<mnutt_>
I'm also currently battling new WebDAV woes. WebDAV uses OPTIONS to determine the server's capabilities, but sandstorm returns a standard response for unauthenticated OPTIONS requests.
<asheesh>
Can't you "just" do the OPTIONS requests authenticated?
<asheesh>
Also I must AFK to hurriedly give a cousin resume feedback.
neynah has joined #sandstorm
<asheesh>
I will be back later/tomorrow. Feel free to ramble more etc but thought I should at least indicate availability level (-:
<mnutt_>
no problem :)
<mnutt_>
often, clients try without Authorization header, wait for a 401 Unauthenticated and then retry with Authorization header
<mnutt_>
I emailed the Transit folks to ask them why it doesn't later try the Authorization header, but it would be great to figure out a way for Sandstorm to pretend to be a regular WebDAV server
<mnutt_>
unfortunately, I can't think of many great options: ideally, the OPTIONS call would return a 401, the client would send the Authorization, and then the second OPTIONS would go to the grain. but I think that would break XHR, which depends on OPTIONS being unauthenticated.
<asheesh>
simonv3: I am using Annotate (or is it Annotator?) to give feedback to my cousin (who is in college) on his resume.
<asheesh>
That is the thing I am busy doing instead of talking with you, mnutt_ !
<mnutt_>
pretty awesome. I wonder if imagemagick could be added to do that for you
<mnutt_>
or some general document-to-image library
|jemc| has quit [Quit: WeeChat 1.4]
|jemc| has joined #sandstorm
<asheesh>
...I just realized that when I'm done reviewing my cousin's resume, I might send them to you, mnutt_ ...
<mnutt_>
asheesh: sounds good! nyc-based?
<asheesh>
Currently Philadelphia; I'll send you more info if/when we finish resume fixing, etc. (-:
<mnutt_>
Ohh man. I did everything I could to try to get Transmit to send Authorization headers, and it wasn't until I changed to WWW-Authenticate: basic realm='ownCloud' that it started working :-/
<mnutt_>
I'm going to ask them to also whitelist Sandstorm, it's way less hacky than trying to get sandstorm to pretend to be ownCloud.
<simonv3>
asheesh: hah, awesome. Unfortunately I won't be able to push any changes over the next few weeks, so publishing will have to wait
xet7 has quit [Ping timeout: 240 seconds]
<asheesh>
Fair enough!
xet7 has joined #sandstorm
kentonv has joined #sandstorm
mcpherrin has quit [Ping timeout: 244 seconds]
<mnutt_>
can someone remind me if there is an http bridge header to inform the grain that a request came from the api and not the browser?
jacksingleton has quit [Ping timeout: 252 seconds]
mcpherrin has joined #sandstorm
|jemc| has quit [Ping timeout: 248 seconds]
mcpherrin has quit [Ping timeout: 255 seconds]
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mcpherrin has joined #sandstorm
jacksingleton has joined #sandstorm
jacksingleton has quit [Ping timeout: 252 seconds]
frigginglorious has quit [Ping timeout: 248 seconds]
jacksingleton has quit [Ping timeout: 248 seconds]
jacksingleton has joined #sandstorm
rustyrazorblade has quit [Quit: rustyrazorblade]
tobald has joined #sandstorm
raoulzecat has joined #sandstorm
jacksingleton has quit [Ping timeout: 240 seconds]
aaronr has joined #sandstorm
synchrone has joined #sandstorm
Jan\ has joined #sandstorm
NhanH has quit [Ping timeout: 240 seconds]
bpierre has quit [Ping timeout: 240 seconds]
nolski has quit [Ping timeout: 240 seconds]
Zertrin has quit [Ping timeout: 240 seconds]
Zertrin has joined #sandstorm
NhanH has joined #sandstorm
nolski has joined #sandstorm
bpierre has joined #sandstorm
xet7 has quit [Ping timeout: 240 seconds]
xet7 has joined #sandstorm
amyers has joined #sandstorm
amyers has quit [Remote host closed the connection]
amyers has joined #sandstorm
xet7 has quit [Ping timeout: 240 seconds]
synchrone has quit [Ping timeout: 240 seconds]
amyers has quit [Read error: No route to host]
amyers has joined #sandstorm
amyers has quit [Remote host closed the connection]
amyers has joined #sandstorm
xet7 has joined #sandstorm
hannes3 has joined #sandstorm
<hannes3>
if i want to use my account on another machine, what is the primary way to do so? when i signed up all i got was a mail
Jan\ has quit [Ping timeout: 250 seconds]
Jan\ has joined #sandstorm
<dwrensha>
hannes3: click "Sign in" in the upper right and enter your email address in the "with email" part.
