Virtuoso 08.03.3323 Release Notes
We are pleased to announce immediate availability of a new Virtuoso Maintenance Release comprising a wide variety of bug fixes and feature enhancements across the following functionality realms.
- Virtuoso DBMS Engine Core
- SPARQL Query Processor
- Plugins
- Data Virtualization
- Faceted Search & Browsing
- Virtual Authentication Layer (VAL)
- SHACL Data Validation
- Data Replication
- HTML-based Admin UI (“Conductor”)
- WebDAV Filesystem
- Native Connectors for Jena and RDF4j
Immediate benefits following installation of this maintenance release include higher-performance & scalability (especially in regards to write and delete operations), smarter security, increased stability, and generally enhanced platform for exploiting the fundamental potential of secure and high-performance data connectivity across on-premise and cloud deployment configurations.
Virtuoso Engine
- Added faster check for version of installed VAD package
- Added initial list of languages to decrease risk of deadlocks
- Added support for handling
X-Forwarded-Proto
header from proxy - Added more secure
bif
s - Added support for
cast
epoch time back todate
/datetime
- Added Elliptic Curve Digital Signature Algorithm (ECDSA) support
- Added support for
CONNECT
to allow http proxy (likesquid
) to tunnelhttps://
requests - Added optimization of
rdf_delete_triple_c()
for faster DELETE operations - Fixed
dc_max_batch_sz
cannot be more thansz
of boxed array - Fixed handling
tableName
attribute, both quoted table names as well as fully qualified table names, as per spec - Fixed return HTTP
503
when faults occur at the protocol level, even ifMaintenance Page
response page doesn’t exist or hasn’t been configured - Fixed missing index index creation following upgrade of older databases
- Fixed missing qualifiers and uppercase names in built-in Stored Procedures
- Fixed only use
charset_recode
if value isvarchar
- Fixed efficiency of RDF label language lookup
- Fixed UTF-8 encoding to prevent loss of encodings
- Fixed
backup_online
syntax inconsistencies - Fixed support for
NCHAR
,NVARCHAR
, andLONG NVARCHAR
fields in SQL Tables - Removed redundant checkpoints when creating new database
- Fixed issue with
backup restore
and DDL from plugins - Fixed lower limit for max batch to 100K
- Fixed allow
srv_stat ('git_head')
on all builds - Fixed issue with registering tables in plugin
- Fixed issue building
getrusage
bif
- Fixed trace to log warnings as
WARN_0
instead ofERRS_0
- Fixed allow certain digest functions to accept string or binary args to avoid explicit type cast
- Fixed small issue in
http_proxy_v2
- Fixed check for https behind proxy for dynamic local
- Fixed issue executing
vec
exec expression inWHERE
clause - Fixed return
404
error ifusage.vsp
is called with bad url - Fixed HTTP logging save correct CEST timezone values
- Fixed Windows binary to prevent crash loading database statistic diagnostic file
-
http_proxy
extended to support HTTPS - Fixed Virtuoso Windows service crash running security vulnerability scan
- Added “
MaxGroupByHashKeys
” INI file param to control number of keys inGROUP BY
or hashJOIN
SPARQL Query Processing
- Added support to fine-tune size of memory pool used by SPARQL
CONSTRUCT
withMaxMemInUse
param - Added missing JSON support function to format output of
ASK
query - Added support to use built-in
rdf_label
to prevent duplicate caching of literal object values - Added
dump_graph
anddump_nquads
as built-in stored procedures - Added use a standard namespace URI for special
bif:
andsql:
with regards to SPARQL Built-in functions - Fixed incorrect handling of UTF8 characters on SPARQL HTTP endpoint
- Fixed issue with
Anytime Query
and default query timeout values presented via /sparql query editor - Fixed issue with SPARQL
ASK
in embedded Stored Procedures in regards to SPASQL - Fixed issue with SPARQL variables containing Unicode characters
- Fixed vectorized functions for creating RDF literals
- Fixed RDF loading re. transaction modalities and ACID
- Fixed LDP metadata generation
- Fixed issue with SPARQL
ASK
- Fixed issue with incomplete RDF box in regards to SPASQL usage by ODBC or JDBC clients
- Fixed decoding of percent-encoded URLs when used as labels
- Fixed caching query via plink
- Fixed issue with bnodes
- Fixed memory leaks in transitive queries
- Fixed SPARQL endpoint to correctly process comments in query
- Fixed scenario where SPARQL query comprising “
HAVING
” clause lead to crashes - Fixed scenario where SPARQL query comprising “
GROUP BY
” and “HAVING
” clause lead to crashes - Fixed
/sparql-auth
endpoint where input Form Page redirected to/sparql
following query execution rather that returning back to/sparql-auth
- Fixed
Access-Control-Allow-Headers
response headers to include DPoP as a supported OpenID Connect modality - Fixed
G IN
hash mechanism to work forO IN
predicate - Fixed
ttlp()
function to correctly handle loading ofxsd:boolean
types
Plugins
- Added
GEOS-isValid
bif
Data Virtualization
R2RML Processor
- Fixed R2RML to always use
rr:graph
setting when specified
Data Transformation Middleware Layer (“Sponger”)
- Fixed Sponger description pages to only make live links when protocol scheme is safe (
http
,https
,ftp
) - Fixed Sponger
/about
HTML page auto refresh issue
Faceted Search and Browsing
- Added
FCT
configuration options for managing HTTP server cache directives returned via response headers - Added amount of ‘memory in use’ in footer
- Added some
nofollow
andnoindex
hints for bots
Virtual Authentication Layer (VAL)
- Added Bitcoin Block Chain (BTC) as a Profile Provider supported by the NetID authentication protocol enhancement
- Added protocol hooks for WebID/NetID
- Fixed only use
RPERM
to handle error in VAL - Fixed WebID authentication API
SHACL Data Validation
- Fixed
SHACL_ENDPOINT_PREPARE_GRAPH
to remove grant fromSPARQL SPONGE
Data Replication
- Fixed Graph replication to delete string object values publisher and subscriber nodes
HTML-based Admin Interface (“Conductor”)
- Fixed irrelevant errors from Conductor DAV browser pages
- Fixed missing checks related to ODS Briefcase instances
WebDAV File System
- Added helper function for fixing DAV
COL_FULL_PATH
Native Connectors for Jena and RDF4j
- Upgraded to use of native
rdf_delete_triple_c()
procedure for faster SPARQL delete operations - Upgraded Native Jena4 Provider to use latest version of
junit
andlog4j
- Fixed Native Jena providers to prevent “
SR491 Too many open statements
” errors in the JDBC driver - Fixed
SQLConnect
handling of empty strings inszDSN
andszUID
Availability
On-premise Installers and Docker Containers
- On-premise Installers for Linux, macOS, and Windows
- Docker Containers
New BYOL (Bring-Your-Own License) and PAGO (Pay-As-You-Go) Virtual Machine Releases for Cloud Computing Platforms (Amazon AWS and Microsoft Azure)
- Virtuoso Virtual Machines for the Amazon AWS Cloud
- OpenLink Data Junction Box Virtual Machine for the Amazon AWS Cloud
- Virtuoso Virtual Machines for the Microsoft Azure Cloud
- OpenLink Data Junction Box for the Microsoft Azure Cloud
Note: To update existing instances of Virtual Machines listed above, simply execute the relevant command:
- Ubuntu —
apt update & upgrade
- RedHat —
yum update