Add PATCH-processing functionality to Resource. #91
Conversation
patrickt
force-pushed
the
feature/process-patch
branch
from
e28c343
to
230c502
Compare
| else o17 r | ||
|
|
||
| o17 r@Resource{..} = do | ||
| trace "o17" |
There was a problem hiding this comment.
We need an updated graph now :)
There was a problem hiding this comment.
@charleso Sure thing; anyone have an OmniGraffle doc for it? 😄
There was a problem hiding this comment.
Curious that web machine doesn't appear to support PATCH right now.
There was a problem hiding this comment.
@tmcgilchrist Apparently webmachine is implemented based strictly on RFC 2616, which does not discuss HTTP PATCH
There was a problem hiding this comment.
Might be nice at some point to tweak for-GET and have the (slightly augmented) airship flow graph.
There was a problem hiding this comment.
Hmm… though for-GET is gorgeous, it departs pretty significantly from the Airship graph, AFAICT.
|
Ugh, sorry for all the whitespace changes. If you really object, I'll disable stylish-haskell (though it's my contention that everyone should have stylish-haskell on everywhere, with editor integration) |
|
Fixes #90. |
I'm all for consistent styles, whatever they are. Definitely bikeshedding, but (EDIT: our) style guides are more aligned with something like elm. So we don't, for example, align the import values to avoid the big-reshuffle if you import something slightly longer (I don't know how stylish-haskell deals with this, so that may be a moot point). In any case, I'm happy to stick with style-haskell (I don't use it at the moment). You could even go so far to have a pre-commit hook script with instructions for setting up locally to ensure it doesn't get missed. And/or an extra build that fails if something isn't formatted correctly (which might be a little extreme). |
|
@charleso Cool, I'll leave it in. What I like about stylish-haskell is that it doesn't futz with indentation. It mostly makes small but helpful changes (aligning type declarations, etc.) |
|
@patrickt re whitespace, as long as it's consistent, I care more about the code being correct. |
|
👍 |
| , processPost :: Webmachine m (PostResponse m) | ||
| , processPost :: Webmachine m (PostResponse m) | ||
| -- | As with 'processPost', but called on PATCH requests. | ||
| , processPatch :: Webmachine m Bool |
There was a problem hiding this comment.
Just curious, and this might be part of the RFC, but do you normally care about the content-type for PATCH in the same way you do with PUT?
There was a problem hiding this comment.
@charleso I'd say yes, though I would imagine most people are passing in application/json.
|
I don't think this fully handles |
|
Also it looks like you should be calling EDIT Some interesting issues raised here on the ruby web machine client. I don't think the linked diagram is quite right but it does show the other areas that need to change to support PATCH |
|
Another note: from the RFC,
This is counterintuitive! We presumably need allowMissingPatch :: Webmachine m Boolin the resource type. EDIT: but we will do this later |
patrickt
force-pushed
the
feature/process-patch
branch
4 times, most recently
from
783bfe3
to
6565269
Compare
This extends the decision tree with actions to take in case of a PATCH request. If one is encountered, the processPatch member is invoked: if its operation effected no change, it returns a 304 Not Modified; if it did, return a 202 Accepted. Also updates to 0.4.4.0.
patrickt
force-pushed
the
feature/process-patch
branch
from
2128fc2
to
7157703
Compare
|
That last commit, connecting o17 to o20, looks right to me. I'd think the PATCH is a rather tricky verb to get right, it'd be good to have an example of how to do it somewhere in the project. |
|
@tmcgilchrist Re: I can definitely add a PATCH example. |
|
Cheers, I'm happy with that. 👍 from me |
Add PATCH-processing functionality to Resource.
This extends the decision tree with actions to take in case of a PATCH
request. If one is encountered, the processPatch member is invoked: if
its operation effected no change, it returns a 304 Not Modified; if it
did, return a 204 No Content. This is all invoked directly after testing for
POST.