libera/#maemo-leste/ Tuesday, 2021-08-24

freemangordonWizzup: latest abook master saves to personal addressbook, evolution correctly sees the created contact :)00:57
Wizzupthat's sweet01:01
WizzupI guess we can sync as well using syncevo01:01
Wizzup(I can sync my fremantle contacts at least)01:01
Wizzupexciting :)01:43
lelMerlijnWajer synchronize a pull request: https://github.com/maemo-leste/connui-internet/pull/2 (WIP: add providers tab to advanced iap settings)01:45
Wizzupparazyd: btw, if the service_id becomes an unknown value to the icd2 provider (i.e. a config is deleted from tor cpa), then the network will no longer be marked as service provider01:47
WizzupSince the libicd plugin will reject it based on the id not being supported, and then UI acts appropriately01:47
Wizzupso we actually don't need to worry about that use case :)01:47
parazydWizzup: That's pretty useful :)07:21
Wizzupfreemangordon: any idea if text added to a gtk combo box (marked const char*) is copied to the widget itself?07:31
WizzupI would hope so, but not sure07:31
parazydWizzup: In what sense?07:34
Wizzuplooks like it is fine07:34
Wizzupparazyd: I wasn't sure if I had to g_strdup() strings I add with gtk_combo_box_append_text07:34
Wizzupbut it looks I don't need to07:34
parazydNo you don't really07:35
Wizzupas in, I didn't allocate them (gconf did)07:35
Wizzupyeah07:35
Wizzupit makes sense I guess07:35
parazydmhm07:35
lelMerlijnWajer synchronize a pull request: https://github.com/maemo-leste/connui-internet/pull/2 (WIP: add providers tab to advanced iap settings)07:52
Wizzupfreemangordon: ^^ if you have a couple of minutes07:53
lelMerlijnWajer edited a pull request: https://github.com/maemo-leste/connui-internet/pull/2 (Add providers tab to advanced iap settings)07:55
lelMerlijnWajer synchronize a pull request: https://github.com/maemo-leste/connui-internet/pull/2 (Add providers tab to advanced iap settings)07:57
sicelofreemangordon: yay! that's beautiful08:23
freemangordonWizzup: please fix block brackets :)13:46
freemangordonactually the whole PR styling needs to be fixed13:46
Wizzupfreemangordon: ok, apart from indent?13:47
freemangordonsee https://github.com/maemo-leste/connui-internet/pull/2/commits/feb8316ca434d858e3c0b9381b7526dafa33bd98#diff-f5f00673365a45906cdb8c63dedef131c800df08c9b1d70e940a7cd933742fdcR67 and https://github.com/maemo-leste/connui-internet/pull/2/commits/feb8316ca434d858e3c0b9381b7526dafa33bd98#diff-f5f00673365a45906cdb8c63dedef131c800df08c9b1d70e940a7cd933742fdcR7713:48
freemangordonthe second has the function name on the same line as function type13:48
freemangordonbrackets have to be on a separate line13:48
Wizzupyeah, I'll make it match yours13:48
freemangordonmhm13:49
Wizzupbut I mean apart from indent13:49
Wizzupsorry, we're just used to different things :D13:49
Wizzupbut it's absolutely np, I'll fix it13:49
freemangordonone question: why gtk_combo_box_new_text() ?13:49
WizzupAs opposed to / instead of?13:49
freemangordonI am almost sure some hildon widget is used instead13:49
freemangordonlemme try to find it13:49
WizzupI think I took this from other combo boxes from connui13:50
freemangordonhmm, could be13:50
* freemangordon checks13:50
WizzupThe only limitation it has that I don't want to fix right now is that when you switch providers to something else and then back to what it was, it won't "remember" the service id (i.e. provider specific config) you selected initially, but I think that shouldn't be a big deal13:51
uvosoh btw please rewiev the mce and h-d pr13:51
freemangordonuvos: will do13:52
freemangordonWizzup: I am almost sure hildon_picker_button/hildon_touch_selector is used elsewhere13:54
Wizzupdo you have an example? I took this from the wlan tab13:54
Wizzupin advanced13:55
freemangordonoh, right13:55
freemangordonpickers are on the main 'page'13:55
freemangordon"Internet connections" that is13:56
freemangordonso yeah, combobox is fine13:56
freemangordonLMK when you fix the coding style13:56
