more content #10952

doc/release-notes/6.5-release-notes.md

@@ -29,6 +29,10 @@ The old "privateUrl" API endpoints for the creation and deletion of Preview (for
 
 See also #8184, #8185, #10950, and #10961.
 
+### More Scalable Dataset Version Differencing
+
+Differencing between dataset versions, which is done during dataset edit operations and to populate the dataset page versions table has been made signficantly more scalable. See #10814 and #10818.
+
 ### PostgreSQL and Flyway Updates
 
 This release bumps the version of PostgreSQL and Flyway used in containers as well as the PostgreSQL JDBC driver used all installations, including classic (non-Docker) installations. PostgreSQL and its driver have been bumped to version 17. Flyway has been bumped to version 10.
@@ -55,6 +59,12 @@ An optimization has been added for the Globus upload workflow, with a correspond
 
 See the [Database Settings](https://guides.dataverse.org/en/6.5/installation/config.html#GlobusBatchLookupSize) section of the Guides, #10977, and #11040 for more information.
 
+### Harvesting "oai_dc" Metadata Prefix When Extended With Specific Namespaces
+
+Some repository extend the "oai_dc" metadata prefix with specific namespaces. In this case, harvesting of these datasets was not possible because an XML parsing error was raised.
+
+Harvesting of these datasets has been fixed by excluding tags with namespaces that are not "dc:" and harvest only metadata with the "dc" namespace. See #10837.
+
 ### Harvested Dataset PID from Record Header 
 
 When harvesting, Dataverse can now use the identifier from the OAI-PMH record header as the persistent id for the harvested dataset.
@@ -63,6 +73,10 @@ This will allow harvesting from sources that do not include a persistent id in t
 
 It is also possible to optionally configure a harvesting client to use this OAI-PMH identifier as the **preferred** choice for the persistent id. See the [Harvesting Clients API](https://guides.dataverse.org/en/6.5/api/native-api.html#create-a-harvesting-client) section of the Guides, #11049 and #10982 for more information.
 
+### Harvested Datasets Can Have Multiple "otherId" Values
+
+When harvesting using the DDI format, datasets can now have multiple "otherId" values. See #10772.
+
 ## Bugs Fixed
 
 ### Updating Files Now Possible When Latest and Only Dataset Version is Deaccessioned
@@ -73,6 +87,10 @@ When a dataset was deaccessioned and was the only previous version it would caus
 
 The superuser-only feature of filtering by a username on the My Data page was not working. The "Results for Username" field now returns data for the desired user. See also #7239 and #10980.
 
+### Better Handling of Parallel Edit/Publish Errors
+
+Improvements have been made in handling the errors when a dataset has been edited in one browser window and an attempt is made to edit/publish it in another. (This practice is discouraged.) See #10793 and #10794.
+
 ### Version Differences Details Sorting Added
 
 In order to facilitate the comparison between the draft version and the published version of a dataset, a sort on subfields has been added. See #10969.
@@ -127,11 +145,23 @@ Previously, not setting these fields meant using a copy of the settings from the
 
 The JSON payload of the "get collection" endpoint has been extended to include properties isMetadataBlockRoot and isFacetRoot. See also [the docs](https://preview.guides.gdcc.io/en/develop/api/native-api.html#view-a-dataverse-collection), #11012, and #11013.
 
-### Get API Token Supports Any Auth
+### Image URLs from the Search API
+
+As of 6.4 (thanks to #10855) `image_url` is being returned from the Search API. The logic has been updated to only show the image if each of the following are true:
+
+1. The DataFile is not Harvested
+2. A Thumbnail is available for the Datafile
+3. If the Datafile is Restricted then the caller must have Download File Permission for the Datafile
+4. The Datafile is NOT actively embargoed
+5. The Datafile's retention period has NOT expired
+
+See also #10875 and #10886.
+
+## API Tokens
 
-The `/api/users/token` endpoint has been extended to support any auth mechanism for retrieving the token information.
+An optional query parameter called "returnExpiration" has been added to the "/api/users/token/recreate" endpoint, which, if set to true, returns the expiration time in the response message. See [the docs](https://preview.guides.gdcc.io/en/develop/api/native-api.html#recreate-a-token), #10857 and #10858.
 
-Previously, this endpoint only accepted an API token to retrieve its information. Now, it accepts any authentication mechanism and returns the associated API token information. See #10914 and #10924.
+The `/api/users/token` endpoint has been extended to support any auth mechanism for retrieving the token information. Previously, this endpoint only accepted an API token to retrieve its information. Now, it accepts any authentication mechanism and returns the associated API token information. See #10914 and #10924.
 
 ## Settings Added