<hannes3>
the same as before?
<dwrensha>
yeah, "signing in" works the same as "creating an account"
<hannes3>
do i need access to my mail account for that (ie do i get another mail or what)?
<hannes3>
do i have to repeat this every time the cookies vanish?
<dwrensha>
you'll get another mail
<dwrensha>
the login token is stored in localStorage
<hannes3>
could i just take that token with me?
<dwrensha>
interesting idea. possibly!
<dwrensha>
Accounts._storedLoginToken() will give you the token
<dwrensha>
and `Meteor.loginWithToken(token)` will log you in with it
<dwrensha>
Looks like the tokens expires after four months.
aaronr has quit [Read error: Connection reset by peer]
aaronr has joined #sandstorm
tobald has quit [Ping timeout: 240 seconds]
tobald has joined #sandstorm
aaronr_ has joined #sandstorm
synchrone has joined #sandstorm
|jemc| has joined #sandstorm
wolcen has joined #sandstorm
bodisiw has joined #sandstorm
aaronr_ has quit [Remote host closed the connection]
raoulzecat has quit [Ping timeout: 250 seconds]
notevil has joined #sandstorm
mnutt_ has joined #sandstorm
<phildini>
It's mildly annoying that I as an app developer don't have a dashboard or something I can log into to perform operations for my app.
<phildini>
perform operations and see metrics.
<dwrensha>
phildini: what kind of operations?
<phildini>
well, as an example, I tested my spk before kenton approved it, but there are bugs that appear in the app store version that don't appear in any of my testing.
<phildini>
But I can't take it down on my own, or halt people from installing it, so right now people are potentially installing a buggy version, and I have no control over that.
<phildini>
makes me feel pretty not-great as an app developer.
<phildini>
additionally, there's no way for them to report bugs in a meaningful way in the interface, I have no idea how many downloads there have been, and it would be awesome if sandstorm could capture errors and report them to me.
<dwrensha>
There is a `spk publish --remove` command.
<phildini>
does that need someone's confirmation on the backend?
<phildini>
also, is that exposed in vagrant-spk?
<dwrensha>
I would guess that it's automatic. I've never tried it.
<dwrensha>
Don't know about vagrant-spk.
<dwrensha>
but presumably you can `vagrant-spk ssh` and run `spk publish --remove` there
<dwrensha>
kentonv: they push a release every month
<synchrone>
ah, forgot dev instances also support appstore :D
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<synchrone>
why generating the username by the way?
<synchrone>
i mean it's ignored, right?
<dwrensha>
yes, it is ignored on the server
<synchrone>
... i probably should not be using "owner" as a username then
<dwrensha>
the git "store" credential helper keys repos by (username+url)
<dwrensha>
that is, I generate a random username so that clients can disambiguate
<synchrone>
so if the username is the same everywhere, you still get different URLs... since you have different Tokens and hence different api subdomain
<dwrensha>
right
<synchrone>
or you're afraid there might be some client which would not differentiate based on url?
xet7 has joined #sandstorm
<synchrone>
i kind of want my API urls to be shorter... way shorter than 138 chars now
funwhilelost has joined #sandstorm
TC01 has quit [Ping timeout: 240 seconds]
erikmaarten has joined #sandstorm
TC01 has joined #sandstorm
TC01 has quit [Ping timeout: 244 seconds]
TC01 has joined #sandstorm
<neynah>
Yay, Contact Otter update. :D
<synchrone>
Contact Otter?
<synchrone>
man, why is that name totally not googleable
<phildini>
it doesn't have CardDAV support currently because CardDAV is awful.
<phildini>
Its implemented inconsistently across clients _and_ servers, and the existing implementations for python are either wrong license or py2 only.
<phildini>
Radicale, for example, is GPL, meaning that I can't ever really look at it again.
<phildini>
Google has multiple APIs for trying to sync contacts across things.
<phildini>
They are also awful, in new and different ways.
<phildini>
It's highly likely that ContactOtter will have contact syncing at some point in the future, but it would actually be easier to write mobile apps for iphone and android and use _those_ address book apis to handle the sync, than it would be to try and navigate the absolute mess that is CardDAV on the modern internet.
