Databases
StackOverflow2013
Postsdbo
POBreaking apart a csv string of id's in MySQL to reference for a subquery
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POBreaking apart a csv string of id's in MySQL to reference for a subquery
    primarykey
    data
    text
    <p>The following query will return the id associated with a given document, a human readable list of permissions that are associated with a document, a string containing id's of groups, roles, and users who have that permission assigned to them in a document (I'm aware that this method of assigning permissions is garbage, alas it's what I was handed), and whether or not everyone has access via a <code>REGEXP</code> check. 1 = yes, everyone should have access to that permission on this document, 0 = not everyone.</p> <pre><code>SELECT `D`.`id` AS `docID`, `P`.`human_name` AS `permissionName`, `PD`.`descriptor_text` AS `permissionAssignments`, `PD`.`descriptor_text` REGEXP '.*role\\((-4,)?-3.*' AS `everyoneAccess` FROM `documents` AS `D` INNER JOIN `permission_lookups` AS `PL` ON `D`.`permission_lookup_id` = `PL`.`id` INNER JOIN `permission_lookup_assignments` AS `PLA` ON `PL`.`id` = `PLA`.`permission_lookup_id` INNER JOIN `permission_descriptors` AS `PD` ON `PLA`.`permission_descriptor_id` = `PD`.`id` LEFT JOIN `permission_descriptor_roles` AS `PDR` ON `PDR`.`descriptor_id` = `PD`.`id` LEFT JOIN `permissions` AS `P` ON `PLA`.`permission_id` = `P`.`id` WHERE `D`.`id` = 74; </code></pre> <p>The output on this specific query looks similar to:</p> <pre><code>74 Read group(1)role(-3)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294,15389) 1 74 Write group(1)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294) 0 74 Add Folder group(1)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294) 0 74 Manage security group(1)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294) 0 74 Delete group(1)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294) 0 74 Manage workflow group(1)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294) 0 74 Folder Details group(1)role(-3)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294,15389) 1 74 Rename Folder group(1) 0 </code></pre> <p>Under the 3rd column, first row (for example), the string: <code>group(1)role(-3)user(13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294,15389)</code> means that the group with id 1 has access to the read permission, the role with id -3 has access (-3 means everyone) to the read permission, and with different role allocations and such, at a resolved state, users with the ids 13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294,15389 have access. So in this instance, everyone has access due to the document having the everyone role assigned to it, but they also gave specific access to the long string of users as a blanket across all permissions.</p> <p>What I want to do, is pull out the csv list of users, and use that in a subquery to pull out their usernames in the same query. I try the following query but without any luck, it only returns the first user in the csv list:</p> <pre><code>SELECT `D`.`id` AS `docID`, `P`.`human_name` AS `permissionName`, `PD`.`descriptor_text` AS `permissionAssignments`, `PD`.`descriptor_text` REGEXP '.*role\\((-4,)?-3.*' AS `everyoneAccess`, IF(`PD`.`descriptor_text` REGEXP '.*user\\([0-9,]+)$', (SELECT group_concat(`username`) FROM users where `id` IN ( SUBSTR( `PD`.`descriptor_text`, INSTR(`PD`.`descriptor_text`, 'user(') + 5, LENGTH(`PD`.`descriptor_text`) - INSTR(`PD`.`descriptor_text`, 'user(') - 5 ) ) ), NULL ) FROM `documents` AS `D` INNER JOIN `permission_lookups` AS `PL` ON `D`.`permission_lookup_id` = `PL`.`id` INNER JOIN `permission_lookup_assignments` AS `PLA` ON `PL`.`id` = `PLA`.`permission_lookup_id` INNER JOIN `permission_descriptors` AS `PD` ON `PLA`.`permission_descriptor_id` = `PD`.`id` LEFT JOIN `permission_descriptor_roles` AS `PDR` ON `PDR`.`descriptor_id` = `PD`.`id` LEFT JOIN `permissions` AS `P` ON `PLA`.`permission_id` = `P`.`id` WHERE `D`.`id` = 74; </code></pre> <p>What can be changed in that 4th field to have it query for a group_concat list of the usernames based off the csv of id's?</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. USScott
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POBreaking apart a csv string of id's in MySQL to reference for a subquery
      1. This table or related slice is empty.
      1. VTDownMod
    2. VO
      singulars
      1. POBreaking apart a csv string of id's in MySQL to reference for a subquery
      1. USKonerak
      1. VTFavorite
    3. VO
      singulars
      1. POBreaking apart a csv string of id's in MySQL to reference for a subquery
      1. This table or related slice is empty.
      1. VTUpMod
    1. COThe reason it doesn't work currently is that `IN` is trying to match the string you provided ("13,14,26,38,56,73,81,88,103,108,138,140,148,158,159,175,191,276,294,15389" for row 1 above), but you actually need a list of values. There is a good bit of discussion regarding string splitting on the [MySQL Reference Manual](http://dev.mysql.com/doc/refman/5.0/en/string-functions.html), but I'm not sure how to apply it to your situation yet.
      singulars
      1. POBreaking apart a csv string of id's in MySQL to reference for a subquery
      1. USraistlin0788
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload