mattcasters opened a new pull request, #8559:
URL: https://github.com/apache/hop/pull/8559

   Fixes #8509.
   
   Several projects can share one execution-information database or search 
index. This adds an optional `projectId` on the project so each run can be 
tagged and the Execution Information perspective can show only that project.
   
   `projectId` is stored in `project-config.json` and shown on the project 
dialog, under the name. A new project suggests the project name (a UUID is fine 
too). An existing project is left blank, so opening the dialog and pressing OK 
does not start filtering. `hop-conf` does the same for a brand-new config file. 
While a project is active, Hop sets `HOP_PROJECT_ID`, including to empty, so a 
parent id does not leak into the child.
   
   The value is copied onto new execution documents only when it is set. It is 
omitted from the JSON when empty. Apache Hop 2.19.0 reads those documents with 
a strict parser, so an extra property would make them unreadable. Documents 
that do contain `projectId` are not readable by 2.19.0.
   
   Filtering:
   
   - **Caching database.** Nullable `project_id` column and index 
`idx_hop_exec_project`. If the column is missing, reads and writes keep the 
2.19.0 column list and nothing is filtered. Show DDL still does not execute 
SQL. It prints `CREATE TABLE` for a new table and, below a comment, `ALTER 
TABLE` plus the index for an existing table. Run one block, not both.
   - **Elastic and OpenSearch.** The same value is a root `projectId` keyword. 
Create index adds it. If the index already exists, the same action adds the 
field to the mapping and does not recreate the index.
   - With a project id set, the list shows that id plus rows and documents that 
have none. Other projects' tagged runs are hidden, including from 
last-execution lookups. With no id, the list is unfiltered.
   
   Neo4j and the remote location are unchanged.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to