| freemangordon | Wizzup: hard to say :) | 08:33 |
|---|---|---|
| freemangordon | libelogind is | 08:33 |
| freemangordon | ugh | 08:38 |
| freemangordon | it tries modesetting | 08:38 |
| freemangordon | I have removed leste-config-mapphone ;) | 08:39 |
| freemangordon | ugh: /usr/bin/osso-abook-home-applet: No such file or directory | 08:50 |
| Wizzup | freemangordon: how did you dist upgrade | 09:56 |
| Wizzup | freemangordon: was it missing? | 09:57 |
| Wizzup | freemangordon: do you know if we have a n900 with bg layout somewhere | 13:03 |
| Wizzup | hw | 13:03 |
| Wizzup | or russian I guess | 13:04 |
| freemangordon | the is no bg hw layout | 13:33 |
| Wizzup | maybe a russian one then | 13:33 |
| Wizzup | doesn't really matter, I can test it without I guses | 13:34 |
| sicelo | maybe the Geany issue is GTK3? do keyboard mappings work fine on gtk3? | 13:43 |
| Wizzup | are you talking about https://github.com/maemo-leste/bugtracker/issues/688 ? | 13:43 |
| sicelo | yes | 13:45 |
| k1r1t0 | Wizzup I have a device with a Russian layout, what should I check? | 13:47 |
| Wizzup | not something to check right now, but later :) | 13:48 |
| freemangordon | Wizzup: I have gtk3 him plugin working (almost) | 15:09 |
| freemangordon | so, if someone is interested in, I can share teh for her to finish | 15:09 |
| freemangordon | Wizzup: still the issue remains with supported layouts (keymaps parsing) | 15:11 |
| Wizzup | freemangordon: cool, in any case, sharing that somewhere in a branch is useful | 15:17 |
| Wizzup | freemangordon: hm, what is the challenge here exactly? | 15:17 |
| freemangordon | neither chromium nor FF were properly supporting gtk3 IM plugins in beowulf :) | 15:38 |
| freemangordon | and the chalange is - how to know what layouts are supported by the current device based on kbd model | 15:38 |
| freemangordon | if you remember we were unable to find a way to get that info from X or otherwise | 15:39 |
| Wizzup | but this applies for gtk2 too, no? | 15:39 |
| freemangordon | it seems a complicated parser is needed | 15:39 |
| freemangordon | yes | 15:39 |
| freemangordon | there are 2 separate ussues | 15:39 |
| freemangordon | *issues | 15:39 |
| Wizzup | ok, so this is not a gtk3 specific problem, I got confused | 15:40 |
| Wizzup | what needs work for the gtk3 plugin? | 15:40 |
| Wizzup | freemangordon: btw looks like switching to and from russian works well for the guy using the maemo switching codew | 15:40 |
| freemangordon | yes, because n900/russion is explicitly supported | 15:40 |
| freemangordon | *russian | 15:40 |
| freemangordon | https://github.com/maemo-leste/hildon-input-method-plugins/blob/master/hildon-im-keyboard-monitor/hildon-im-keyboard-monitor.c#L479 | 15:41 |
| freemangordon | (not my comment there ;) ) | 15:41 |
| freemangordon | if you look at what it does, on n900 Russian is a kind of hack (according to my limited knowledge on how keymaps work) | 15:42 |
| freemangordon | but again, that's not related to gtk | 15:43 |
| freemangordon | for gtk3 we obviously need IM plugin | 15:43 |
| bencoh | I'm about to say something silly, but what about leveraging ibus? | 15:45 |
| bencoh | and add a (custom?) touch-friendly virtual keyboard to ibus? | 15:45 |
| bencoh | it would work pretty much everywhere, and allow changing the hw keyboard layout through ibus as well | 15:46 |
| freemangordon | how we will show it on edit field entered? | 15:46 |
| bencoh | hm? | 15:46 |
| bencoh | you're basically asking if ibus can make it pop upon clicking a text field? | 15:47 |
| freemangordon | mhm | 15:47 |
| bencoh | no idea | 15:47 |
| freemangordon | forgive my ignorance | 15:47 |
| bencoh | I dunno either | 15:47 |
| bencoh | :) | 15:47 |
| freemangordon | I don;t really know how all this is supposed to work | 15:48 |
| freemangordon | (ibus) | 15:48 |
| Wizzup | I think it would be better to improve him first | 15:48 |
| Wizzup | h-i-m that is | 15:49 |
| freemangordon | right :) | 15:49 |
| bencoh | I usually ibus on various computers, both for hw layout (latin and non-latin) and semi-virtual input (like japanese / anthy) | 15:49 |
| bencoh | and the anthy module has some kind of UI | 15:49 |
| freemangordon | bencoh: and you have a window with that kbd somewhere all the time, no? | 15:49 |
| bencoh | for japanese? no, basically you change the input mode, start typing, and it changes your input on-the-fly, and allow you to choose the correct input (hiragana/katakana/kanji, japanese stuff) | 15:50 |
| freemangordon | I was talking about vkb | 15:51 |
| bencoh | vkb I haven't tried with ibus, that's what I think we would have to add | 15:51 |
| freemangordon | which brings us back to my initial question - how does it know when to pop :) | 15:52 |
| freemangordon | brb | 15:52 |
| Wizzup | freemangordon: if I can help with the gtk3 im, lmk | 16:26 |
| Wizzup | I'd love to get that in place some time soon, it's a big thing for us | 16:26 |
| bencoh | (and re: vkb implemented with ibus, looks like it's not really a thing, quite the opposite, they kinda expect 3rd-party non-ibus vkb to be used if anything) | 16:31 |
| bencoh | (someone proposed adding a virtual vs physical keyboard hint to ibus) | 16:31 |
| bencoh | (so just ignore my first question :) ) | 16:32 |
| uvos | raising vkb on field click in all toolkits is prosible via at-spi | 16:58 |
| uvos | https://github.com/maemo-leste/hildon-input-method/pull/7 | 16:58 |
| Wizzup | darnit, was writing him a msg | 17:00 |
| Wizzup | :D | 17:00 |
| uvos | sphone dosent do video calls (and imo video calls should probubly be handed by a different application since the required ui is so different) | 17:02 |
| uvos | sphone dosent do multi party calls, atho it sorta dose, you can have any number of call objects at a time, and the ui will present a list of calls. really this is intended to support call holding, but nothing prevents a backend haveing multiple endpoints active this way in a conference style call | 17:04 |
| bencoh | uvos: "for instance clicking on a field will raise the vkb only once, after closing to vkb again you need to click somewhere else before you can raise it again on the same field" sounds good to me | 17:55 |
| bencoh | since one can always raise the vkb manually as well | 17:55 |
| bencoh | so user might want to select (copypaste) something from said field without vkb getting in the way | 17:55 |
| uvos | sure but that pr is still more of a proof of concept, there are a lot of things it still needs | 18:26 |
| uvos | atspi dose for instances allow you to get the content of an input field | 18:26 |
| uvos | and this would allow you to then edit in in him | 18:26 |
| uvos | but this is not supported | 18:26 |
| buZz | found another convert :) | 22:15 |
| Wizzup | bencoh: cool :) | 23:04 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!