mirror of
https://github.com/MariaDB/server.git
synced 2025-01-29 02:05:57 +01:00
MariaDB server is a community developed fork of MySQL server. Started by core members of the original MySQL team, MariaDB actively works with outside developers to deliver the most featureful, stable, and sanely licensed open SQL server in the industry.
amazon-web-servicesdatabasefulltext-searchgalerageographical-information-systeminnodbjsonmariadbmysqlrdbmsrelational-databasessqlstorage-enginevector-database
aa32336198
The problem in this bug is when we create temporary tables. When temporary tables are created for unions, there is some inferrence being carried out regarding the type of the column. Whenever this column type is inferred to be REAL (i.e. FLOAT or DOUBLE), MySQL will always try to maintain exact precision, and if that is not possible (there are hardware limits, since FLOAT and DOUBLE are stored as approximate values) will switch to using approximate values. The problem here is that at this point the information about number of significant digits is not available. Furthermore, the number of significant digits should be increased for the AVG function, however, this was not properly handled. There are 4 parts to the problem: #1: DOUBLE and FLOAT fields don't display their proper display lengths in max_display_length(). This is hard-coded as 53 for DOUBLE and 24 for FLOAT. Now changed to instead return the field_length. #2: Type holders for temporary tables do not preserve the max_length of the Item's from which they are created, and is instead reverted to the 53 and 24 from above. This causes *all* fields to get non-fixed significant digits. #3: AVG function does not update max_length (display length) when updating number of decimals. #4: The function that switches to non-fixed number of significant digits should use DBL_DIG + 2 or FLT_DIG + 2 as cut-off values (Since fixed precision does not use the 'e' notation) Of these points, #1 is the controversial one, but this change is preferred and has been cleared with Monty. The function causes quite a few unit tests to blow up and they had to b changed, but each one is annotated and motivated. We frequently see the magical 53 and 24 give way to more relevant numbers. mysql-test/r/create.result: bug#24791 changed test result With the changes made for FLOAT and DOUBLE, the original display lengths are now preserved. mysql-test/r/temp_table.result: bug#24791 changed test resullt Test case added mysql-test/r/type_float.result: bug#24791 changed test result delta 1: field was originally declared as DOUBLE with no display length, so the hardware maximum is chosen rather than 53. delta 2: fields exceed the maximum precision and thus switch to non-fixed significant digits delta 3: Same as above, number of decmals and significant digits was not specified when t3 was created. mysql-test/t/temp_table.test: bug#24791 Test case sql/field.h: bug#24791 The method max_display_length is reimplemented as uint32 max_display_length() { return field_length; } in Field_double and Field_float. Since all subclasses of Field_real now have the same implementation of this method, the implementation has been moved up the hierarchy to Field_real. sql/item.cc: bug#24791 We switch to a non-fixed number of significant digits (by setting decimals=NOT_FIXED_DECIMAL) if the calculated display length is greater than the display length of a value with the maximum precision. These values differ for double and float, obviously. sql/item_sum.cc: bug#24791 We must increase the display length accordinly whenever we change number of decimal places. |
||
|---|---|---|
| bdb | ||
| BitKeeper | ||
| BUILD | ||
| client | ||
| cmd-line-utils | ||
| config/ac-macros | ||
| dbug | ||
| Docs | ||
| extra | ||
| heap | ||
| include | ||
| innobase | ||
| libmysql | ||
| libmysql_r | ||
| libmysqld | ||
| man | ||
| myisam | ||
| myisammrg | ||
| mysql-test | ||
| mysys | ||
| ndb | ||
| netware | ||
| os2 | ||
| pstack | ||
| regex | ||
| scripts | ||
| server-tools | ||
| sql | ||
| sql-bench | ||
| sql-common | ||
| SSL | ||
| strings | ||
| support-files | ||
| tests | ||
| tools | ||
| VC++Files | ||
| vio | ||
| win | ||
| zlib | ||
| .bzrignore | ||
| .cvsignore | ||
| CMakeLists.txt | ||
| configure.in | ||
| configure.in.rej | ||
| COPYING | ||
| EXCEPTIONS-CLIENT | ||
| Makefile.am | ||
| README | ||
This is a release of MySQL, a dual-license SQL database server. MySQL is brought to you by the MySQL team at MySQL AB. License information can be found in these files: - For GPL (free) distributions, see the COPYING file and the EXCEPTIONS-CLIENT file. - For commercial distributions, see the LICENSE.mysql file. For further information about MySQL or additional documentation, see: - The latest information about MySQL: http://www.mysql.com - The current MySQL documentation: http://dev.mysql.com/doc Some manual sections of special interest: - If you are migrating from an older version of MySQL, please read the "Upgrading from..." section first! - To see what MySQL can do, take a look at the features section. - For installation instructions, see the Installing and Upgrading chapter. - For the new features/bugfix history, see the Change History appendix. - For the currently known bugs/misfeatures (known errors) see the Problems and Common Errors appendix. - For a list of developers and other contributors, see the Credits appendix. A local copy of the MySQL Reference Manual can be found in the Docs directory in GNU Info format. You can also browse the manual online or download it in any of several formats at the URL given earlier in this file. ************************************************************ IMPORTANT: Bug or error reports should be sent to http://bugs.mysql.com.