Partial updates use atom with PATCH and seems nice. I just dont personally like too many query parameters and URL patterning them with GET, even less with POST...
15335
Jan Algermissen
algermissen1971
May 10, 2010 9:23 pm
... From a very brief look: Sounds somewhat ok, here is what is missing: - a link relation (or other hypermedia control) that defines the implied URI template...
15334
mike amundsen
mamund
May 10, 2010 9:13 pm
Hi Guilherme! <snip> Can you help me to identify the representation size influence? (I can see we need a handy patch media type to do 2, but if its present, I...
15333
Guilherme Silveira
guilherme_az...
May 10, 2010 9:05 pm
Hello Mike! ... /status-changes/) back to the server using the appropriate media type. Did I miss something or is it the suggestion I made? If so, ++ on this...
15332
piyushpurang
May 10, 2010 8:56 pm
Hi, I wanted to have the groups opinion on the http://googlecode.blogspot.com/2010/03/making-apis-faster-introducing-partial.html Sounds very attractive but...
15331
Guilherme Silveira
guilherme_az...
May 10, 2010 8:56 pm
... GETting the schema URI returns a schema, but in which media type+profile is this schema described? Conneg can be done as usual with this GET, but the same...
15330
Kris Zyp
kriszyp
May 10, 2010 8:48 pm
... Hash: SHA1 I put down some thoughts on the application of REST principles to programming (and how I've applied those in the Persevere framework), if anyone...
15329
Kris Zyp
kriszyp
May 10, 2010 8:41 pm
... Hash: SHA1 ... I believe one should be able to assume that the content type of the representation returned from a server from GET for URI is acceptable in...
15328
Philipp Meier
llucifer_2003
May 10, 2010 8:30 pm
... Which makes me wonder if this is related only to status changes. I think it's a more general pattern: whenever you expect that the processing of a...
15327
Craig McClanahan
craig_mcclan...
May 10, 2010 6:53 pm
... longer than you want the client to have to wait for an initial response. I like to return a 202 Accepted status in this case, complete with a link to a ...
15326
Jørn Wildt
jorn_lind_ni...
May 10, 2010 6:40 pm
(3) is the one I am gravitating towards. It expresses exactly what I want and in addition to this it also serves as a source of the history of the resource -...
15325
Eb
amaeze77
May 10, 2010 6:02 pm
... I concur. This is what I use whenever I start asking myself which way to go. Eb...
15324
mike amundsen
mamund
May 10, 2010 5:53 pm
IMO, this boils down to one of three options: 1) resend the entire representation (via PUT /{existing-item-uri}) back to the server w/ the "status" data...
15323
Tim Williams
quwiltw
May 10, 2010 2:42 pm
... You've selected to build on top of a pretty inefficient style, so it's to be expected I suppose. ... It seems to me that it's just changing the resource...
15322
Avdi Grimm
avdi
May 10, 2010 2:28 pm
I'm sure this has been covered in the past, but I missed it... what are some general guidelines for giving a client hints about what sort of data to supply to...
15321
Jorn Wildt
jorn_lind_ni...
May 10, 2010 2:26 pm
... http://tools.ietf.org/html/rfc5023#section-13.1.1 Oh, cool. Thanks. But ... then we are back to the situation where I must POST my whole resource in order...
15320
Philipp Meier
llucifer_2003
May 10, 2010 2:22 pm
... This would by a nice use case for the PATCH method. This way one could avoid the re-transmission of the whole posting. -billy....
15319
Jan Algermissen
algermissen1971
May 10, 2010 2:14 pm
... Yes. It is an anti-pattern because the message is not self descriptive. See http://www.nordsc.com/blog/?p=414 Jan ... Jan Algermissen, Consultant NORD...
15318
Jan Algermissen
algermissen1971
May 10, 2010 2:13 pm
... Yes. It is an anti-pattern because the message is not self descriptive. See http://www.nordsc.com/blog/?p=414 Jan ... Jan Algermissen, Consultant NORD...
15317
Jan Algermissen
algermissen1971
May 10, 2010 2:12 pm
... AtomPub has native support for that use case: http://tools.ietf.org/html/rfc5023#section-13.1.1 Use <app:draft/> to mark the posting as draft and simply...
15316
Guilherme Silveira
guilherme_az...
May 10, 2010 2:08 pm
... resource a list of "transaction objects" - objects representing operations that have taken place. Actually, in a lot of systems, every resource list is a...
15315
Jorn Wildt
jorn_lind_ni...
May 10, 2010 1:48 pm
... Aha ... this would make the "publication" (or rather "publicationS") resource a list of "transaction objects" - objects representing operations that have...
15314
Guilherme Silveira
guilherme_az...
May 10, 2010 1:13 pm
... Sorry, Im pretty bad with the difference between verbs and nouns in english as its my second language. The correct relation name would be "publication"....
15313
Jorn Wildt
jorn_lind_ni...
May 10, 2010 12:15 pm
... Except that your introduction of a publish URL is one step better than assuming a specific sub-resource. Thanks! So a more general solution is: if you have...
15312
Philipp Meier
llucifer_2003
May 10, 2010 11:38 am
... I consider the 'empty' POST to the publish URL as a little too RPC like. I would rather post a text/uri-list to a resource that is linked with relation...
15311
Guilherme Silveira
guilherme_az...
May 10, 2010 11:33 am
Hello Jorn, What do you think of? POST /drafts (+ post content) ... GET (previous location header) ... POST (following the link 'publish') ... So you do not...
15310
Jorn Wildt
jorn_lind_ni...
May 10, 2010 8:16 am
... Yes. That's like my (5) except using complete URI's (which is better than my idea of using IDs). /Jørn...
15309
Erlend Hamnaberg
ngarthl
May 10, 2010 7:25 am
POST a text/uri-list with the blog posts you want to publish? ... POST a text/uri-list with the blog posts you want to publish? On Mon, May 10, 2010 at 8:30...
15308
Jorn Wildt
jorn_lind_ni...
May 10, 2010 6:31 am
I am working on a Blog project where I use ATOM for representing feeds and blog posts. I have a set of draft posts and published posts. In order to change a...
15307
Etan Wexler
etan_wexler
May 10, 2010 2:12 am
Roy Fielding mistyped (ha!) a word in one of his playful messages<http://tech.groups.yahoo.com/group/rest-discuss/message/15023>; ... Ah, yes, a similar thing...
