When the label of the context cannot be resolved, the context tag is not added.To ensure tests do not make any external calls, the code has been refactored so that the proper mock could be defined.
Ensure an index on id field
Log request body
Fixed direct indexing to projects with no funding streams and with escaped '/' in the grant code. Fixed UKRI acronym/names
added rules to match funder name ukri. When ukri or rcuk is found ukri is the funder
no need to commit immediately, the periodic task will do it
direct indexing api relies on a single solr client instance, which gets reconfigured on the expiration of the cached info retrieved from the IS, indicating the solr collection to be used
Added missing funders (including EEA, see #5564)
Fixed typo in log
catch exception thrown by rest service
added property for community api address
reintegrated branch solr75 -r53793:HEAD
moving business logic to OpenAIRESubmitterUtils class
catch and ignore all the exceptions to keep the autocommit task scheduled
update the list of funder
added method to filter out Zenodo communities from context
added call to method to filter out zenodo community from result context
Updated docs to include software and orps
check for zenodo record not direct indexed
Fixed missing bean error at container startup because of indexclientmap not found
Cache eviction on IS lookup queries
using ScheduledExecutorService for the autocommit task
reschedule index commit task when it dies
do not allow to set autocommit frequency values lower than a defined minimum
close the connection!
do not commit for every direct index request, instead delegate the commit to a periodic task
Added MIUR and fixed Finland (prefix is aka, not aff anymore)
Added OPEN SOURCE access right to be used for software
Updated record template and code to fix: 1. bestlicense and license must be bestaccessright and accessright; 2. Persons are not entities anymore and are now stored in the creator field; 3. Added empty journal field and other minor fields to reflect the current records in the production and beta indexes.
simple code refactoring
fetch also context profiles typed as 'ri'
logs and exceptions
renemaed some packages
implementation
cleaning
first import
Use misc-utils escaper: the one in velocity is too old and does not cover all XML entities!
upgraded version of jdbc driver for postgres, this includes a lot of changes, Because we have to add @transactional annotation in some properties to avoid connection closed exception
merged branch dsm into trunk
#3392: added funder param to FET-H2020 projects
Adding an almost geenric approach to obtain funder prefixes and ids. For names, however, we are still hard-coding them
Again, fixes name of FCT for direct indexing
Fixed FCT name
stream claims from the jdbcTemplate instead of load them in memory
fix bug for fet-h2020 context
Added method to get an id of the submitted result
Log json payload of the request upon exception
Fixed toString() and comments
trivial fixes
use single quote instead of double quote
integrated (hopefully) all required changes from dnet40
Fixed implementation for keeping track of progress
fixed parameter order
Node for writing context updates on hbase and refactoring
ApplyClaims with progress provider
Let's not fail for a non well-formed claim: we go on and keep track of it.
Updated toString
Fixed order of arguments
Fixed semantics and added test
Fixed ApplyClaimRels job param and mapping datasets and pubs to results
Using database service classes to avoid exception 'Method org.postgresql.jdbc4.Jdbc4Connection.isValid(int) is not yet implemented.'. We are locked to an old jdbc driver (for hibernate?)
Getters and setters
nodes and workflows for the new claims
Classes and configuration for writing claimed relationships into HBASE
Claims metadata are collected from the file system, not from the database.
updated the direct index api
removed an invalid migration of the apis
migrated to dnet45
codebase used to migrate to java8 the production system