| 22 | + </para> |
| 23 | + |
| 24 | + <sect2> |
| 25 | + <title>Migration to Version 8.2.23</title> |
| 26 | + |
| 27 | + <para> |
| 28 | + A dump/restore is not required for those running 8.2.X. |
| 29 | + </para> |
| 30 | + |
| 31 | + <para> |
| 32 | + However, a longstanding error was discovered in the definition of the |
| 33 | + <literal>information_schema.referential_constraints</> view. If you |
| 34 | + rely on correct results from that view, you should replace its |
| 35 | + definition as explained in the first changelog item below. |
| 36 | + </para> |
| 37 | + |
| 38 | + <para> |
| 39 | + Also, if you are upgrading from a version earlier than 8.2.14, |
| 40 | + see the release notes for 8.2.14. |
| 41 | + </para> |
| 42 | + |
| 43 | + </sect2> |
| 44 | + |
| 45 | + <sect2> |
| 46 | + <title>Changes</title> |
| 47 | + |
| 48 | + <itemizedlist> |
| 49 | + |
| 50 | + <listitem> |
| 51 | + <para> |
| 52 | + Fix bugs in <literal>information_schema.referential_constraints</> view |
| 53 | + (Tom Lane) |
| 54 | + </para> |
| 55 | + |
| 56 | + <para> |
| 57 | + This view was being insufficiently careful about matching the |
| 58 | + foreign-key constraint to the depended-on primary or unique key |
| 59 | + constraint. That could result in failure to show a foreign key |
| 60 | + constraint at all, or showing it multiple times, or claiming that it |
| 61 | + depends on a different constraint than the one it really does. |
| 62 | + </para> |
| 63 | + |
| 64 | + <para> |
| 65 | + Since the view definition is installed by <application>initdb</>, |
| 66 | + merely upgrading will not fix the problem. If you need to fix this |
| 67 | + in an existing installation, you can (as a superuser) drop the |
| 68 | + <literal>information_schema</> schema then re-create it by sourcing |
| 69 | + <filename><replaceable>SHAREDIR</>/information_schema.sql</filename>. |
| 70 | + (Run <literal>pg_config --sharedir</> if you're uncertain where |
| 71 | + <replaceable>SHAREDIR</> is.) This must be repeated in each database |
| 72 | + to be fixed. |
| 73 | + </para> |
| 74 | + </listitem> |
| 75 | + |
| 76 | + <listitem> |
| 77 | + <para> |
| 78 | + Fix TOAST-related data corruption during <literal>CREATE TABLE dest AS |
| 79 | + SELECT * FROM src</> or <literal>INSERT INTO dest SELECT * FROM src</> |
| 80 | + (Tom Lane) |
| 81 | + </para> |
| 82 | + |
| 83 | + <para> |
| 84 | + If a table has been modified by <command>ALTER TABLE ADD COLUMN</>, |
| 85 | + attempts to copy its data verbatim to another table could produce |
| 86 | + corrupt results in certain corner cases. |
| 87 | + The problem can only manifest in this precise form in 8.4 and later, |
| 88 | + but we patched earlier versions as well in case there are other code |
| 89 | + paths that could trigger the same bug. |
| 90 | + </para> |
| 91 | + </listitem> |
| 92 | + |
| 93 | + <listitem> |
| 94 | + <para> |
| 95 | + Fix race condition during toast table access from stale syscache entries |
| 96 | + (Tom Lane) |
| 97 | + </para> |
| 98 | + |
| 99 | + <para> |
| 100 | + The typical symptom was transient errors like <quote>missing chunk |
| 101 | + number 0 for toast value NNNNN in pg_toast_2619</>, where the cited |
| 102 | + toast table would always belong to a system catalog. |
| 103 | + </para> |
| 104 | + </listitem> |
| 105 | + |
| 106 | + <listitem> |
| 107 | + <para> |
| 108 | + Improve locale support in <type>money</> type's input and output |
| 109 | + (Tom Lane) |
| 110 | + </para> |
| 111 | + |
| 112 | + <para> |
| 113 | + Aside from not supporting all standard |
| 114 | + <link linkend="guc-lc-monetary"><varname>lc_monetary</></link> |
| 115 | + formatting options, the input and output functions were inconsistent, |
| 116 | + meaning there were locales in which dumped <type>money</> values could |
| 117 | + not be re-read. |
| 118 | + </para> |
| 119 | + </listitem> |
| 120 | + |
| 121 | + <listitem> |
| 122 | + <para> |
| 123 | + Don't let <link |
| 124 | + linkend="guc-transform-null-equals"><varname>transform_null_equals</></link> |
| 125 | + affect <literal>CASE foo WHEN NULL ...</> constructs |
| 126 | + (Heikki Linnakangas) |
