#taskwarrior

/

      • swimmer joined the channel
      • diyfupeco has quit
      • diyfupeco joined the channel
      • lydgate has quit
      • lydgate joined the channel
      • tbabej
        pbecking1: Additionally, I think the _ids, _uuids, ids, uuids, _unique, _projets, _tags reports should not respect context as well. They're not meant for human consumption - used in CmdCompletion and other user scripts, and imho, introduction of context application there would break stuff.
      • Also, I'm already annoyed by my command completion not being able to complete stuff out of the context. It's pretty irritating if you're used to stuff being completed.
      • ^patyx7 joined the channel
      • gour joined the channel
      • gour has quit
      • gour joined the channel
      • Haudegen has quit
      • Haudegen joined the channel
      • sjoshi joined the channel
      • Vargos joined the channel
      • Pest has quit
      • Haudegen has quit
      • Haudegen joined the channel
      • pbecking1
        tbabej: Agree (hepers)
      • viq has quit
      • tbabej
        pbecking1: That's good, because patches are in your INBOX.
      • akendo joined the channel
      • Haudegen joined the channel
      • viq joined the channel
      • MZAWeb_ joined the channel
      • sjoshi has quit
      • pbecking1
        tbabej: Thank you. Will apply later. Will push sometime in the near future.
      • mmarker joined the channel
      • tbabej
        pbecking1: Also, there's patch 0113 which is not merged yet. Just a reminder.
      • pbecking1
        tbabej: Actually 0013 is applied and not pushed.
      • tbabej: I have a few commits (to say the least) that are going to be pushed to merge Date and ISO8601.
      • tbabej
        pbecking1: Sooner you push, sooner we play.
      • Actuallly I suggest we sit on the refactored version at least for a week before releasing. Or do another beta.
      • pbecking1
        tbabej: Exactly, this has to be right.
      • rivarun joined the channel
      • rivarun
        hello. i have a task that says "12 days" on a "S" column and its background is blue. what do these two mean?
      • kalessin has quit
      • tbabej
        rivarun: Depends on your color configuration (theme).
      • rivarun
        unfortunately those colors are not human readable, but nevermind the color. it's the S that i can't find in the documentation
      • tbabej
        rivarun: The headers of each report are defined by report.X.labels and report.X.columns variables
      • rivarun: So, it's report specific, and if you personalize it, it can be literally any mapping header <-> column
      • kalessin joined the channel
      • pbecking1
        rivarun: Which report did you run?
      • tbabej
        rivarun: Nevertheless, 'S' stands for scheduled as far as I can see
      • rivarun
        pbecking1: just `task +work`
      • tbabej
        rivarun: That's invoking the default command.
      • rivarun
        tbabej: the "default" command?
      • tbabej
        Which is, in your case, I believe, one of 'ls', 'next', 'overdue' or 'ready'.
      • rivarun: Yes :) look it up in "man taskrc".
      • pbecking1
        tbabej: sorry, didn't mean to butt in. I failed to read ahead.
      • seschwar joined the channel
      • tbabej
        pbecking1: Not a problem, I bet rivarun enjoyed this ultimate level of support :)
      • Haudegen joined the channel
      • rivarun: Long story short, the 12 days in S (scheduled) column means the task is scheduled in 12 days.
      • rivarun
        tbabej: i'm still confused. i couldn't find "S" in `man taskrc`, nor `task columns`. can i query `task config` for report.X.labels?
      • also yes, my default would be `task next`
      • pbecking1
        rivarun: task show report.next.labels
      • bqf
        rivarun: there's lots of "S"s in that manpage. i would say about 70.
      • aw, now pbecking1 answered the question before i could add the useful part, and i just look like a dick. oh well.
      • Rafal92 joined the channel
      • Rafal92
        hello, what is a procedure for issuing a pull request?
      • pbecking1
        Rafal92: Send me a patch.
      • rivarun
        pbecking1: that's good, except it still doesn't show what S maps to
      • pbecking1
        rivarun: Right. Now compare positionally to: task show report.next.columns
      • Rafal92
        pbecking1: via what? irc?
      • rivarun
        pbecking1: now that's the answer, thank you very much
      • pbecking1
        Rafal92: Good old fashioned email. Or telegram or postcard.
      • Rafal92: Send to support@taskwarrior.org
      • bleader
        postcard don't get delivered to email addresses :'(
      • pbecking1
        bleader: They could though - the postal service could scan both sides and email.
      • Then burn the thing for fuel.
      • bleader
        heh
      • Rafal92
        pbecking1: PR sent, this is a really simple fix intended for systems without timegm. You guys seem to forget to include util.h in Nibbler.cpp and ISO8601.cpp thus compilation fails on some systems
      • pbecking1
        Rafal92: Thank you. Which systems?
      • bqf
        Rafal92: what systems are those? all Taskwarrior commits are built on arch, cygwin, debian, fedora, freebsd, osx, and ubuntu right now.
      • Rafal92
        pbecking1, bqf: android-ndk (r10e, platforms: 19, 21)
      • pbecking1, bqf: I really don't know how to test this apart from creating a new build target
      • pbecking1, bqf: in fact you already head all the implementation apart from including those header where needed :-)
      • *had
      • bqf
        Rafal92: are you working on an android client by any chance? because if so, prepare to be hugged to death by the Taskwarrior community. ;)
      • pbecking1
        Rafal92: Thanks, it's a good patch. But we have no way of automating that testing (currently), so the only reason I can accept it is because it's a #include, and not a code change, ie provably safe.
