|
| 1 | +--- |
| 2 | +title: JSON Hyper-Schema Draft 6 migration FAQ |
| 3 | +layout: page |
| 4 | +--- |
| 5 | + |
| 6 | +FAQ for migrating from draft-luff-json-hyper-schema-00 (draft-04) to draft-wright-json-schema-hyperschema-01 (draft-06). |
| 7 | + |
| 8 | +* TOC |
| 9 | +{:toc} |
| 10 | + |
| 11 | +### Q: What are the incompatible changes between draft-04 and draft-06? |
| 12 | + |
| 13 | +Between drafts 04 and 06 we undertook a major re-examining of Hyper-Schema, which has never been as widely adopted as JSON Schema Validation. |
| 14 | + |
| 15 | +You will notice that some things are still in flight and under discussion for draft-07. We feel that draft-06 is a good set of changes for collecting feedback, with the most notable compatibility gaps easily addressed as extension keywords in the meantime. |
| 16 | + |
| 17 | +#### Changes from draft-04 to draft-05 |
| 18 | + |
| 19 | +keyword | change | consequence |
| 20 | +---- | ---- | ---- |
| 21 | +`"base"` | replaces looking up the nearest "self" link to determine the base URI for `"href"` | if you were relying on "self" links to change the base, set `"base"` explicitly |
| 22 | +`"rel"` | "full" relation removed | use ["item"](https://github.com/json-schema-org/json-schema-spec/issues/295) |
| 23 | +`"rel"` | "instances" and "create" relations removed | use ["collection"](https://github.com/json-schema-org/json-schema-spec/issues/295) |
| 24 | +`"rel"` | "root" relation removed | use a fragment in your `"href"` URI Template |
| 25 | +`"fragmentResolution"` | *removed* | media type determines how fragments are interpreted |
| 26 | +`"pathStart"` | *removed* | *[no replacement]* |
| 27 | +`"method"` | [changed back to HTML form semantics](https://tools.ietf.org/html/draft-zyp-json-schema-03#section-6.1.1.4.1) of "get" and "post" rather than all HTTP methods | *[changed again in draft-06 due to feedback that this was confusing]* |
| 28 | + |
| 29 | +#### Changes from draft-05 to draft-06 |
| 30 | + |
| 31 | +keyword | change | consequence |
| 32 | +---- | ---- | ---- |
| 33 | +`"method"` | *removed* | for HTTP method proposals, see issues [#73](https://github.com/json-schema-org/json-schema-spec/issues/73) and [#296](https://github.com/json-schema-org/json-schema-spec/issues/296) (use either `"method"` or `"allow"` as an extension keyword if needed); indication of how to use `"schema"` and `"encType"` no longer necessary |
| 34 | +`"schema"` | *removed* | use `"hrefSchema"`, `"submissionSchema"`, or `"targetSchema"` | |
| 35 | +`"encType"` | *removed* | use `"submissionEncType"` for request bodies; no longer needed for URI query strings |
| 36 | +`"hrefSchema"` | *added* | replaces `"method": "get", "schema": {...}`, with additional functionality | |
| 37 | +`"submissionSchema"` | *added* | replaces `"method": "post", "schema": {...}` |
| 38 | +`"submissionEncType"` | *added* | replaces `"method": "post", "encType": "..."` |
| 39 | +`"href"` | preprocessing removed | *to be replaced and expanded in future drafts* |
| 40 | + |
| 41 | +#### Proper use of `"targetSchema"` |
| 42 | + |
| 43 | +While `"targetSchema"` did not change its meaning in either recent draft, it has been widely misinterpreted. So it may feel like a change to use it as specified. |
| 44 | + |
| 45 | +Due to draft-04 emphasizing individual HTTP methods as `"method"` values, many users interpreted `"targetSchema"` as a hint of the response to the method in `"method"`. This was never correct; all drafts define this keyword as describing the representation of the target resource, which appears as a response to HTTP GET, but may or may not appear in other responses. |
| 46 | + |
| 47 | +Draft-06 clarifies this usage and provides guidance on its use with different HTTP methods. This includes using `"targetSchema"` as a request description for PUT and PATCH. With draft-04, many users used `"schema"` to describe PUT and PATCH requests which is not needed. |
| 48 | + |
| 49 | +However, see also [#296](https://github.com/json-schema-org/json-schema-spec/issues/296) for a proposal for hinting at "Accept-Patch", which is needed to properly use `"targetSchema"` with HTTP PATCH. |
| 50 | + |
| 51 | +### Q: What are key issues under consideration for draft-07? |
| 52 | + |
| 53 | +There are a number of relatively concrete proposals, although it is unlikely that all will be resolved in draft-07 |
| 54 | + |
| 55 | +* [#73](https://github.com/json-schema-org/json-schema-spec/issues/73) `"allow"` for HTTP method hints, proposed as its own keyword |
| 56 | +* [#296](https://github.com/json-schema-org/json-schema-spec/issues/296) `"usageHints"` for general protocol usage hints, including HTTP "Allow" and "Accept-Patch" |
| 57 | +* [#295](https://github.com/json-schema-org/json-schema-spec/issues/295) clarifying usage of "collection" and "item" in `application/schema+json` |
| 58 | +* [#140](https://github.com/json-schema-org/json-schema-spec/issues/140) `"anchor"` and `"anchorPointer"` for adjusting the link's context URI |
| 59 | +* [#74](https://github.com/json-schema-org/json-schema-spec/issues/74) `"deprecation"` property |
| 60 | +* [#51](https://github.com/json-schema-org/json-schema-spec/issues/51) `"$data"`, particularly for use with `"const"` and/or `"default"` in `"hrefSchema"` and `"submissionSchema"` |
| 61 | +* [#204](https://github.com/json-schema-org/json-schema-spec/issues/204) includes specific possible uses of `"default"` related to Hyper-Schema |
| 62 | + |
| 63 | +There are some important philosophical discussions about the scope and goals of Hyper-Schema, which hopefully will be resolved to help us make the right decisions for draft-07 and beyond: |
| 64 | + |
