summaryrefslogtreecommitdiffstats
path: root/common/resourceaccess.h
Commit message (Collapse)AuthorAge
* Insall sincommon_export.hChristian Mollekopf2016-01-31
|
* Explicitly state visibilityChristian Mollekopf2016-01-31
| | | | | | | | | This cuts the exportet symbols from 12k to 2k and the library size from 16Mb to 14Mb, which isn't great but still an improvement. About 30% of the symbols are coming from the Store:: interface. nm -C -D /work/install/lib64/libsinkcommon.so.0.1
* Renamed Akonadi2 to SinkChristian Mollekopf2016-01-20
| | | | (except for documentation).
* ResourceNotification -> NotificationChristian Mollekopf2016-01-19
|
* Reuse the queuedInvoke method for less boilerplateChristian Mollekopf2016-01-18
|
* Draft of inspection APIChristian Mollekopf2016-01-18
|
* Improved resource access cachingChristian Mollekopf2015-12-03
| | | | | | | | * Smarter caching. ResourceAccess instances close after a timeout, if not reused. * Introduced a start command to avoid race condition when sending commands to a resource that is currently shutting down. * We resend pending commands after we lost access to the resource * unexpectedly.
* Cache ResourceAccess instances.Christian Mollekopf2015-12-01
| | | | | | | | | We want one connection per application per resource instance. With this change the connection is maintained for the lifetime of the client process (it should probably time out instead), but we at least avoid creating a connection per operation/query, which results in a significant performance boost (~10% for reading, 90% for writing in dummyresourcebenchmark).
* Introduced a QueryRunner objectChristian Mollekopf2015-11-27
| | | | | | | | The QueryRunner object lives for the duration of the query (so just for the initial query for non-live queries, and for the lifetime of the result model for live queries). It's supposed to handle all the threading internally and decouple the lifetime of the facade.
* Let clients tell the resource when they no longer require a revision.Christian Mollekopf2015-10-19
|
* Change replayChristian Mollekopf2015-10-10
| | | | | So far only includes modifications and additions, removals are not yet stored as separate revisions.
* Moved sendCommand implementations to ResourceAccessChristian Mollekopf2015-08-12
| | | | So we have commands in one place, and not in a header file.
* Allow to inject a fake ResourceInstanceChristian Mollekopf2015-07-28
|
* Call callbacks for already completed commands before abortingChristian Mollekopf2015-07-23
|
* Fixed all warningsChristian Mollekopf2015-07-07
|
* KAsync has moved to it's own kasync.git repositoryDan Vrátil2015-05-18
|
* Adapt to KAsync namespace changeDan Vrátil2015-05-15
|
* Avoid starting the resource for shutdown.Christian Mollekopf2015-04-29
|
* ResourceAccess: rewrite connection code using jobsChristian Mollekopf2015-04-28
| | | | | Another nice showcase implementation how things get easier, testable and composable using async.
* Always queue commands in resourceaccess.Christian Mollekopf2015-04-12
| | | | | We want to keep the command until we know it arrived in the resource, so we can resend it otherwise.
* Use QByteArray instead of QStringChristian Mollekopf2015-04-09
| | | | | All identifiers should be latin1 and we make this explicit by using QByteArray. QString is reserved for strings that can be UTF-8 or alike.
* Don't try to restart the resource on every disconnect.Christian Mollekopf2015-03-31
| | | | | There's a chance that the resource actually wanted to shut-down. Instead ResourceAccess should only reopen the connection if it still has work to do.
* Propagate errors for commands.Christian Mollekopf2015-01-25
|
* A way to ensure all messages have been processed.Christian Mollekopf2015-01-25
| | | | | As queries become reactive this should become less important. We can then just wait until all results become available. For tests it is in either case useful though.
* Use jobs to track progress of write commands.Christian Mollekopf2015-01-18
|
* Writing from facade.Christian Mollekopf2015-01-15
|
* Turned synchronizeResource command into a job.Christian Mollekopf2015-01-06
|
* Buffers wrapped into entity buffer, async command progress tracking.Christian Mollekopf2014-12-28
|
* Write-Read loop from clientside.Christian Mollekopf2014-12-21
| | | | | | | | | It's a huge hack but starts to show results. Most urgently we need: * reliable command results * the 3 buffers instead of the 1 * A way to implement storage as preprocessor (or a place to impelement it after the preprocessors).
* use a dptr, API for sending commands, queue commands until connectedAaron Seigo2014-12-16
|
* move client classes into akonadi2common and add the base class for resource ↵Aaron Seigo2014-12-16
plugins we can divide up libakonadi2common later once we have a full collection of classes this makes writing code a bit simpler now as we don't have to figuer out which libraries to link against or how class dependencies should look. when we have more infrastructure in place this will mostly become self-evident
generated by cgit v1.2.3 (git 2.25.1) at 2025-07-18 22:23:25 +0000