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/
diehlpk has joined #ste||ar
nikunj has quit [Ping timeout: 248 seconds]
MDez has joined #ste||ar
<MDez>
Hello! I'm a technical writer interested in participate in the Google Season of Docs for 2019! Last week I reached out on email and was suggested to join here and discuss my project ideas with people.
<MDez>
One of the things i'm interested in is re-structuring the API Reference page. It's an enormous page right now with lots of data and I think it could benefit from having a companion navigation / link table or something similar. I imagine it's difficult for users to look up a specific API reference unless they run a find operation on the page (Which of course, means remembering how it's correctly spelled and such).
<diehlpk>
MDez, Hi
<diehlpk>
sounds interesting
<diehlpk>
One thing we need to decide how we group things
<diehlpk>
here heller and hkaiser and simbergm could give some hints
<MDez>
Okay, excellent!
<hkaiser>
MDez: sounds like a good start
<hkaiser>
this page is generated, though, so it would require some tooling to make that happen
<hkaiser>
MDez: I don't know anything about how generation of the docs work, simbergm is the one to poke, he's in Euorpe - so vast asleep at this point
hkaiser has quit [Quit: bye]
<diehlpk>
Extensive cross-references: semantic markup and automatic links for functions, classes, citations, glossary terms and similar pieces of information
<diehlpk>
Hierarchical structure: easy definition of a document tree, with automatic links to siblings, parents and children
<diehlpk>
I think it should be possible to do this, because the mention these features
<diehlpk>
parsa, Do you know how the docs are generated?
<diehlpk>
MDez, parsa could help to since he is the co-mentor of simbergm
<MDez>
Okay, all those features sound like they could be useful. I'll dig some more into sphinx and see if I can flow something out.
<diehlpk>
sounds good
K-ballo has quit [Quit: K-ballo]
diehlpk has quit [Ping timeout: 252 seconds]
MDez has quit [Quit: Going offline, see ya! (www.adiirc.com)]
jbjnr has joined #ste||ar
jaafar has quit [Quit: Konversation terminated!]
mdiers_ has quit [Read error: Connection reset by peer]
mdiers_1 has joined #ste||ar
mdiers_1 has quit [Ping timeout: 258 seconds]
mdiers_ has joined #ste||ar
<simbergm>
MDez: hi! yeah, the API documentation is another big pain point
<simbergm>
that was the one point where we regressed significantly moving from boostbook to sphinx
<simbergm>
we currently use doxygen and the breathe plugin in sphinx to generate the API documentation
<simbergm>
there's also something called exhale which builds on top of breathe and adds a lot of niceties: https://github.com/svenevs/exhale
<simbergm>
we had problems with it before (e.g. it errored out on overloaded functions), but I think those are fixed now
<simbergm>
it could be worth another try
<simbergm>
another thing to keep in mind is that we'll automatically get a bit more structure as we start splitting HPX into modules
<simbergm>
we have the first one in place for our upcoming release and each module gets its own section for (at least) its API docs
<hkaiser>
simbergm: just wanted to apologize for being MIA
<hkaiser>
was travelling and generally too busy :/
<simbergm>
:o
<simbergm>
no need to apologize!
<hkaiser>
I really appreciate you are carrying things forward!
<simbergm>
if you're thinking about the small vector thing, I figured you might have other more important things to think about
<hkaiser>
that as well, sure
<hkaiser>
I meant more generally
<simbergm>
well, if there was anything else, I didn't notice or don't mind :)
<hkaiser>
I feel like I don't invest as much time into HPX as I should
<simbergm>
obviously we appreciate all the help we can get from you, but you shouldn't need to apologize if you don't have the time to spend more time with HPX
<simbergm>
hkaiser: anyway, I had a question for you: are there any requirements on the shape parameter passed to the bulk_*_execute functions?
<simbergm>
I'm trying to get a scheduling hint into the bulk_*_execute functions, and right now I'm thinking making the shape iterator return tuple<hint, tuple_that_we_currently_use>
<simbergm>
am I going to make (future) standards conformance more difficult by doing so
<simbergm>
?
hkaiser has quit [Quit: bye]
<heller>
I am sympathizing with hkaiser :(
<heller>
and probably need to apoligize myself
<heller>
now, let's finally get the packet based buildbot up and running
<zao>
You all are doing acceptably well!
<jbjnr>
I believe that "acceptable well" is a euphemism for "poor"
<jbjnr>
^acceptably
<K-ballo>
I for one won't be apologizing
<jbjnr>
why would you - if you were going to though? have you done something worth confessing about?
<diehlpk_work>
simbergm, One GSoD applicant showed up yesterday evening
<diehlpk_work>
Have you seen the log?
<simbergm>
MDez? yeah, replied, not sure if he/she has seen my reply though
aserio has joined #ste||ar
MDez has joined #ste||ar
<MDez>
simbergm, yes, I just reviewed the log and saw. For the record I am he/him, I believe we exchanged email last week as well. MBenavidez@gmail.com :)
<MDez>
I'll try and dig into those links over the weekend and get familiar with the tool.
<simbergm>
MDez: ah, good to hear that you found your way here! the "d" in your nick made it hard to connect the dots...