<dwrensha>
... and after disabling etags in sharelatex's socket.io server, the problem goes away
<kentonv>
dwrensha: sounds like sharelatex fails at etags
<kentonv>
the standard very specifically states that they must be quoted
<dwrensha>
it might have to do with my hack for putting stuff in the same process. not sure
<kentonv>
that seems unlikely... what would that have to do with ETag headers?
<dwrensha>
shrug
mnutt_ has joined #sandstorm
<kentonv>
lolol bootstrap's grid system lets you have a series of boxes that flow like text but if they aren't all exactly the same height then you have to manually indicate where the line breaks are for every screen size which kind of defeats the purpose. Arghhh.
<kentonv>
(because they are floats, not inline-blocks)
<dwrensha>
kentonv: new sharelatex spk should be in your queue
myguidingstar has left #sandstorm ["Anything in parenthesis can (not) be ignored."]
roflbox_ has joined #sandstorm
mrdomino_ has joined #sandstorm
jadewang has quit [Ping timeout: 240 seconds]
decipherstatic has joined #sandstorm
kentonv has quit [Ping timeout: 260 seconds]
prosodyCagain has quit [Ping timeout: 246 seconds]
derf- has quit [Ping timeout: 246 seconds]
mrdomino has quit [Ping timeout: 246 seconds]
dvn- has quit [Ping timeout: 246 seconds]
roflbox has quit [Ping timeout: 246 seconds]
coyotebush has quit [Ping timeout: 246 seconds]
jus has quit [Ping timeout: 246 seconds]
decipherstatic_ has quit [Ping timeout: 246 seconds]
coyotebush has joined #sandstorm
derf- has joined #sandstorm
roflbox_ is now known as roflbox
jus has joined #sandstorm
prosodyCagain has joined #sandstorm
dvn has joined #sandstorm
augustl_ has joined #sandstorm
bpierre_ has joined #sandstorm
saneki_ has joined #sandstorm
dvn has quit [*.net *.split]
simonv3 has quit [*.net *.split]
saneki has quit [*.net *.split]
augustl has quit [*.net *.split]
bpierre has quit [*.net *.split]
oao has quit [*.net *.split]
jjb has joined #sandstorm
jadewang has joined #sandstorm
augustl_ is now known as augustl
kentonv has joined #sandstorm
bpierre_ is now known as bpierre
dvn has joined #sandstorm
simonv3 has joined #sandstorm
rhapsodhy has quit [Remote host closed the connection]
rhapsodhy has joined #sandstorm
simonv3 has quit [Quit: Connection closed for inactivity]
ArcTanSusan has joined #sandstorm
h2o3 has joined #sandstorm
aaronr has joined #sandstorm
ArcTanSusan has quit [Quit: ArcTanSusan]
jadewang has quit []
jadewang has joined #sandstorm
jadewang has quit [Remote host closed the connection]
amyers has joined #sandstorm
asmyers has joined #sandstorm
amyers has quit [Ping timeout: 255 seconds]
tobald has joined #sandstorm
mnutt_ has quit [Quit: mnutt_]
asmyers has quit [Ping timeout: 250 seconds]
asmyers has joined #sandstorm
asmyers has quit [Remote host closed the connection]
asmyers has joined #sandstorm
XgF has quit [Ping timeout: 250 seconds]
XgF has joined #sandstorm
mnutt_ has joined #sandstorm
h2o3 has quit [Ping timeout: 250 seconds]
tobald has quit [Quit: Ex-Chat]
simonv3 has joined #sandstorm
kentonv has quit [Ping timeout: 250 seconds]
kentonv has joined #sandstorm
jadewang has joined #sandstorm
rolig has quit [Quit: Quit]
rolig has joined #sandstorm
NwS has joined #sandstorm
ArcTanSusan has joined #sandstorm
ArcTanSusan has quit [Client Quit]
ArcTanSusan has joined #sandstorm
ArcTanSusan has quit [Client Quit]
ArcTanSusan has joined #sandstorm
erikmaarten has joined #sandstorm
NOTevil has joined #sandstorm
* asheesh
stretches.
jus has quit [Ping timeout: 240 seconds]
ArcTanSusan has quit [Quit: ArcTanSusan]
jus_ has joined #sandstorm
<asheesh>
Hey erikmaarten, sorry I missed you yesterday. See email etc.!
M-eternaleye is now known as eternaleye
eternaleye has quit [Changing host]
eternaleye has joined #sandstorm
rancar2 has joined #sandstorm
funwhilelost has joined #sandstorm
erikmaarten has quit [Ping timeout: 244 seconds]
kentonv has quit [Quit: Leaving]
<zarvox>
asheesh: kentonv wants me to tell you that Debian broke X on his machine and he blames you
<zarvox>
:)
jadewang has quit [Remote host closed the connection]
jadewang has joined #sandstorm
bb010g has joined #sandstorm
erikmaarten has joined #sandstorm
<asheesh>
Howdy erikmaarten
<asheesh>
Did the vagrant-spk stuff I emailed you help?
