IRC log for #rest, 2016-04-27

https://trygvis.io/rest-wiki/

All times shown according to UTC.

Time	Nick	Message
00:36		fuzzyhorns joined #rest
01:00		quimrsto_ joined #rest
01:37		gigo1980________ joined #rest
02:01		quimrstorres joined #rest
03:00		gigo1980________ joined #rest
04:02		quimrstorres joined #rest
05:32		Macaveli joined #rest
06:34		GriffinHeart joined #rest
06:40		gigo1980________ joined #rest
06:45		HermanToothrot joined #rest
07:02		rickharrison joined #rest
07:12		_ollie joined #rest
07:25		interop_madness joined #rest
07:34		gigo1980________ joined #rest
08:03		quimrsto_ joined #rest
08:06		nesselbosch joined #rest
08:38		interop_madness joined #rest
10:05		quimrsto_ joined #rest
10:24		quimrsto_ joined #rest
10:27		quimrsto_ joined #rest
10:37		Macaveli joined #rest
12:19		Macaveli joined #rest
12:33		fuzzyhorns joined #rest
12:37		Macaveli joined #rest
12:52		Devastator_ joined #rest
13:13		tbsf joined #rest
14:12		Macaveli joined #rest
14:29		tbsf joined #rest
15:06		quimrsto_ joined #rest
15:08		tbsf joined #rest
15:11		quimrsto_ joined #rest
15:13		quimrsto_ joined #rest
15:29		GriffinHeart joined #rest
15:29		Coldblackice_ joined #rest
15:31		quimrsto_ joined #rest
15:32		tbsf joined #rest
16:22		Macaveli joined #rest
16:26		tbsf joined #rest
16:36		whartung joined #rest
16:42		quimrst__ joined #rest
18:13		quimrsto_ joined #rest
18:47		quimrstorres joined #rest
20:05		SJr joined #rest
20:09	SJr	I have a system where system A is authoritative for some info, and updates system B with that information. It does so via a REST PUT request currently. We would like to make the system more reliable and have a way of system B pulling the information from system A, by reading a series of events in a stream and updating (like the Atom queues in "REST in Practice" and choreography in "Building Microservices")
20:10	SJr	The problem I have is that I'm not sure how to coordinate the two processes. For every update system A has a version number, that could be used in an ETag and conditional update.
20:11	SJr	The problem is that I want semantics where the ETag (which would be version) is <= not just = and I'm not sure the best way to do that.
20:11	SJr	i.e. system A tells system B, update entity if version <= X
20:25		NordiQ joined #rest
20:28	NordiQ	Hi all. Have a conceptual question. Let's say I have a server that polls external resources and then the client needs to be able to force the server to refresh these resources, how would you do it to stay within the spirit of the restfull api?
20:29	NordiQ	Using headers?
21:02		interop_home joined #rest
21:02		interop_home joined #rest
21:06	whartung	SJr: in what case would you not want to update if the ETag is different?
21:06	whartung	is there a case where B will have a higher version than A?
21:10	SJr	Yeah
21:10	SJr	If the background process of B pulling is ahead of what A is posting.
21:10	SJr	I admit it's a very weird edge case, but I wanted to think about it.
21:12	whartung	well there's nothing that says you can't implement etag however you want, frankly.
21:12	SJr	The HTTP RFC?
21:16	whartung	well, the issue there is basically cache validation with intermittent proxies
21:18	SJr	Yeah and I think that would be fine. Because the only change to the regular behaviour I think is that we won't update in certain cases where we otherwise should.
21:33		Coldblackice joined #rest
22:03		Coldblackice joined #rest
22:10		fuzzyhorns joined #rest
22:26		Coldblackice joined #rest
22:56		rowan7 joined #rest
23:15		fuzzyhorns joined #rest
23:16		quimrstorres joined #rest

https://trygvis.io/rest-wiki/