The release notes will tell you what’s new in each version, and any changes that you must be aware of when upgrading. For reference, Vidispine ticket numbers are printed as (#1234).
- Using empty:/// URI does not work for truncating audit log and transfer logs (#3609).
- HTTP 400 when setting external ids or metadata field groups in metadata (#3599).
- Incorrect file list counts if a file belongs to multiple items for Elasticsearch (#3595).
- Incorrect search result if the search value contains “literal” spaces for Elasticsearch (#3594).
- Auto-completion not case insensitive for string-exact for Elasticsearch (#3593).
- SAML authentication broken with Bouncy Castle 1.60 (#3588).
- Too strong SSH key generated by Vidispine on Java 1.8u91 and later (#3583).
- Proxy links for manually getting file data does not respect ‘apiNoauthUri’ property (#3096).
- Sorting users by username/realname not working when searching for users (#2961).
- Tearing in video output when deinterlacing source (#3606).
Sequence rendering using Vidinet¶
Sequence rendering can now be performed using Vidinet. This works just like when using a Vidinet transcoder to transcode or conform (#3221).
Stream files from agent¶
You can now get access to read files directly from the VSA. Previously you could only access the files by streaming them through Vidispine. This make sense, if you for example are on the same on-premise network as the VSA and you have Vidispine server running in the cloud. The access is requested with a new methodType: AUTO-VSA.
Paging using cursors¶
Only metadata searches in the
generic interval supports
It is now possible to boost field values in item and collection searches by adding boost factors in the search documents or metadata field definitions.
PUT /item Content-Type: application/xml <?xml version="1.0" encoding="UTF-8"?> <ItemSearchDocument xmlns="http://xml.vidispine.com/schema/vidispine" version="2"> <operator operation="OR"> <field> <name>title_field</name> <value boost="10">phoenix</value> </field> <field> <name>description_field</name> <value>phoenix</value> </field> </operator> <sort> <name>_score</name> <value>descending</value> </sort> </ItemSearchDocument>
<MetadataFieldDocument xmlns="http://xml.vidispine.com/schema/vidispine"> <name>title_field</name> <type>string</type> <boost>10.0</boost> </MetadataFieldDocument>
Key-value metadata improvements¶
Support for key-value metadata has been added to:
- Shape tags (#3440)
- Libraries (#3412)
- Task definitions (#3439)
- Support for bulk upload of files to a placeholder shape (#3434).
- Option to configure which file extensions to
not treat as sidecar files(#3430).
- Added missing permissions for multipart upload using temporary S3 credentials (#3548).
- Support for retrieving temporary S3 credentials from the EC2 instance profile (#3490).
- Ability to filter on groups that are roles when searching for groups (#3487).
- Support for pagination when searching for users (#3273).
- Support for creating multiple (item) relationships in a single request (#3452).
- Support for disabling the built in SSH server (#3448).
- Support for specifying destination storage of output file for transcode jobs (#3424).
- Support for metadata dataset models in JSON-LD format (#3416).
- Proper error message if transcoder discovery settings is not correct (#3483).
- Reduce the possible “Lock wait timeout” when running jobs (#3370).
- SMB 2 support (#3510).
- Authentication failure when VSA connects over SSH to Vidispine running on Java 1.8u91 or later (#3569).
- Generating a pre-signed Azure URL results in a HTTP URL, expected HTTPS (#3551).
- Only sidecar files with lower case file extensions are picked up as sidecars (#3429).
- Resolution and aspect ratio not marshalled properly with useJackson=false when getting shapes as JSON (#3422).
- GET or DELETE of a non-existent metadata lock returns 500 (#3303).
- Possible metadata migration failure due to long running database queries (#3280).
- Component import to shape fails because of duplicate job creation (#3278).
- Don’t allow transient metadata fields to be modified (#3277).
- HTTP 500 response when importing a Final Cut XML that has a project element with no children (#3226).
- Missing mediainfo section for shapes imported by placeholder imports (#3206).
- GET storage ignores the status parameter (#3148).
- NPE instead of proper error message if export template fails to evaluate (#2889).
- Incorrect file key logged for API requests (#3568).
- The transcoders StatsD metrics not compliant with collectd (#3192).
Upgrading from 4.15¶
- Solr: No changes to the documents. Re-indexing is not required.