aboutsummaryrefslogtreecommitdiffstats
path: root/searchcore/src/tests/proton/documentdb/feedhandler
Commit message (Collapse)AuthorAgeFilesLines
* Add helper classes containg a FNET_Transport and a FastOS_ThreadPool to ↵Henning Baldersheim2022-02-211-7/+3
| | | | reduce test boiler plate.
* Use a common FNET_Transport owned by Proton in both SceduledExecutor and ↵Henning Baldersheim2022-02-201-1/+9
| | | | | | | | | | TransactionLogServer. This reduces the number of Transport object by 1 per document type and netto 1 in Proton. Each of them contains 2 threads. In addition it uses a common Transport for the RpcFileAcquirer objects used during config fetching. This prevents creating 3 temporary Transport objects on every reconfig.
* Add DoneCallback to removeDocuments.Henning Baldersheim2021-11-301-1/+1
|
* Add DoneCallback to heartBeat so it can be tested without relying on sync.Henning Baldersheim2021-11-301-2/+2
|
* Update document db feeding metrics.Tor Egge2021-11-261-0/+14
|
* GC some unused code and less sync_all_executors.Henning Baldersheim2021-11-211-35/+3
|
* Do not inherit Syncable. sync -> sync_all_executors to make its call sites ↵Henning Baldersheim2021-11-161-1/+1
| | | | more evident.
* Update 2017 copyright notices.gjoranv2021-10-073-3/+3
|
* When making callbacks like notifying about new bucket show the gurad that is ↵Henning Baldersheim2021-02-141-1/+1
| | | | | | held. Then this guard can used instead of possibly making a deadlock if trying to take it yourself.
* Use vespalib::duration for timeoutsHenning Baldersheim2021-01-311-35/+5
|
* Various readability and code cleanup. closure -> lambdaHenning Baldersheim2021-01-231-1/+0
|
* Move idestructorcallback.h to vespalib module and ↵Henning Baldersheim2021-01-141-1/+1
| | | | search::IDestructorCallback to vespalib::IdestructorCallback.
* Move the feed reject helper to document module.Henning Baldersheim2021-01-111-47/+1
|
* Allow updates with only remove/clear updates and operations to single value ↵Henning Baldersheim2021-01-051-2/+72
| | | | | | | | | numeric fields to pass even when feed rejection is in effect. The rationale is that these updates will not affect memory or disk footprint negatively, but it can have a significant positive effect. Note that this will only be allowed for buckets that are in sync.
* use SimpleValue instead of EngineOrFactoryArne Juul2020-12-021-6/+6
|
* use factory.from_spec() instead of tensor::makeTensor<T>Arne Juul2020-11-271-6/+7
|
* Stop using PartitionId in searchcore module.Tor Egge2020-10-161-1/+0
|
* handle both engine- and factory-based tensorsArne Juul2020-10-151-4/+4
| | | | | | * use EngineOrFactory::get() instead of DefaultTensorEngine::ref() * avoid direct use of DenseTensorView etc where possible * use eval::Value instead of tensor::Tensor where possible
* Add startCommit methodHenning Baldersheim2020-09-211-0/+1
|
* Rename storeOperation to appendOperationHenning Baldersheim2020-09-211-2/+2
|
* Reduce code visibility.Henning Baldersheim2020-09-151-1/+0
|
* Revert "Revert "Configure compression and chunk size""Henning Baldersheim2020-09-031-1/+2
|
* Revert "Configure compression and chunk size"Arnstein Ressem2020-09-021-2/+1
|
* Configure compression and chunk sizeHenning Baldersheim2020-09-021-1/+2
|
* Style chnagesHenning Baldersheim2020-08-271-1/+0
|
* Use a fixed visibility-delay of 100ms during replayHenning Baldersheim2020-08-201-1/+0
|
* Do the time tracking for immediate commit in the replay state instead of ↵Henning Baldersheim2020-08-191-36/+32
| | | | having it spread all over.
* Remove default tensor spec in doctype builder.Geir Storli2020-06-171-2/+2
|
* Revert "Two phase put in attribute writer"Harald Musum2020-06-151-2/+2
|
* Remove default tensor spec in doctype builder.Geir Storli2020-06-151-2/+2
|
* - Add async interface to putHenning Baldersheim2020-04-291-3/+3
| | | | | - Use MessageTracker for keeping context. - implement putAsync, but still use it synchronously.
* Make RemoveOperation an interface and move implmentation to ↵Henning Baldersheim2020-04-081-12/+11
| | | | RemoveOperationWithDocId
* Add shutdown to thread interface.Henning Baldersheim2020-03-231-1/+0
| | | | Let the optimize config control index and summary executor too.
* WipeHistory has not been used for a very long time.Henning Baldersheim2019-12-161-5/+0
|
* GC unused filesHenning Baldersheim2019-09-301-1/+0
|
* doc: -> id:Henning Baldersheim2019-08-161-4/+4
|
* Use enum class for storage::spi::Result::ErrorType.Tor Egge2019-07-301-5/+5
|
* Wire in the shared executor in the threading service.Henning Baldersheim2019-05-141-1/+3
|
* Adjust wrong tensor type exception message.Tor Egge2019-03-071-1/+1
|
* Check for assignable tensor type when setting tensor in TensorFieldValue.Tor Egge2019-03-071-3/+46
|
* Remove files used by old unit test runner.Tor Egge2019-02-192-2/+0
|
* Reinstate update verificationHenning Baldersheim2018-06-141-4/+0
|
* Temporarily disable testHenning Baldersheim2018-06-131-0/+4
|
* 1 - Use a backing buffer for the DocumentUpdate that always is source of truth.Henning Baldersheim2018-06-121-1/+1
| | | | | | | | | | | 2 - Use this buffer for re-serialization. 3 - Make deserialization lazy where possible. Currently lazy on replay and when arriving over the storageapi. Still needs to eager over documentapi. 4 - Deserialize eagerly in the persistence thread since those are plentyfull and not bottlenecked, instead of in the single master thread. Use real repo.
* Merge branch 'master' into balder/reduce-code-visibility-rebasedHenning Baldersheim2018-06-121-0/+1
|\
| * Reduce code visibilityHenning Baldersheim2018-06-111-0/+1
| |
* | Reduce code visibilityHenning Baldersheim2018-06-121-0/+1
|/
* Perform fixup on put and update operations using non-current document type repo.Tor Egge2018-05-291-12/+38
| | | | | This ensures that document type used do access fields is the same for live feeding and replay.
* Extend unit test coverage to check validation of updates with fallback toTor Egge2018-04-051-11/+42
| | | | creating document.
* Add extra check if document type for update operation might be differentTor Egge2018-04-052-9/+78
| | | | (e.g. too old or too new) to ensure that we are able to handle replay.