[DBAL-1063] Allow defining duplicate indexes on a table #764
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Hello, thank you for creating this pull request. I have automatically opened an issue http://www.doctrine-project.org/jira/browse/DBAL-1102 We use Jira to track the state of pull requests and the versions they got |
| Duplicate indexes are considered indexes that pass ``isFullfilledBy()`` or ``overrules()`` | ||
| in ``Doctrine\DBAL\Schema\Index``. | ||
| This is required to make the index renaming feature introduced in 2.5.0 work properly and avoid | ||
| race conditions in the ORM schema tool / DBAL schema manager which pretends users from updating |
Ocramius
added a commit
that referenced
this pull request
Jan 3, 2015
[DBAL-1063] Allow defining duplicate indexes on a table
Ocramius
added a commit
that referenced
this pull request
Jan 3, 2015
|
Backported into |
This was referenced Dec 6, 2015
1 task
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR removes silent dropping of "duplicate" indexes on a table if two indexes fulfill each other or one overrules the other.
This is necessary to make detection of renamed indexes work properly (introduced in 2.5). Currently there are situations where false positive index renamings are detected because of race conditions.
Example
An ORM entity contains an association which adds a foreign key constraint to the underlying table. Foreign key constraints always create implicit indexes on the particular columns.
The entity also contains a mapping for a custom named index fulfilling the foreign key column(s).
The ORM schema tool first adds the foreign key constraint to the table, implicitly creating an index. Afterwards it adds the custom named index which is dropped silently by the table instance because it is considered a "duplicate".
So the table instance created by the schema now contains the auto generated index but not the custom named one.
When using the schema tool to update the schema, the DBAL schema manager is utilized to create the "online" schema. The schema manager detects two indexes in the database, the auto generated one and the custom named one (manually created by the user).
When the schema manger creates the table instance it eventually adds the custom named index first which results in the auto generated index being dropped silently.
When comparing both "offline" and "online" schema, the comparator detects an index rename which actually is none.
This leads to failing
DROP INDEXandCREATE INDEXSQL.Besides this issue "duplicate" indexes are totally legit and DBAL should not decide whether or not to create an index. This should be user responsibility.
This implementation is much more transparent and less error prone and opens up more flexibility for the user to define indexes on tables.
/cc @flack @Ocramius @beberlei