| Commit message (Collapse) | Author | Age | |
|---|---|---|---|
| * | Commit also if we did revision cleanup | Christian Mollekopf | 2016-06-26 |
| | | |||
| * | Use specialpurpose preprocessor which gives us trash folder support. | Christian Mollekopf | 2016-06-26 |
| | | |||
| * | Share special purpose preprocessor implementation. | Christian Mollekopf | 2016-06-25 |
| | | |||
| * | Catch errors | Christian Mollekopf | 2016-06-21 |
| | | |||
| * | Fixed cleanup of date indexes | Christian Mollekopf | 2016-06-20 |
| | | |||
| * | Workaround mdb_put failures | Christian Mollekopf | 2016-06-20 |
| | | | | | | | | I got failures during a sync during the initial mdb_put that stores the db name for verification on open (invalid parameter passed). It seems the verification step actually prevents the problem in the first place. | ||
| * | Avoid unnecessary commits | Christian Mollekopf | 2016-06-20 |
| | | |||
| * | Made the log-level output a lot less noisy | Christian Mollekopf | 2016-06-20 |
| | | |||
| * | Fix cache update on removal | Christian Mollekopf | 2016-06-20 |
| | | |||
| * | Commit all databases in the same interval | Christian Mollekopf | 2016-06-20 |
| | | |||
| * | Don't create a transaction for every appended message. | Christian Mollekopf | 2016-06-20 |
| | | | | | | We create now only one transaction for the whole sync which is also not ideal. | ||
| * | Don't try to remove empty values from the index. | Christian Mollekopf | 2016-06-19 |
| | | | | | Indexes don't allow empty keys, so this will always fail. | ||
| * | Don't use a static for the threadboundary | Christian Mollekopf | 2016-06-19 |
| | | | | | Otherwise we risk calling into already destroyed models. | ||
| * | Moved mailpreprocessors to a shared location | Christian Mollekopf | 2016-06-19 |
| | | |||
| * | A working mailtransport resource | Christian Mollekopf | 2016-06-17 |
| | | |||
| * | If the changereplay failed we have to stop. | Christian Mollekopf | 2016-06-17 |
| | | | | | | Otherwise we would never replay changes that failed because we were offline or so. | ||
| * | Initial version of the mailtransport resource | Christian Mollekopf | 2016-06-17 |
| | | |||
| * | Simpler facade and domaintypeadaptorfactory instantiation | Christian Mollekopf | 2016-06-16 |
| | | |||
| * | Move to trash | Christian Mollekopf | 2016-06-15 |
| | | |||
| * | Automatically fetch children if necessary. | Christian Mollekopf | 2016-06-15 |
| | | | | | | Otherwise a treeview doesn't show the expander until you click on the item (which triggers fetchMore). | ||
| * | Avoid crashing. | Christian Mollekopf | 2016-06-15 |
| | | | | | ...that was a bit too optimistic, sill happening when running kube. | ||
| * | Fixed LocalStorageFacade. | Christian Mollekopf | 2016-06-15 |
| | | | | | | We can't make any assumptions about the lifetime of the facade, so make sure we have all we need in the lambda. | ||
| * | Avoid clearing the liveQuery flag. | Christian Mollekopf | 2016-06-15 |
| | | |||
| * | Remember the revision we loaded in the result set. | Christian Mollekopf | 2016-06-15 |
| | | | | | Otherwise we reload the same entities over and over. | ||
| * | Moved thread-boundary crossing to the model. | Christian Mollekopf | 2016-06-15 |
| | | | | | | That way we avoid any unnecessary queuing for the sync API, and enable fine-tuning in the model code at a later stage. | ||
| * | Synchronous API | Christian Mollekopf | 2016-06-15 |
| | | |||
| * | Prepared trash test | Christian Mollekopf | 2016-06-15 |
| | | |||
| * | Fixed issues found by clang analyzer | Christian Mollekopf | 2016-06-14 |
| | | |||
| * | Don't match invalid properties. | Christian Mollekopf | 2016-06-13 |
| | | |||
| * | Get folder moves to work, and fix the mime message moving. | Christian Mollekopf | 2016-06-13 |
| | | |||
| * | A more convenient preprocessor API | Christian Mollekopf | 2016-06-11 |
| | | |||
| * | Create the drafts folder if necessary and merge it with the source | Christian Mollekopf | 2016-06-10 |
| | | | | | version | ||
| * | Allow preprocessors to inject entities. | Christian Mollekopf | 2016-06-10 |
| | | | | | | Currently only working when creating an entity, the new entity is always preprended in the store. | ||
| * | Moved query logic to EntityReader to make it reusable in the resource. | Christian Mollekopf | 2016-06-09 |
| | | |||
| * | Defined resource capabilities. | Christian Mollekopf | 2016-06-08 |
| | | |||
| * | Avoid extra dependency in public header. | Christian Mollekopf | 2016-06-06 |
| | | |||
| * | A more reasonable limit for revisions | Christian Mollekopf | 2016-06-05 |
| | | |||
| * | A central place to generate uuid's | Christian Mollekopf | 2016-06-05 |
| | | |||
| * | Allow the maildir resource to skip the folder rename test | Christian Mollekopf | 2016-06-05 |
| | | |||
| * | These checks are no longer necessary | Christian Mollekopf | 2016-06-05 |
| | | |||
| * | Skip internal keys by default while scanning. | Christian Mollekopf | 2016-06-05 |
| | | |||
| * | Fixed pipelinetest | Christian Mollekopf | 2016-06-03 |
| | | |||
| * | Track modified properties to detect necessary replay changes | Christian Mollekopf | 2016-06-03 |
| | | |||
| * | A way to retrieve the last revision during changereplay. | Christian Mollekopf | 2016-06-03 |
| | | |||
| * | Using the entityStore to read anything except the current revision is | Christian Mollekopf | 2016-06-03 |
| | | | | | probably a bad idea. | ||
| * | Deal with errors in the change-replay job. | Christian Mollekopf | 2016-06-03 |
| | | |||
| * | Avoid triggering the changereplay directly from the constructor. | Christian Mollekopf | 2016-06-02 |
| | | |||
| * | Load the right revision for changereplay, not just the latest one | Christian Mollekopf | 2016-06-02 |
| | | |||
| * | Reading of individual revisions from the store | Christian Mollekopf | 2016-06-02 |
| | | |||
| * | Non blocking change-replay | Christian Mollekopf | 2016-06-02 |
| | | |||
 |
index : sink | |
| A data access layer handling synchronization, caching and indexing |
| summaryrefslogtreecommitdiffstats |