| *** Colgate_ is now known as Colgate | 03:08 | |
| *** ecloud is now known as ecloud_wfh | 07:26 | |
| dcaliste | Hello chriadam, just a word to say that I remember why I've prefered the etag modification for the CalDAV fix ;) Please hold the merge, the detached-by-sync tag version has a bug. I'm going to add yet another test (on up-sync this time) and correct it while not going back to etag mangling. | 07:52 |
|---|---|---|
| chriadam | dcaliste: hi :-) | 07:52 |
| chriadam | sure, I haven't had a chance to review the latest change anyway, I was planning to do that tomorrow | 07:53 |
| chriadam | thanks very much for looking into that one, by the way. sorry I took so long to respond. | 07:53 |
| dcaliste | The problem is for UI detached exception (thus local addition) that are always up-sync now, because the detached-by-sync tag or equivalent is not added when upsyncing, while with the etag mangling it was done automatically ;) | 07:53 |
| dcaliste | No problem, you was quick, don't worry, I'm not waiting for reply within hours, days are fine! | 07:54 |
| dcaliste | s/you was/you were/ ouch :( | 07:54 |
| dcaliste | This issue of up-syncing can be solved easily by reverting the suppression of the search in additions list, but since we have this tagging, there should be a nice way to remove this SQL query and this search in the addition list for every synced events. | 07:56 |
| chriadam | what do you mean by "the detached-by-sync tag .. is not added when upsyncing" ? I had assumed that was added when we downsynced it | 07:56 |
| chriadam | and just checked before we attempted to upsync | 07:57 |
| dcaliste | Yes, this tag is added when down-syncing and indeed checked in delta calculation. | 07:57 |
| chriadam | right | 07:57 |
| dcaliste | But I've removed the search in addition list. | 07:57 |
| dcaliste | Because it was not necessary with the etag mangling. | 07:58 |
| dcaliste | And it saved one SQL query and a for loop. | 07:58 |
| chriadam | ah right I see | 07:59 |
| chriadam | yes, it is a shame that linear search through memory is required for the addition suppression :-( | 07:59 |
| chriadam | but, probably not too big a problem imo | 07:59 |
| dcaliste | I'm sure that with tagging we can avoid that. I don't like the tagging solution but I cannot see any other way (based on time stamps???) to differentiate local addition from down-sync additions. | 08:00 |
| dcaliste | Thus, since we have tagging for good I'm sure we can remove this SQL query and search in addition list. | 08:00 |
| chriadam | I don't see any other way either. if you can avoid the overhead, great :-) | 08:01 |
| dcaliste | Yes and as a bonus, it make the delta easier to read in my opinion. Well, I'm going to add a test for up-sync like the one I've added for down-sync and modify a bit the tagging mechanism to work in both cases. | 08:02 |
| dcaliste | Otherwise, I've seen work again on C API for secrets, nice. With GObject, even better ;) I'll try to find time to help for review. | 08:04 |
| chriadam | dcaliste: that would be appreciated, I have no idea with glib ;-) | 08:08 |
| dcaliste | I'm using GLib / GObject at work so no problem. Particularly, with introspection, you have JS and Python for free with the exact same API without doing anything more. It's really powerful. | 08:10 |
| *** feodoran_ is now known as feodoran | 10:40 | |
| x2s | monich: still interested in the Xperia looses mobile network thing? I think the Update 2018.11.22 part of this could be of interest for you: https://together.jolla.com/question/170394/xperia-x-loss-of-mobile-network/?answer=193030#post-id-193030 | 15:44 |
| monich | x2s: that's quite a reading, thanks! | 15:47 |
| *** DennisRoczek|afk is now known as DennisRoczek | 21:12 | |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!