freemangordonuvos: reviewing those will take more time I have ATM, will try to do h-d at least tomorrow13:57
uvosmce is way more important13:57
uvosis blocking everything13:57
freemangordonok13:58
uvosthe hildon pr just replaces a hack i wrote some time ago with soemthing better13:58
freemangordonI am not sure how usefull I will be, given I never used cmake :)13:58
uvosso it dosent change any real functionality13:58
freemangordonbut will do my best13:58
uvosthank you13:58
freemangordonWizzup: also, please keep 80 chars per line limit14:17
uvosfreemangordon: maybe you want to way in on https://github.com/maemo-leste/bugtracker/issues/32914:18
uvosfreemangordon: at least for new things the clear consensus so far was linux style14:18
uvos(which includes relaxed 80 col  limit since some time)14:19
freemangordonuvos: this is not new14:19
Wizzuphe did weigh in on it uvos :p14:19
Wizzuphttps://github.com/maemo-leste/bugtracker/issues/329#issuecomment-59478740514:19
uvoswell if he has nothing further to add we need to close 329 by voteing on a style or something14:20
uvossomething to put into guidelines14:20
WizzupI think the concensus we have is that we stick to whatever the repo has14:20
Wizzupand if the maintainer wants to switch, they can do that14:20
uvosthats not how 329 reads14:21
freemangordon:nod:14:21
Wizzupmaybe we can narrow it down to like two or three styles14:21
Wizzupand then make it clear which uses which14:21
uvosand besides a prefered style is sensible14:21
Wizzupnot if we can't agree on one :P14:21
uvoseven if it comes with a dont change coherent repos preface14:21
WizzupI'm fine changing my PR for fmg's style, but I won't use it for new repos (only I) work on14:21
bencoh:)14:22
freemangordonwell, at the end of the day I think it comes to which style is the easiest to read14:22
uvosi dont think there is a style that is easiest to read14:22
uvosthat just whatever you are used to14:22
WizzupI will switch to your style if you switch to my editor ;)14:23
freemangordonnot really14:23
Wizzup(joke)14:23
bencohacctually a 80-ish limit really is easier to read14:23
freemangordonWizzup: and give up on QtCreator?14:23
Wizzup:p14:23
uvosi dissagree on modern displays that just creates whitespace14:23
bencohit basically means your eyes to need to move too far14:23
freemangordonbencoh: :nod:14:23
uvosto many line breaks is bad too14:23
WizzupI just want to get to a sitation where we can use tooling with a set of indentation rules14:23
bencohand that you can have several opened files in columns14:23
uvossure14:23
Wizzupand then you can run it when you work, revert it with the standard rules, and commit code14:23
uvosbut 80 coll is to narrow imo14:24
uvos110 at least14:24
uvos200 is def to long14:24
uvosetc14:24
bencoh(s/to need/don't need/ )14:24
freemangordonuvos: no matter the display, eyes still have limit on the minimum font size that's ok for them14:24
freemangordonand that changes with aging14:24
bencohuvos: oh, right, the limit itself is obviously a matter of taste, and 80 is legacy14:24
uvosright14:24
uvosits very mutch a matter of taste14:25
freemangordon80 chars limit forces you to beautify the code you write, more or less14:25
bencohrelaxed-limit in linux is 80 soft-limit, ~100 hard-limit now, right?14:25
uvos11014:25
bencohfreemangordon: indeed :)14:25
uvossoft ish at 11014:25
freemangordonthat's why I prefer to stick to it14:25
uvos80 is totaly gohne from docs iirc14:25
bencohoh? sad14:26
bencohtbh I always wonder how people manage to keep under 80 with tabsize 8 (I use 4)14:26
uvoswel linux uses tab14:27
uvosso that counts as one14:27
freemangordonwhich is a huge waste of space IMO14:27
Wizzupmaybe we can pick 2 rules and settle on those, and make some bash aliases or helper scripts to let people switch from styles before they commit?14:27
uvostab?14:27
bencoh8 ? definitely a waste14:27
Wizzupas much as I like discussing code formatting14:27
freemangordonyes14:27
WizzupI don't think this leading us anywhere14:27
freemangordonWizzup: ok14:27
bencohWizzup: didn't you toy with indent (the tool) at some point?14:27