      • Rafal92
        bqf: indeed I am, I want to run have taskwarrior as a native binary in the internal storage and call it as needed; reading will be from task export command :-)
      • pbecking1: I understand; I'm going to create a fork on github during a weekend and configure travis ci to build using android-ndk so that I have everything under control
      • bqf
        Rafal92: yeah, good approach. be sure to use the current 2.5.0 dev branch, import is much improved for this exact use case there.
      • Rafal92: and if you manage to build Taskwarrior with gnutls support for android, tshirtman will want to speak to you ;)
      • Rafal92
        bqf: thanks, I'll check it
      • bqf: first I have to build gnutls from source for my pc then see what's needed on android
      • bqf
        Rafal92: you'll want to speak to tshirtman then, he has most of that figured out by now. i can send you the build log i have for gnutls on a desktop machine, but he has most of it ready for android NDK already.
      • i think he's still stuck on packaging the right libraries. which shouldn't be a huge problem.
      • Rafal92
        bqf: that's great, I will talked with him then as well; he may be interested in taking part in the development
      • jrabbit
        Rafal92: or we can use taskc ;)
      • gour has quit
      • gour joined the channel
      • tbabej
        Rafal92: you and tshirtman should meet then
      • Rafal92: there's already some effort ongoing for building task on android
      • Rafal92
        tbadej: building without gnutls is quite easy though, with that library is quite the opposite
      • tbabej: and this is quite a milestone since syncing with taskserver is a must-have for an android app
      • jrabbit
        this is why im pushing my pure python taskd library :P
      • but using tw might be simpler if the dang thing ever works
      • tbabej
        Rafal92: I know tshirtman spent some long nights fighting that :)
      • Rafal92: but the effort seems currently out of steam, so if you can chime in, you would be indeed seen as a hero
      • just wait for djp to appear
      • jrabbit
        Rafal92: well no matter the sync method we should flesh out an interface or something?
      • Rafal92: are you on the py/android or pure android front?
      • Rafal92
        jrabbit: I'd like it to be as much android as possible
      • and frankly speaking I can start development right away and add sync later
      • I didn't know that android app is so much desired
      • it was just an idea at work :-)
      • jrabbit: what I see taskc is in fact a proxy for taskserver
      • jrabbit: reimplementing this in Java would make gnutls not required probably as android has some tools to that stuff
      • jrabbit
        im sure
      • Rafal92: if you android a bunch maybe its better to start with mirakel?
      • they have partial taskserver support that was removed
      • Rafal92
        jrabbit: I've tried mirakel and this was quite a poor quality software
      • CI has been failing for one year
      • jrabbit
        well they did remove the sync for a reason :P
      • Rafal92
        jrabbit: yeah, they used own database for tasks and then translate them to taskwarrior format
      • my approach is to have taskwarrior as a database and implement some proxy (adapter) making system calls
      • so all taskwarrior functionality should be available, only android app needs to support it
      • demure joined the channel
      • sliptoni` has quit
      • sliptonic joined the channel
      • tbabej
        Rafal92: Yep, I looked into Mirakel and it was quite a mess, I don't wonder they started slimming down the app.
      • demure
        I don't suppose there is a decent option besides mirakel for android?
      • tbabej
        demure: No, hence the current discussion.
      • demure
        *nod*
      • I probably joined part way in
      • tbabej
        Rafal92: The initial plan for TW client on android we had is to build the task, then use kivy to build a Python UI on top of it. (given there are already good python libraries that work on top of task).
      • jrabbit
        Rafal92: you don't need tw you can just have your own pending and tx files but yeah
      • tbabej
        Rafal92: However, if you plan to build a native android app, you will have my help too.
      • jrabbit: That's true, but there's quite a lot of logic you get 'for free' by having task under the hood.
      • Rafal92
        tbabej: exactly, once project is started I will get you all know :-)
      • jrabbit
        if we can solidify behind one project or a group of projects or whatever we can ask mirakel to point to us
      • right now it's just like "sorry no more taskd"
      • tbabej
        jrabbit: Yep, I support solidifying behind a small number of higher quality projects. Currently, the tools page has quite a lot tools, hard to tell what is maintained and what not, etc.
      • fling has quit
      • djp
        tbabej: what's the moniker for the "core" or "supported" projects?
      • tbabej
        djp: Feel free to invent it.
      • djp: I hear you're good at that.
      • djp
        we'll see about that
      • MZAWeb_ has quit
      • nitefall joined the channel
      • tshirtman
        Rafal92: my work can be seen in the task branch of python-for-android's project on github (kivy/python-for-android)
      • most things have been resolved for building, i just need to get back to it to change the .so generated names, for android to pick them up at packaging time, probably a small problem, but i've been busy with other things (like, work)
      • Haudegen has quit
      • tbabej
        tshirtman: oh, work, the ultimate time waster
      • pbecking1
        Here comes an opinion:
      • bqf
        burgers are awesome.
      • pbecking1
        [1] I have no Android skills/knowledge/interest