<erikmaarten>
hi Asheesh! Been out for a while, will have a look!
<asheesh>
Great!
<asheesh>
Thanks for emailing me back, btw!
mnutt_ has quit [Quit: mnutt_]
<gwillen>
Hello #sandstorm! I have another question about the Wekan port.
<gwillen>
Right now, when you open the 'more' dialog for a card, it displays the card's URL as a chunk of text relative to the current URL (rather than an actual link that works). Is this because the grain doesn't know it's own URL, and/or because the grain's URL is not durable so you can't make a permalink like that?
<gwillen>
I guess the latter, which I just realized... is there a good way to handle this? Would it pollute the ACL table to insane levels if it could automatically be a share link that comes straight to that card, or some other sort of durable link?
kecolus has joined #sandstorm
mnutt_ has joined #sandstorm
asmyers has quit [Remote host closed the connection]
asmyers has joined #sandstorm
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
asmyers has quit [Remote host closed the connection]
asmyers has joined #sandstorm
<erikmaarten>
asheesh: took a while to get through the update process, but it works now! Splendid, thanks so much!
<erikmaarten>
So easy when you know what to do :)
<zarvox>
The grain doesn't know it's own URL, and the grain's URL is not durable. Theoretically, you could link to a specific thing by appending the grain-relative path to the grain URL, but the app doesn't know the grain URL (and I *think* it would be a confinement leak if it did)
funwhilelost has joined #sandstorm
asmyers has quit [Ping timeout: 244 seconds]
<asheesh>
erikmaarten: Great! (-:
simonv3 has quit [Quit: Connection closed for inactivity]
kentonv has joined #sandstorm
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
funwhilelost has joined #sandstorm
<kentonv>
man, this debian update broke _everything_
<kentonv>
ok, just Xorg and somehow 50% of ruby things, but still
DanC has joined #sandstorm
<DanC>
I'm installing sandstorm on an Ubuntu box... very slick; it's come a long way since I last did this... but I can't dereference the admin link.
<DanC>
I'm behind a firewall. Should I expect this to work?
<kentonv>
DanC: if you're using sandcats, you'll need to open ports on the firewall
<kentonv>
80 and 443
<DanC>
I can't do that.
<DanC>
I don't need a sandcats address... I just want sandstorm on this machine
<DanC>
suggestions on how to proceed?
<DanC>
the installation is sufficiently slick that it makes me expect it to just work, even from behind a firewall. Some sort of
<DanC>
Some sort of "We need to reach you on ports 80 and 443, so if you're behind a firewall, blah blah..." is perhaps in order
jadewang has quit [Remote host closed the connection]
<kentonv>
are you installing on your local desktop or on a machine on the network?
<DanC>
local desktop
<DanC>
I'd like it to offer services to the LAN
<kentonv>
well, if you want it LAN-accessible then you will need to set up DNS somehow. Sandcats can only automate it if the machine is visible on the public internet.
<kentonv>
if you only care about accessing it from the local machine then you can use "localhost"
<kentonv>
(with some config changes)
NOTevil has quit [Quit: ...]
<kentonv>
do you have access to your network's DNS server?
erikmaarten has quit [Ping timeout: 240 seconds]
asmyers has joined #sandstorm
* DanC
returns...
<DanC>
this machine has a DNS name in the internal network
<DanC>
xyz.kumc.edu
jadewang has joined #sandstorm
<DanC>
maybe I'll just start my installation over.
<DanC>
hmm... it doesn't seem to have told me how to uninstall
<geofft>
if the server says "Suborigin: foo" in the header, then it gets treated as a separate origin (scheme, host, port, suborigin)
<asheesh>
Will leave open & read in a bit; I am mostly AFK
<geofft>
the use case is I think very different sandstorm (or example.com/~username web hosts) but might work great, nonetheless
<geofft>
*very different from sandstorm
<DanC>
suborigin... interesting idea... thinking out loud: how does deployment work? which end relies on what security properties, and does it work out if the other end is ignorant?
<DanC>
or malevolent
<geofft>
if the browser is malevolent, it can already screw over the user
<geofft>
I think the error case is only that you lose the same-origin protection between grains in old browsers, but I haven't read closely
simonv3 has joined #sandstorm
<DanC>
when there was just alice and bob, I sould sometimes get my head around security protocol properties. But now there's Alice, Bob, Alice's browser, Bob's server, 123 ad networks, the russian mafia, google maps, and 12 jquery plug-ins
<geofft>
hahahaha
asmyers has quit [Ping timeout: 246 seconds]
<DanC>
if only it were a joke. :-{
<geofft>
yeah it's funny 'cause it's true
bb010g has quit [Quit: Connection closed for inactivity]
<DanC>
btw... I sorta want the sandcats address that was just generated for me; will it revert back to being available in 15 minutes so I can grab it later?
