Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Longer timeout so we have time to attach a debugger | Christian Mollekopf | 2017-01-18 |
| | |||
* | Ignore the component if we use a log context | Christian Mollekopf | 2017-01-18 |
| | |||
* | Debug output | Christian Mollekopf | 2017-01-18 |
| | |||
* | Refactored the changereplay | Christian Mollekopf | 2017-01-18 |
| | | | | | | * use a log context * clearer and simpler control flow * No infinite recursive calling | ||
* | Don't overwrite the first job with the continuation | Christian Mollekopf | 2017-01-18 |
| | |||
* | Ensure resources kill themselves if noone connects. | Christian Mollekopf | 2017-01-17 |
| | |||
* | Also try to send the mail on changereplay. | Christian Mollekopf | 2017-01-17 |
| | |||
* | Avoid unnecessary replay requests. | Christian Mollekopf | 2017-01-17 |
| | | | | A single request will replay until the latest revision. | ||
* | Handle errors without crashing. | Christian Mollekopf | 2017-01-17 |
| | |||
* | Avoid sending the mail twice | Christian Mollekopf | 2017-01-13 |
| | |||
* | Simplification | Christian Mollekopf | 2017-01-12 |
| | |||
* | Fixed build issues with pedantic enabled | Christian Mollekopf | 2017-01-12 |
| | |||
* | The template arguments are no longer required. | Christian Mollekopf | 2017-01-12 |
| | |||
* | syncThen is no longer necessary | Christian Mollekopf | 2017-01-12 |
| | |||
* | Logout after changereplay | Christian Mollekopf | 2017-01-12 |
| | |||
* | Only apply the name-based special purpose detection to top-level | Christian Mollekopf | 2017-01-12 |
| | | | | folders. | ||
* | Flag definitions | Christian Mollekopf | 2017-01-12 |
| | |||
* | Modifications for multiple entities. | Christian Mollekopf | 2017-01-12 |
| | | | | This allows to apply a modification to all entities matching some query. | ||
* | Debug output | Christian Mollekopf | 2017-01-12 |
| | |||
* | Set enabled state from subscription | Christian Mollekopf | 2017-01-11 |
| | |||
* | Logout after synchronization | Christian Mollekopf | 2017-01-10 |
| | |||
* | Maildir always has the full payload. | Christian Mollekopf | 2017-01-10 |
| | |||
* | Simplify jobs | Christian Mollekopf | 2017-01-10 |
| | | | | | | For some reason this also makes the synchronization to throw an error on login failure again. Something with the job error propagation is not quite right. | ||
* | Debug output | Christian Mollekopf | 2017-01-10 |
| | |||
* | Constructor to sync a single entity | Christian Mollekopf | 2017-01-10 |
| | |||
* | Correctly resolve the folder | Christian Mollekopf | 2017-01-10 |
| | |||
* | Remember if we have all of the data or only the headers. | Christian Mollekopf | 2017-01-10 |
| | |||
* | Improved imap mail sync algorithm. | Christian Mollekopf | 2017-01-10 |
| | | | | | | | | | | | | | | | * when requesting individual mails we sync the full content * when requesting individual folders we get 2 weeks of full content + headers for everything else. * when requesting a sync for all folders we only get 2 weeks of full content. Getting the headers for 50k messages takes about 180s on my system with kolabnow (network being the bottleneck), so that's managable. Getting the full content would take in the range of hours. This way we have something to show, and a way to request more data, without making the system overly complex yet. Certainly not the final solution, but a good start. | ||
* | Make sure the path is existing, otherwise the lockfile will not be | Christian Mollekopf | 2017-01-10 |
| | | | | created. | ||
* | Fixed revision cleanup | Christian Mollekopf | 2017-01-10 |
| | |||
* | Create the lockfiles in the data directory. | Christian Mollekopf | 2017-01-09 |
| | | | | The ended up in the home directory. | ||
* | No more normalizedPath | Christian Mollekopf | 2017-01-07 |
| | |||
* | Debug output | Christian Mollekopf | 2017-01-07 |
| | |||
* | Share query syntax | Christian Mollekopf | 2017-01-07 |
| | |||
* | Debug output | Christian Mollekopf | 2017-01-06 |
| | |||
* | Avoid unnecessary warnings | Christian Mollekopf | 2017-01-05 |
| | |||
* | Sinksh | Christian Mollekopf | 2017-01-05 |
| | |||
* | Remove resources and identities with the account | Christian Mollekopf | 2017-01-05 |
| | |||
* | sinksh account removal | Christian Mollekopf | 2017-01-05 |
| | |||
* | Don't turn modifications into creations | Christian Mollekopf | 2017-01-04 |
| | |||
* | Move the mail between folders and update specialpurpose accordingly. | Christian Mollekopf | 2017-01-03 |
| | |||
* | Comparison operator for ApplicationDomainType | Christian Mollekopf | 2017-01-03 |
| | |||
* | React to removals in reduced live-query. | Christian Mollekopf | 2017-01-03 |
| | |||
* | Debug messages | Christian Mollekopf | 2016-12-23 |
| | |||
* | Avoid fetching again if a fetch is already in progress. | Christian Mollekopf | 2016-12-23 |
| | |||
* | Better debug output | Christian Mollekopf | 2016-12-23 |
| | |||
* | Use the resourcefacade to query for resources. | Christian Mollekopf | 2016-12-23 |
| | | | | | | Otherwise we'd have to maintain two query paths, and getResource already broke because we didn't retrieve the capabilities. | ||
* | More Log::Context | Christian Mollekopf | 2016-12-22 |
| | |||
* | Introduced Log::Context | Christian Mollekopf | 2016-12-22 |
| | | | | | | | To have hierarchical debug output we have to pass around something at run-time, there is no reasonable alternative. Log::Context provides the identifier to do just that and largely replaces the debug component idea. | ||
* | Better query debug output | Christian Mollekopf | 2016-12-22 |
| |