| 65 | +* [#294](https://github.com/json-schema-org/json-schema-spec/issues/294) how analogous Hyper-Schema should or shouldn't be to HTML, particularly in regard to forms vs anchor semantics |
| 66 | +* [#288](https://github.com/json-schema-org/json-schema-spec/issues/288) whether link URI Templates must be resolvable without knowing whether input data will be used |
| 67 | +* [#226](https://github.com/json-schema-org/json-schema-spec/issues/226) whether and how to handle APIs that do not strictly conform to HTTP semantics |
| 68 | + |
| 69 | + |
| 70 | +Additionally, there are two further proposals for JSON Schema vocabularies which could impact or complement Hyper-Schema: |
| 71 | + |
| 72 | +* [Documentation](https://github.com/json-schema-org/json-schema-spec/issues/136), which could take over some static API description work |
| 73 | +* [UI](https://github.com/json-schema-org/json-schema-spec/issues/67), which would deal with presentation issues for forms |
| 74 | + |
| 75 | +### Q: Why were several major changes made to Hyper-Schema just before publication? |
| 76 | + |
| 77 | +A: During final review, it became apparent that there was no consensus on how to use the spec as written. The late changes were necessary to publish a spec with unambiguous meaning, so that we could get feedback on its contents rather than differing interpretations. Originally we attempted to simply clarify what was there, but then we realized there was no agreement on what was there in the first place. |
| 78 | + |
| 79 | +### Q: Why doesn't the spec mention or behave like HTML anymore? |
| 80 | + |
| 81 | +A: While there are [unresolved questions around HTML analogies](https://github.com/json-schema-org/json-schema-spec/issues/294), we came to a consensus that the existing analogies caused more harm than good, for two reasons: |
| 82 | + |
| 83 | +1. The change between draft-03 and draft-04 to let `"method"` indicate any HTTP method instead of HTML's `<form method="...">` "get" and "post" values broke the original analogy to HTML, and trying to change it back was not well-received |
| 84 | +2. Only being able to use `"schema"` and `"encType"` for either the URI query string (but no other part of the URI) or the request body, but not having any way to work with both at once, was overly restrictive for API design |
| 85 | + |
| 86 | +#### Splitting `"schema"` |
| 87 | + |
| 88 | +Instead of having `"schema"` perform two different things depending on `"method"`, we split it into two keywords, one for each use. This allows using both simultaneously when needed, which is a use case not present in HTML forms. |
| 89 | + |
| 90 | +`"hrefSchema"` replaces the `"method": "get"` use, but leverages URI Template variables so that client data-driven dynamic URIs are no longer limited to the query string. `"encType"` is no longer needed with this approach. |
| 91 | + |
| 92 | +`"submissionSchema"` directly replaces the `"method": "post"` use, with `"submissionEncType"` replacing `"encType"`. |
| 93 | + |
| 94 | +#### Removing `"method"` |
| 95 | + |
| 96 | +Draft-05 tried to restore the draft-03 behavior of `"method"`, but feedback told us that users found the change very confusing. With `"schema"` split, the draft-05 behavior of `"method"` was no longer needed. |
| 97 | + |
| 98 | +We could have switched by to draft-04's `"method"` behavior, but in addition to producing more confusion from all of the back and forth, the draft-04 approach to `"method"` was not consistent with the rest of the LDO design anyway. Most notably, it caused problems with the usage of `"targetSchema"` as described above. |
| 99 | + |
| 100 | + |
| 101 | +### Q: So how do I indicate which HTTP methods are supported on a link? |
| 102 | + |
| 103 | +A: Ideally, this is implicitly conveyed by your link relation type, which is the primary indicator of semantics across machine-oriented hypermedia in general. [RFC 5988](https://tools.ietf.org/html/rfc5988) provides guidance on creating custom (a.k.a. "extension") link relations. |
| 104 | + |
| 105 | +Several URI schemes and namespaces, such as the [UUID namespace in the `urn:` scheme](https://tools.ietf.org/html/rfc4122), or the [`tag:` scheme](https://tools.ietf.org/html/rfc4151), are particularly suitable for creating unique identifiers. |
| 106 | + |
| 107 | +And of course, there are ways to detect this at runtime such as HTTP's `"Allow"` response header, or attempting a method and handling a `405 Method Not Allowed` error accordingly. |
| 108 | + |
| 109 | +### Q: No, really. How do I _explicitly_ indicate which HTTP methods are supported on a link? |
| 110 | + |
| 111 | +A: Pick a proposal such as [`"allow"`](https://github.com/json-schema-org/json-schema-spec/issues/73) or [`"usageHints"`](https://github.com/json-schema-org/json-schema-spec/issues/296) to implement as an extension keyword and let us know how it works for you. This will help us determine the right permanent solution in future drafts. |
| 112 | + |
| 113 | +### Q: If `"targetSchema"` is not the response, how do I describe responses? |
| 114 | + |
| 115 | +A: You should have hyper-schemas for your various success and error responses, but connecting them to links is more of a documentation question than a usage question: each response will indicate its own schema, so you don't need to know it in advance at runtime. |
| 116 | + |
| 117 | +There has never been a Hyper-Schema keyword to explicitly associate responses with operations such as HTTP methods. The use cases for this seem to be about generating API documentation, so this is most likely a candidate for a [JSON Schema API Documentation vocabulary](https://github.com/json-schema-org/json-schema-spec/issues/136). |
0 commit comments