uvoshow is tab a waste of space you can set it to any with you like14:27
freemangordonWizzup: we are really just chattering IIUC14:27
Wizzupthe one thing my above suggestion requires is if we pick 2 or 3 cdoe styles, we also apply them to a repo, so that subsequent runs do not generate daditional noise14:27
Wizzupok14:27
Wizzupthen keep going :P14:27
freemangordon:)14:28
bencoh:]14:28
uvosbencoh: indent -linux -l 120 had some consus on the bug above14:28
bencohyeah, having a few reference styles would be a good idea (and I'm all with "stick-to-the-original-repo-choice" thing)14:28
freemangordonWizzup: but, seriously, what we *must* have as a rule is - do PRs in coding style of the repo you're working with14:28
Wizzupfreemangordon: yes, but I want the style to be strictly defined14:29
freemangordonah, I see14:29
uvossometimes the codeing style the repo works with is hard to decern14:29
Wizzupso if you can find a set of indent(1) flags for your style, that would be nice14:29
uvoswe need to fix that too14:29
uvosh-d is an offender14:29
freemangordonh-d is...14:29
freemangordonWizzup: I am using QtCreator, not really sure how to extract the style from there14:30
Wizzupso my suggestion was, let's try to define 2 or 3 sets of indent flags14:30
Wizzupfreemangordon: maybe try running indent with some flags until it matches14:30
Wizzupit doesn't have to *now*, but it would make our lives easier14:30
freemangordonok14:30
bencohfreemangordon: I didn't think you would use it, interesting (actually I think you told me once already, but still surprised)14:30
freemangordonthat's the best C/C++ editor IMO14:31
freemangordonWizzup: I can export its settings in some qt3.xml file14:31
bencohand yeah, we might not run the full project through indent, but at least people pushing commits would have a way to fix indent in some automatic manner before commiting14:31
Wizzupbencoh: yes14:31
Wizzupalso we can fix the entire project ina single commit and make git blame ignore it14:32
bencohhow so? git log/blame/whatever read some config file stored in repository?14:32
Wizzupyes14:33
Wizzupsee the issue linked14:33
bencohoh ... it's semi-automatic then14:33
freemangordonhttps://pastebin.com/TnfF0zAD14:34
bencohinteresting14:34
bencoh(but only works for blame)14:34
bencoh(I guess?)14:34
Wizzupbencoh: I think so14:34
WizzupThat's also really the only thing that matters14:34
* freemangordon tries indent14:35
freemangordonWizzup: can't make it work properly17:12
freemangordonit inserts space after * on half of the occasions :(17:13
freemangordonlike:17:18
freemangordon-osso_abook_contact_dispose(GObject *object)17:18
freemangordon+osso_abook_contact_dispose(GObject * object)17:18
Wizzupfreemangordon: maybe there is a flag for that?17:24
freemangordonthere is17:24
freemangordonand it is set17:24
freemangordon-par17:24
freemangordonbut it is ignored17:25
WizzupI don't see that flag in my man page17:25
freemangordon--pointer-align-right17:26
bencohI don't see that flag either, funny17:27
freemangordonin leste it is there17:28
Wizzupfreemangordon: it should be the default behaviour17:29
Wizzupit looks like17:29
freemangordonmhm17:30
freemangordonbut still a space is insterted afte *17:30
freemangordonncs seems to be ignored too17:33
freemangordon-ncs that is17:33
freemangordonthis is what I tried with:17:38
freemangordonindent -di0 -bad -bap -lp -bli0 -l80 -ncs -npcs -nprs -ts8 -nbs -i2 -par -ip0 -as -slc -nut -hnl -ci4 osso-abook-contact.c17:38
freemangordonand it breaks it badly17:38
uvosuff that style is impossible to read18:17
uvos(for me anyhow)18:17
Wizzupwhich is why we can automate it away18:17
uvoswierd about the *18:18
uvosi dont think i have seen that error18:18
tmlind_uvos: fyi, just pushed out https://github.com/tmlind/linux/commits/droid4-pending-v5.14 and https://github.com/tmlind/linux_openpvrsgx/commits/droid4-pending-pvr-omapdrm-v5.1421:47
tmlind_seems to work for me based on brief testing21:47
* tmlind_ sleepy, later22:00

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!