| 127 | + </para> |
| 128 | + |
| 129 | + <para> |
| 130 | + <varname>transform_null_equals</> is only supposed to affect |
| 131 | + <literal>foo = NULL</> expressions written directly by the user, not |
| 132 | + equality checks generated internally by this form of <literal>CASE</>. |
| 133 | + </para> |
| 134 | + </listitem> |
| 135 | + |
| 136 | + <listitem> |
| 137 | + <para> |
| 138 | + Change foreign-key trigger creation order to better support |
| 139 | + self-referential foreign keys (Tom Lane) |
| 140 | + </para> |
| 141 | + |
| 142 | + <para> |
| 143 | + For a cascading foreign key that references its own table, a row update |
| 144 | + will fire both the <literal>ON UPDATE</> trigger and the |
| 145 | + <literal>CHECK</> trigger as one event. The <literal>ON UPDATE</> |
| 146 | + trigger must execute first, else the <literal>CHECK</> will check a |
| 147 | + non-final state of the row and possibly throw an inappropriate error. |
| 148 | + However, the firing order of these triggers is determined by their |
| 149 | + names, which generally sort in creation order since the triggers have |
| 150 | + auto-generated names following the convention |
| 151 | + <quote>RI_ConstraintTrigger_NNNN</>. A proper fix would require |
| 152 | + modifying that convention, which we will do in 9.2, but it seems risky |
| 153 | + to change it in existing releases. So this patch just changes the |
| 154 | + creation order of the triggers. Users encountering this type of error |
| 155 | + should drop and re-create the foreign key constraint to get its |
| 156 | + triggers into the right order. |
| 157 | + </para> |
| 158 | + </listitem> |
| 159 | + |
| 160 | + <listitem> |
| 161 | + <para> |
| 162 | + Preserve blank lines within commands in <application>psql</>'s command |
| 163 | + history (Robert Haas) |
| 164 | + </para> |
| 165 | + |
| 166 | + <para> |
| 167 | + The former behavior could cause problems if an empty line was removed |
| 168 | + from within a string literal, for example. |
| 169 | + </para> |
| 170 | + </listitem> |
| 171 | + |
| 172 | + <listitem> |
| 173 | + <para> |
| 174 | + Use the preferred version of <application>xsubpp</> to build PL/Perl, |
| 175 | + not necessarily the operating system's main copy |
| 176 | + (David Wheeler and Alex Hunsaker) |
| 177 | + </para> |
| 178 | + </listitem> |
| 179 | + |
| 180 | + <listitem> |
| 181 | + <para> |
| 182 | + Honor query cancel interrupts promptly in <function>pgstatindex()</> |
| 183 | + (Robert Haas) |
| 184 | + </para> |
| 185 | + </listitem> |
| 186 | + |
| 187 | + <listitem> |
| 188 | + <para> |
| 189 | + Ensure VPATH builds properly install all server header files |
| 190 | + (Peter Eisentraut) |
| 191 | + </para> |
| 192 | + </listitem> |
| 193 | + |
| 194 | + <listitem> |
| 195 | + <para> |
| 196 | + Shorten file names reported in verbose error messages (Peter Eisentraut) |
| 197 | + </para> |
| 198 | + |
| 199 | + <para> |
| 200 | + Regular builds have always reported just the name of the C file |
| 201 | + containing the error message call, but VPATH builds formerly |
| 202 | + reported an absolute path name. |
| 203 | + </para> |
| 204 | + </listitem> |
| 205 | + |
| 206 | + <listitem> |
| 207 | + <para> |
| 208 | + Fix interpretation of Windows timezone names for Central America |
| 209 | + (Tom Lane) |
| 210 | + </para> |
| 211 | + |
| 212 | + <para> |
| 213 | + Map <quote>Central America Standard Time</> to <literal>CST6</>, not |
| 214 | + <literal>CST6CDT</>, because DST is generally not observed anywhere in |
| 215 | + Central America. |
| 216 | + </para> |
| 217 | + </listitem> |
| 218 | + |
| 219 | + <listitem> |
| 220 | + <para> |
| 221 | + Update time zone data files to <application>tzdata</> release 2011n |
| 222 | + for DST law changes in Brazil, Cuba, Fiji, Palestine, Russia, and Samoa; |
| 223 | + also historical corrections for Alaska and British East Africa. |
| 224 | + </para> |
| 225 | + </listitem> |
| 226 | + |
| 227 | + </itemizedlist> |
| 228 | + |
| 229 | + </sect2> |
| 230 | + </sect1> |
| 231 | + |
4 | 232 | <sect1 id="release-8-2-22">
|
5 | 233 | <title>Release 8.2.22</title>
|
6 | 234 |
|
|
0 commit comments