| Bug #90172 | Inaccurate TYPE description in documentation for INNODB_SYS_INDEXES | ||
|---|---|---|---|
| Submitted: | 22 Mar 2018 10:31 | Modified: | 8 Jun 2018 17:39 |
| Reporter: | Alexey Kopytov | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Server: Documentation | Severity: | S3 (Non-critical) |
| Version: | 5.6, 5.7, 8.0 | OS: | Any |
| Assigned to: | CPU Architecture: | Any | |
[22 Mar 2018 10:44]
MySQL Verification Team
Hello Alexey, Thank you for the report. Thanks, Umesh
[4 Jun 2018 18:11]
Daniel Price
Posted by developer: The referenced documentation has been updated. The changes should appear online soon. Thank you for the bug report.
[6 Jun 2018 13:07]
Alexey Kopytov
Daniel, The changes appear to be live already at https://dev.mysql.com/doc/refman/8.0/en/innodb-indexes-table.html I see there's a typo: "1 = automatically generated clustered index (GEN_CLUST_INDEX); unique non-clustered index;" is missing "2 = " before unique
[8 Jun 2018 17:39]
Daniel Price
Posted by developer: Thank you, Alexey. The typo is fixed. The change should appear online soon.
Description: The manual provides inaccurate description for the TYPE column in the INFORMATION_SCHEMA.INNODB_SYS_INDEXES table (INFORMATION_SCHEMA.INNODB_INDEXES in 8.0): " A numeric identifier signifying the kind of index. 0 = Secondary Index, 1 = Clustered Index, 2 = Unique Index, 3 = Primary Index, ... " The TYPE column values are read from the 'type' field of the dict_index_t structure representing InnoDB's internal descriptor for each index. It is in fact a bitmap of the following flags defined in dict0mem.h: --- /** Type flags of an index: OR'ing of the flags is allowed to define a combination of types */ /* @{ */ #define DICT_CLUSTERED 1 /*!< clustered index; for other than auto-generated clustered indexes, also DICT_UNIQUE will be set */ #define DICT_UNIQUE 2 /*!< unique index */ ... --- Which means the user manual is a bit misleading about the values in the TYPE column. It should say that: 0 means a non-unique secondary index 1 means an automatically generated clustered index (which is always named GEN_CLUST_INDEX) 2 means a unique non-clustered index 3 means a clustered index (which may be either PRIMARY KEY, or a unique secondary index, depending on table schema) How to repeat: Read description of the TYPE column at https://dev.mysql.com/doc/refman/5.7/en/innodb-sys-indexes-table.html and https://dev.mysql.com/doc/refman/8.0/en/innodb-indexes-table.html Compare with the DICT_* flag descriptions in storage/innobase/dict0mem.h.