~drizzle-trunk/drizzle/development : changes

~drizzle-trunk/drizzle/development » Changes from revision 1638.9.1

From Revision 1638.9.1 to 1632

Rev	Summary	Authors	Date
1638.9.1	Add FOREIGN KEY constraints to the table proto. set them, ad... Add FOREIGN KEY constraints to the table proto. set them, add code in statement_transform for them, which means we now get foreign keys in the transaction_log. We don't go changing any existing code manipulating foreign key data structures and instead just create our own (minimal changes FTW).	Stewart Smith	14 years ago
1638.8.1	make the compare_length() of index parts of VARCHAR and BLOB... make the compare_length() of index parts of VARCHAR and BLOB (i.e. those with charsets) be a length in characters instead of bytes. This makes the logic converting a table proto back into a CREATE TABLE statement remotely sane. This does mean that if we increase the number of bytes utf8_general_ci uses, engines may have issues with their on disk formats (if they're not too smart)	Stewart Smith	14 years ago
1638.7.2	Merged bug600088-multibyte-comments-in-show-create into Merged bug600088-multibyte-comments-in-show-create into bug600635-table-proto-default-null-false.	Stewart Smith	14 years ago
1638.7.1	in the table proto, having default_null set to false explici... in the table proto, having default_null set to false explicitly (instead of just having it as a default) had a different behaviour than if it was not set and defaulting to NULL. This meant that you'd get empty defaults instead of no default. This only showed up in table_function tables.	Stewart Smith	14 years ago
1638.6.2	fix multibyte comments in SHOW CREATE TABLE. Was a simple of... fix multibyte comments in SHOW CREATE TABLE. Was a simple off-by-one bug.	Stewart Smith	14 years ago
1638.6.1	add test for bug600088: multibyte comments not shown correct... add test for bug600088: multibyte comments not shown correctly in SHOW CREATE TABLE	Stewart Smith	14 years ago
1638.3.5	update schema_dictionary.show result to show CURRENT_TIMESTA... update schema_dictionary.show result to show CURRENT_TIMESTAMP instead of NOW() as this is waht we have in the proto now. This is back to MySQL style behaviour (not that it matters at all)	Stewart Smith	14 years ago
1638.3.4	update type_timestamp result to show CURRENT_TIMESTAMP inste... update type_timestamp result to show CURRENT_TIMESTAMP instead of NOW() as this is waht we have in the proto now. This is back to MySQL style behaviour (not that it matters at all)	Stewart Smith	14 years ago
1638.3.3	store CURRENT_TIMESTAMP instead of NOW() in the table proto ... store CURRENT_TIMESTAMP instead of NOW() in the table proto as this is what we end up displaying in SHOW CREATE so more accurately represents what's going on. Although it could go either way (it totally does not matter), this paves the way for SHOW CREATE TABLE running through statement_transform.	Stewart Smith	14 years ago
1638.3.2	use default_expression and update_expression instead of defa... use default_expression and update_expression instead of default_value and update_value in the table proto for NOW()	Stewart Smith	14 years ago
1638.3.1	TimestampFieldOptions in the table proto contains nothing th... TimestampFieldOptions in the table proto contains nothing that we use. remove it.	Stewart Smith	14 years ago
1638.2.1	remove unused row_type from table proto (although referenced... remove unused row_type from table proto (although referenced in a bunch of places)	Stewart Smith	14 years ago
1638.1.1	have table wide default collation/collation_id in table prot... have table wide default collation/collation_id in table proto as well as move to only storing it when it's not the default (utf8_general_ci). This means that the replication stream now doesn't produce COLLATE= in the transformation of create table back to SQL.	Stewart Smith	14 years ago
1638	Fixes from Brian to increase strictness/correctness, Toru to... Fixes from Brian to increase strictness/correctness, Toru to remove current_session from blitzdb, a race condition bugfix from Joe Daly, and moving of when plugins call program_options from Monty.	Monty Taylor	14 years ago
1637	Merge in changes to call error on bad data input. Merge in changes to call error on bad data input.	Brian Aker	14 years ago
1636	Merge in work for count_cuted_fields Merge in work for count_cuted_fields	Brian Aker	14 years ago
1635	Merge up of Brian to trunk. Mainly work on ha_open, and some... Merge up of Brian to trunk. Mainly work on ha_open, and some current_session work.	Brian Aker	14 years ago
1634	Merge of the last bits of program_options plugin module inte... Merge of the last bits of program_options plugin module integration: printing help messages.	Monty Taylor	14 years ago
1633	Merge of Patrick's schema_dictionary test changes, Stewart's... Merge of Patrick's schema_dictionary test changes, Stewart's HA Status variable move and more module program_options work.	Monty Taylor	14 years ago
1632	Undid attempt at fixing sporadic freebsd issue. Undid attempt at fixing sporadic freebsd issue.	Monty Taylor	14 years ago

Older »