WIP Prepare to use unified schema in admin CSV export/import

Adjust page assignment checks to not exclude serial data from aggregations

For some reason the search query with clause 'assigned !=0' did not return rows where assigned is NULL. This should work ac

Adjust page assignment checks to not exclude serial data from aggregations

For some reason the search query with clause 'assigned !=0' did not return rows where assigned is NULL. This should work according to specs and does work as expected in standalone queries. Fixed by using an explicit IS NULL check.

show more ...

Remove islookup property of schemas

CSV import must still be adjusted

Fix schema assignment check in cloud

Fix schema assignment check in aggregations

Update cloud queries

Fix code standard violation

Restructure search query

Move schema assignment clauses into an ever-present JOIN, conditional on pid. Also remove the $idColumn parameter, previously used to check if schema assignment clauses shou

Restructure search query

Move schema assignment clauses into an ever-present JOIN, conditional on pid. Also remove the $idColumn parameter, previously used to check if schema assignment clauses should be added to queries.

show more ...

More flexible page access check in search queries

Add page access clauses in a subquery, conditional on non-empty page id

Manual coding style fixes

Method and variable visibility, camel case method names. Some exclusions due to inheritance

Automatic coding style fixes

defer loading of the JSON editor

to make it compatible with DokuWiki master

fix curly braces warning on 7.4

Fix permission check in page aggregations

Access Dokuwiki's refactored lexer

Lexer changed significantly between current stable Greebo and master. Now we can use both versions.

Make more SQL queries contingent on the join column ($idColumn)

Fix data access for lookup types

Pass current time instead of 0 to fetch data of latest page revision

Use current time explicitly when fetching data for the latest revision of a page

Remove redundant serial code

There is actually very little need to split the handling of lookup (not page specific) and serial (page bound) data, because row ids are globally unique and never shared

Remove redundant serial code

There is actually very little need to split the handling of lookup (not page specific) and serial (page bound) data, because row ids are globally unique and never shared between those two types.

show more ...

Consolidate table access classes

Fix multi columns

Use the correct column for joining tables depending on data type. This is an initial "it works" implementation and should be revised and done more carefully.

Using page revisions in inline editor guarantees that correct data access is initialized

Extract data type detection into public methods

Fix SQL for multi fields