<DanC>
grumble... "Firefox can't establish a connection to the server at local.sandstorm.io:6080"
<DanC>
I can ping local.sandstorm.io
<DanC>
and telnet to 6080
funwhilelost has joined #sandstorm
<DanC>
ah. all better now
<DanC>
On the left sidebar, go to "APIs & auth" <- google dev console seems to have been redesigned or something. I see no such thing
<gwillen>
zarvox: is there some way the app could ask sandstorm for a sharing link (or some other form of durable link to its grain) to append stuff to?
<gwillen>
zarvox: really it seems like we want a more general way for an app to create durable deep links into itself
<DanC>
" IpNetwork represents the capability to make outbound requests to the network. " ooh! when did that get added? I thought I was tuned into all the "sandstorm has a new thingy" channels
kecolus has quit [Quit: Leaving]
<gwillen>
DanC: they redesign that damn thing every 6 months, I always have to write down where I find things because the docs are out of date
<gwillen>
(the dev console)
<DanC>
on the one hand, why expect a web site design to stay still? on the other hand, since they're a platform, all sorts of other projects are playing the same catch-up game; that's why.
<gwillen>
well, if they would keep the docs up to date it wouldn't be a problem
<DanC>
sandstorm is unbelievably cool, btw
<DanC>
I just went from zero-to-rocketchat in what...
<DanC>
15min. including chatting and whatnot
<DanC>
now to the task that started me down this road today: does ethercalc grok dates from our log files as well as gnumeric (which I can't share with dev team) or google sheets (which I can't keep in the enterprise)?
<DanC>
I don't suppose ethercalc is smart enough to do attenuation... i.e. read-only sharing, or sharing of just one sheet?
<dwrensha>
DanC: read-only sharing is indeed implemented in EtherCalc
<gwillen>
dwrensha: huuuh, I'm not sure
<DanC>
dwrensha, see? unbelievable
<DanC>
sandstorm has gone from "when will they have XYZ?" to "wow... ABC too?!" at some point. it's over some sort of hump, for me.
<gwillen>
asheesh: so when I have a url of the form https://sandstorm.foobar.com/grain/xxx, is that a durable reference to that grain, even if it reboots?
<gwillen>
I think maybe I knew this but forgot
<DanC>
best kickstarter investment I ever made :)
<gwillen>
(so the id that changes is the hidden one I'd see if I looked at the iframe?)
<dwrensha>
gwillen: yes, grain IDs stay the same
<gwillen>
ok, cool! Then my question is, can apps learn their grain IDs, or in some other way get sandstorm to make an deep share link for them?
<DanC>
alas, ethercalc does not grok 2015-10-19 00:23:46 as a datetime; thinks it's a string
<dwrensha>
gwillen: I don't think grains currently have a good way to get their own grain ID, though
<gwillen>
for the sake of simplicity let me assume the link does NOT need to convey privileges
<gwillen>
and that it only needs to be viewable by people who already have them
<gwillen>
dwrensha: is there a reason for them not to be able? Could lib-sandstorm or whatever the JS shim is, have a "build me a durable sublink" function easily?
<dwrensha>
gwillen: right, a /grain/ link does not share access
* gwillen
nods
<gwillen>
for our purposes I think that'll do fine, rather than having to bikeshed something figuring out how the ACLs should work
xet7 has quit [Quit: Leaving]
<gwillen>
wekan already has a 'give me a deep permalink' feature, and I'm guessing whoever did the port just replaced it with a textual URL path because the functionality was not available
<dwrensha>
gwillen: you could append wekan's relative URL to a /shared/ link generated by Sandstorm "Share Access" button
<gwillen>
dwrensha: that's true -- for our purposes either would be fine (although I think the sharing thing could be more surprising, and surprising ACL additions are not great)
<gwillen>
dwrensha: all I want is a button in wekan that emits the link without the end-user having to copy-paste things together
<gwillen>
(that is to say, I want the existing such button to work :-)
<dwrensha>
gwillen: do you want to send these URLs to user who already have access, or to users who don't have access yet, or both?
<gwillen>
the former is entirely sufficient for our purposes
<gwillen>
(it may not be sufficient for everyone's purposes forever but it's all we need)
<zarvox>
gwillen: sorry, wasn't paying attention to IRC, reading backlog now
<gwillen>
zarvox: np!
home_ has joined #sandstorm
<zarvox>
gwillen: If you want to write up your deeplinking requirements in an issue against sandstorm-io/sandstorm, that would be great. If not, say so, and I'll do my best to paraphrase the above discussion!
<gwillen>
zarvox: I'll do it! :-)
<dwrensha>
I suppose that Wekan could use our existing setPath API, and then the user could just copy the address from their brower's URL bar
<gwillen>
dwrensha: oh, that's clever
<gwillen>
probably requires more extensive fiddling in wekan though