<hannes3>
i only know baikal
[d__d] has joined #sandstorm
sugarpuff has joined #sandstorm
jadewang has quit [Remote host closed the connection]
HalfEatenPie has joined #sandstorm
cstrahan has joined #sandstorm
jadewang has joined #sandstorm
neynah has joined #sandstorm
jadewang has quit [Ping timeout: 252 seconds]
synchrone has quit [Quit: Leaving.]
synchrone has joined #sandstorm
<kentonv>
synchrone: Yeah the long URLs are unfortunate but I'm not sure what else we can do.
<synchrone>
for instance, I still don't see the need of having both Token and a hash of it, in the form of a subdomain
<synchrone>
(maybe i just need to re-iterate over irc logs)
<synchrone>
ah, right, DNS
<synchrone>
phildini: caldav might be a mess... but so would be more or less any protocol with such features, and it's a standardized one. The only sensible way of pushing non-vendor-locked-in software, which are the meat-and-bone of open ecosystems
<synchrone>
not to mention in impressive amount of resources you'd need to spend to document and standardize a protocol
<phildini>
true, but there's no indication any of that would be useful with sandstorm anyway? you can expose an api over the http-bridge, CardDAV needs "more things" than what gets exposed.
<phildini>
So I would be writing my own CardDAV server implementation, and my own sandstorm adapter for a protocol that I'm not wild about to begin with.
<phildini>
(and which most of the common clients don't fully implement, as well as not publishing which parts they do-and-do-not implement)
<kentonv>
phildini: FWIW, spk publish --remove takes effect immediately. The logic on the server is that publishing requires intent from both the publisher and us; if either intent goes away then it is un-published.
<phildini>
if my goal is "make Contact Otter useful to the greatest number of people", then providing mobile apps which "know how to speak sandstorm" seems more reasonable than "try to figure out CardDAV client compliance"
<synchrone>
phildini: sandstorm adapter isn't that much of an effort, really, I did it with Radicale
<synchrone>
and it works with clients right now
<phildini>
I don't see radicale on the app store?
<synchrone>
not yet... it's getting approved
<synchrone>
but you can download my SPK
<phildini>
also, I have no desire to change my license, so looking at radicale is not something I'm terribly interested in.
<synchrone>
the beauty of an open proto is that you can plug anything on the backend
<synchrone>
something sabredav-based
<synchrone>
anyway, you are entitled to do as you see fit, being the app author. I can just express my point of view about the way collaboration can work
<phildini>
absolutely! :D I look forward to playing with radicale on sandstorm when it launches. :)
<synchrone>
also, point out that it was what I did: took a backend-only Radicale, combined with a pure-frontend UI implementation and that was it: https://github.com/synchrone/sandstorm-radicale
<synchrone>
FE was messy, like a jquery project from 2006, but it works fine, I did not need to hack it too much
<phildini>
neat! It's awesome to see the ecosystem growing.
<synchrone>
there is another Calendar UI implementation (AgenDAV), but since there was no Contacts UI ... I figured I'd stick with infCloud
synchrone has quit [Quit: Leaving.]
<kentonv>
synchrone: You said Radicale is "getting approved". What are you waiting on exactly? I don't see it in the app market queue.
synchrone has joined #sandstorm
<kentonv>
synchrone: You said Radicale is "getting approved". What are you waiting on exactly? I don't see it in the app market queue.
<synchrone>
and then it got superceded by another one
<kentonv>
ah, yeah, so that's fixed now. :)
<synchrone>
so I just need to tell @kentonv: hey, my app passed the local test on my machine (I guess I should do it on the mailing list) and we're all set
tdfischer has quit [Ping timeout: 264 seconds]
<kentonv>
and do (vagrant-)spk publish
<synchrone>
so should I do it, like, now?
<kentonv>
if you think it's ready
<synchrone>
ah, okay
<synchrone>
well, ther's a FE resource-speedup-via-static-publishing feature missing, but it was not in my MVP
<kentonv>
we plan to add features to Sandstorm proper to make resource serving faster at some point. For now I wouldn't worry about it. Using static publishing for this is kinda hacky.
<synchrone>
okay then
tdfischer has joined #sandstorm
<synchrone>
uglify.js gets the job somewhat done
<synchrone>
erm... vagrant-spk publish doesn't work for me