summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* split out functions that get inlined into a separate fileAaron Seigo2014-12-16
| | | | allowed including the header more than once ;)
* first run at a plugin loaderAaron Seigo2014-12-16
|
* typo in commentAaron Seigo2014-12-16
|
* split out the resource bits into their own fileAaron 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
* it's a name, not a resource (that is going to be a class soon)Aaron Seigo2014-12-15
|
* licensesAaron Seigo2014-12-15
|
* lgplAaron Seigo2014-12-15
|
* LGPL the library classesAaron Seigo2014-12-15
|
* set a good trend for akonadi2 library name: akonadi2_resource_<name>Aaron Seigo2014-12-15
| | | | also, correct Qt library usage
* namespace everything in libs with Akonadi2Aaron Seigo2014-12-15
|
* Make lmdb work with threads.Christian Mollekopf2014-12-15
|
* avoid integer overflow ... at least on 64 bit.Aaron Seigo2014-12-15
|
* shared libraries, install targetsAaron Seigo2014-12-15
|
* make libhawd a shared libAaron Seigo2014-12-15
|
* restore the test to its former gloryAaron Seigo2014-12-14
|
* Create base directory if necessary.Christian Mollekopf2014-12-14
|
* cleanupChristian Mollekopf2014-12-14
|
* Threaded query processing.Christian Mollekopf2014-12-14
|
* mkpath instead of mkdir, as we may be making multiple dirsAaron Seigo2014-12-14
|
* compile ixes, make akonadi2common an actual shared libAaron Seigo2014-12-14
|
* buildAaron Seigo2014-12-14
|
* the all-in-one-file version of unqliteAaron Seigo2014-12-14
|
* Async: bring back synchronous Future::waitForFinished()Dan Vrátil2014-12-14
|
* Async: make the processing truly asynchronousDan Vrátil2014-12-14
| | | | | | Now calling exec() starts the first job and returns a pending Future immediately. Caller can then use Async::FutureWatcher to wait for the future to become finished, i.e. for all jobs to finish execution.
* put unqlite db's in their own subdir to avoid name collisionsAaron Seigo2014-12-14
|
* Merge branch 'unqlite'Aaron Seigo2014-12-14
|\
| * and now unqlite storage worksAaron Seigo2014-12-14
| |
| * build unqlite with thread supportAaron Seigo2014-12-14
| |
| * remove readAll in favor of using scanAaron Seigo2014-12-14
| |
| * try out unqliteAaron Seigo2014-12-11
| |
* | remove readAll in favor of using scanAaron Seigo2014-12-14
| |
* | Async: make it build againDan Vrátil2014-12-12
| |
* | Missing includesDan Vrátil2014-12-12
| |
* | Drop a temporary fileDan Vrátil2014-12-12
| |
* | Async: add some simple documentationDan Vrátil2014-12-12
| |
* | Async: move some more code out-of-lineDan Vrátil2014-12-12
| |
* | Async: move some common code from executors to shared base classDan Vrátil2014-12-12
| |
* | Async: add Reduce taskDan Vrátil2014-12-12
| |
* | ClientAPI: Demonstrate how we can build powerful and flexible API's on top ↵Christian Mollekopf2014-12-12
| | | | | | | | of ResultEmitter
* | Async: move the actual task exection into Executor implementationDan Vrátil2014-12-11
| | | | | | | | | | | | | | As of now, Job is only front interface to a chain of Executor subclasses. Each Executor subclass specializes for given type of execution (then, each, reduce, ...), and the chain is then executed recursively, as we did with the original Job implementation.
* | Async: change syntax of callablesDan Vrátil2014-12-11
| | | | | | | | | | | | We now pass our own Async::Future to each task, instead of expecting tasks to return their future. This allows us to re-use the same Future for repeated invocations, like in the Each task.
* | Async: move some definitions out-of-lineDan Vrátil2014-12-11
| |
* | Async: simplify the template magic around In... being empty or voidDan Vrátil2014-12-11
| |
* | Prototype of Async libraryDan Vrátil2014-12-11
| |
* | Fix build with FlatBuffers in non-standard locationDan Vrátil2014-12-11
|/
* shhhAaron Seigo2014-12-11
|
* buildAaron Seigo2014-12-11
|
* keep the databases separateAaron Seigo2014-12-11
| | | | | | | | | | | | | otherwise, to keep all the dbs in the same file we need to call mdb_env_set_maxdbs before mdb_env_open ... and even then the docs say: "Currently a moderate number of slots are cheap but a huge number gets expensive: 7-120 words per transaction, and every mdb_dbi_open() does a linear search of the opened slots." Ugh. that and the mem mapping probably gets screwed oever .. whatever. this way hawd works again :)
* reenable the full testAaron Seigo2014-12-11
|
generated by cgit v1.2.3 (git 2.25.1) at 2025-06-30 17:28:25 +0000