| dcaliste | Good morning, pvuorela. | 08:00 |
|---|---|---|
| pvuorela | dcaliste: good morning! | 08:03 |
| dcaliste | I think the discussion may not be long this morning : just a recap of the existing MRs. | 08:04 |
| dcaliste | I've currently restarted to work on implementing in mKCal an async SqQlite driver. | 08:04 |
| dcaliste | The more I work on it, the more I think the drop of ::Ptr for Notebooks is not necessary. | 08:05 |
| pvuorela | alright. how's that going? | 08:05 |
| dcaliste | I think the async class will be ready for a first review later this week or next week. | 08:05 |
| pvuorela | nice | 08:06 |
| dcaliste | You can see the work going on in the cleanup branch in my fork on Github. | 08:06 |
| dcaliste | I've extracted the Private class from SqliteStorage class and put it in a dedicated file sqlitestorage_p.h and .cpp. | 08:06 |
| dcaliste | Like that the sync or the async share the same underlying implementation. | 08:07 |
| dcaliste | It requires in the _p class to move away from Ptr to avoid race issue and ownershiop issues between the main thread and the worker thread, as we already discuss. | 08:08 |
| dcaliste | The drop of ::Ptr for notebooks (and incidences) is thus happening there but not in the main thread API. | 08:08 |
| pvuorela | hm, ok. | 08:09 |
| dcaliste | There are copies of incidences on save, but ownership transfer on load. | 08:09 |
| dcaliste | I don't think I can avoid the copy on save with a true async API without locking mechanism. | 08:10 |
| dcaliste | And I don't want to introduce locks. | 08:10 |
| dcaliste | But in my opinion, it's not a big deal since the async API is for UI purpose and we never save a large bunch of incidences at the same time. | 08:10 |
| dcaliste | On the contrary for sync plugins, where the sync API will still be fine. | 08:11 |
| dcaliste | You'll have a better idea I think when I'll submit the PR. | 08:12 |
| dcaliste | So, I would say that the pending PRs on ::Ptr drop for notebooks should be on hold at the moment. | 08:13 |
| pvuorela | sure | 08:13 |
| dcaliste | We can decide later when the async API will be added if we keep them (and rebase them) or if we close them. | 08:13 |
| dcaliste | As I said, I think I don't have any additional points at the moment for discussion. I'll ping you when the async PR is ready. Thank you for your help pvuorela ! | 08:25 |
| pvuorela | yea, me neither. commented on one old pr but that's not much to discuss here. | 08:28 |
| pvuorela | thanks again, looking forward for the PR. | 08:28 |
| *** amccarthy is now known as Guest1553 | 10:25 | |
| *** amccarthy_ is now known as amccarthy | 10:25 | |
| attah | Any idea what dark incantation i need to make a WebView show as a miniature on the cover? | 17:58 |
| attah | Currently i have cover:null, but for whatever reason, when i release the minimizing swipe the cover goes white | 18:00 |
| attah | (up until then it is fine) | 18:00 |
| attah | ...and i just double-checked, it is not a page-specific problem or related to my html/javascript injection :( | 18:07 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!