hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar.cct.lsu.edu | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | Buildbot: http://rostam.cct.lsu.edu/ | Log: http://irclog.cct.lsu.edu/ | GSoD: https://developers.google.com/season-of-docs/
nikunj has joined #ste||ar
nikunj has quit [Remote host closed the connection]
jaafar has joined #ste||ar
rori has joined #ste||ar
heller has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
heller has joined #ste||ar
aserio has joined #ste||ar
K-ballo1 has joined #ste||ar
K-ballo has quit [Ping timeout: 245 seconds]
K-ballo1 is now known as K-ballo
<Yorlik> simbergm: Was it you who set up breathe for the docs?
<Yorlik> NVM - found the answer in the source ..
<simbergm> Yorlik: I'm the guilty one
<Yorlik> OK
<Yorlik> I just was shocked looking at this huge CMake for the docs
<simbergm> if there's something you'd like to see documented outside of the source let me know
<Yorlik> And i wondered if there is a simple answer
<Yorlik> to template documenting
<Yorlik> I tried it with breathe and failed horribly
<Yorlik> I use doxygen in general
<Yorlik> But I have a more high level documentation in sphinx and trid using breathe to pull docs from the doxy xml
<Yorlik> Buit templates explode on me. Simple structs are fine
<Yorlik> Hints 7 Tps / ideas ?
<simbergm> you can use vanilla sphinx+breathe+(auto)doxygen without cmake and I think you'd get decent results
<simbergm> yeah, the doxygen/breathe bridge is a bit flakey
<Yorlik> Its only the templates
<simbergm> we use what we have but there's a ton of warnings
<Yorlik> Did you run into any issues concerning templates?
<simbergm> yep
<simbergm> I don't have a good solution for that
<Yorlik> How did you manage to get them documented?
<Yorlik> A bad working solution would also be nice :)
<simbergm> it hasn't blown up with hard errors though
<simbergm> and shouldn't in most cases
<Yorlik> I'm getting this in the doc output: doxygenstruct: Cannot find class “agns::memory::ContiguousPoolAllocator<typename T, FreelistType FL>” in doxygen xml output f
<Yorlik> I used this code:
<Yorlik> .. doxygenstruct:: agns::memory::ContiguousPoolAllocator<typename T, FreelistType FL>
<simbergm> I'm sure we have similar warnings
<Yorlik> and :members: etc ..
<simbergm> or is that a hard error?
<Yorlik> Its printed on the page. The rest of the page is OK
<Yorlik> But the templkate is undocumented
<simbergm> everything we have is through breathe, so no manual .. doxygenstruct:s
<Yorlik> Its in the Doxygen output though - I double checked
<Yorlik> Uh .. How did you do that? Is it in the plain breathe docs?
<Yorlik> Like just follow the recipe?
<Yorlik> OK .. doxygenindex:: seems to work
<Yorlik> OMG - it's not even respecting my namespaces ....
<Yorlik> That's painful
<simbergm> It is painful :/ I'm not happy with our api docs but sphinx was overall better than boostbook
<Yorlik> I thinka high level documentation with links to doxygen and an occasional manual snippet is good enough
<Yorlik> But then the code really needs to have good comments
<Yorlik> I have big hpes in the GSOD
<Yorlik> hopes
aserio has quit [Ping timeout: 276 seconds]
aserio has joined #ste||ar
diehlpk_mobile has joined #ste||ar
<Yorlik> simbergm: OK - finally got it working by fuzzing it (trying stuff randomly)
<Yorlik> I tried a ton of themes, but there is no decent theme for API documentation.
<Yorlik> We need a C++ developer who uses breathe and has a pro webdev partner doing an awesome theme for C++ documentation :)
diehlpk_mobile has quit [Ping timeout: 276 seconds]
<simbergm> Yorlik yep that'd be great...
<Yorlik> After trying a ton I came back to the rtd theme - I think its the best for APIs, concerning the typography
<Yorlik> For plain text stuff I like agogo and some others
<simbergm> BTW, exhale is something we want to try again, could give slightly more usable api docs
<simbergm> It's another sphinx plugin
<Yorlik> I saw it, but didn't try it yet. I only know it'Äs on top of breathe
jaafar has quit [Ping timeout: 276 seconds]
<Yorlik> Do any of you have experience with Apache Ignite? I am pondering to use it for storage, alternatively Cassandra.
aserio has quit [Ping timeout: 276 seconds]
<zao> As a rule, I don't even bother with anything Apache <X> for personal or work purposes.
<zao> It's where over-engineered enterprise junk goes to die :)
<zao> (I'm sure there may be worthwhile projects there, just that I never ran into any)
<Yorlik> And how does that apply to Ignite or Cassandra?
<Yorlik> Are junk? Overengineered?
<Yorlik> Dying?
<zao> I've never heard of Ignite, but in general if anything is prefixed with "Apache", it's nothing I want.
<Yorlik> So your answer to my question is a general "Apache stuff is mostly shit" ?
<zao> Not necessarily "shit", but often over-engineered and abandoned by the original owners into the ASF's welcoming arms.
<zao> As for these two, again, no idea.
<zao> Just a good guideline, I've found.
<Yorlik> I'm sure you have better alternativs then, especially for distributed in memory databases?
<Yorlik> Redis?
<zao> Not my department, sorry.
<Yorlik> Allright. I'll watch out for overengineered Trolls in their code then ;)
<heller> Yorlik: exasol :P
<Yorlik> Hey Thomas!
<heller> hey chris!
<Yorlik> So - you want to sponsor us a permanent ExaSol Cluster License? ;)
<heller> Ha
nikunj97 has joined #ste||ar
rori has quit [Quit: WeeChat 1.9.1]
jaafar has joined #ste||ar
jaafar_ has joined #ste||ar
jaafar has quit [Ping timeout: 248 seconds]
aserio has joined #ste||ar
hkaiser has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
<parsa> what's the difference between `hpx::shared_future(std::move(f))` and `hpx::make_shared_future(std::move(f))`?
<parsa> nvm
<parsa> probably pre-c++17 the template arg requirement is the answer
<hkaiser> parsa: I think constructing a shared future gives you a shared_future<T>, your make_shared_future gives you a shared_future<shared_future<T>>
<parsa> hkaiser: both give me hpx::lcos::shared_future<int>. my code wouldn't work otherwise
<hkaiser> parsa: ok
<hkaiser> however note that a shared_future<T> is constructable from a shared_future<shared_future<T>>
<parsa> not sure what that means. are you saying both shared_futures will be unwrapped and the result is put in a new shared_future?
<parsa> :s/both/the nested/
<hkaiser> yes
<hkaiser> you can assign a sf<sf<T>> to a sf<T>
<hkaiser> (I think so at least, let me have a look)
<hkaiser> parsa: no, I was wrong
<hkaiser> also, make_shared_future<move(sf)) is a no-op: https://github.com/STEllAR-GROUP/hpx/blob/master/hpx/lcos/future.hpp#L1471-L1476
<hkaiser> in my initial comment I mixed it up with make_ready_future, which is totally different
hkaiser has quit [Quit: bye]
jaafar_ is now known as jaafar
aserio has quit [Quit: aserio]
bita_ has joined #ste||ar
bita has quit [Ping timeout: 276 seconds]