feat(event): Ability to create and sign events for use with js-ceramic #3

dbcfd · 2023-03-14T17:42:38Z

No description provided.

oed · 2023-03-24T09:51:33Z

What does "document" mean in this context?

Looks like you've mainly implemented "commits" so far, can we call them "events"?
See our spec, we now call them Init, Data, and Time Event
Documents as they exist in js-ceramic we wanted to not have part of this code base, instead we should mainly focus on streams and events

dbcfd · 2023-03-24T15:46:40Z

What does "document" mean in this context?

* Looks like you've mainly implemented "commits" so far, can we call them "events"?
  See our [spec](https://developers.ceramic.network/protocol/streams/event-log/), we now call them Init, Data, and Time Event

* Documents as they exist in js-ceramic we wanted to not have part of this code base, instead we should mainly focus on streams and events

Yep, this work so far focuses on commits, and then the format for stream id type document (3). I think what you're suggesting is

Rename the package to "ceramic-events"
Rename Commit to Event. Right now js-ceramic doesn't distinguish between an init event and a data event, but rather a deterministic commit (init event w/o data), signed commit (init event w/ data), and signed commit + stream_id (data event). It seems like if we make this implementation match the spec, it won't work with js-ceramic.

oed · 2023-03-24T16:28:20Z

Yes, I think that's roughly what I said.

It seems like if we make this implementation match the spec, it won't work with js-ceramic.

Not sure I follow this. js-ceramic should definitely be following the spec. cc @AaronGoldman

dbcfd · 2023-03-28T00:43:54Z

@oed Reworked PR to match more closely with terms used in spec. As implemented, this can be used with js-ceramic for creating, updating, and reading MIDs.

nathanielc

This looks great, I have a few comments/suggestions below.

My biggest question is how we can update the dag-jose crate to be of use here.

event/src/event.rs

nathanielc · 2023-03-31T17:36:22Z

core/src/jws.rs

+}
+
+#[derive(Serialize, Deserialize)]
+pub struct Jws {


I'd like to understand what needs to change in the dag-jose crate in order to be able to use its types here?

I understand this Jws struct also add the signing/verification logic however I do not see the dag-jose crate used anywhere in the code base. As such seems like we should fix that crate to make it amenable to this code as its purpose was to be a building block for Ceramic events.

Yeah, I was really surprised once I got everything implemented, that the events we pass to ceramic don't actually use dag-jose. They just have the data so that ipfs can encode them as dag-jose.

Seems like there's a mismatch there.

Cargo.toml

core/Cargo.toml

core/src/jwk.rs

Cargo.toml

event/src/args.rs

event/src/lib.rs

Cargo.toml

core/src/lib.rs

core/src/stream_id.rs

nathanielc

LGTM, feel free to merge when its green.

Cargo.toml

nathanielc · 2023-04-28T23:22:34Z

ok I updated this PR to pass CI. There are two major changes:

rebased against feat: add /metrics endpoint #32 since that PR contains a large refactor
fixed failing tests

Once #32 merged we can merge this PR

event/src/deterministic_event.rs

event/src/event.rs

oed · 2023-04-29T08:17:59Z

event/src/event.rs

+pub struct Event {
+    /// Cid of the data for the event
+    pub cid: Cid,
+    /// The data for the event to be encoded in the block
+    pub linked_block: DagCborEncoded,
+    /// JWS signature of the event
+    pub jws: Jws,
+}


Probably don't need to change this now, but events can also contain one cacaoBlock (currently), and possible multiple CACAO blocks in the future.

I wonder if a better structure for an event is just a CAR file with all the relevant IPLD blocks?

If not clear already: both jws and linked_block should be IPLD objects.

oed · 2023-04-29T08:22:07Z

core/src/stream_id.rs

+    /// Optional commit for this stream
+    pub commit: Option<Cid>,


Do we need to introduce "commitIds" yet (or at all)?

There is a painpoint around how commitIds are specified currently because they kind of break the multiformat nature of streamIds.

Me and @AaronGoldman talked about introducing a new way to reference a specific event inside of a streamId. This approach would allow streamIds to always be a real multiformat:
https://github.com/AaronGoldman/CIP/blob/stream-sets-synchronization/CIPs/CIP-124/cip-124.md#eventid-sort-order

I mainly have it there to make sure we can read cid's with commit id's specified. If we don't think we'll have stream id's with commits, I'd gladly remove it.

StreamIds with commits are not really used anywhere in the core protocol right now. They are mainly used to show the id of a commit to the user.

…amic Add event functionality (see https://developers.ceramic.network/protocol/streams/event-log/#events) which is similar to commit functionality currently found in ceramic.

dbcfd force-pushed the feat-document branch from 412899e to a5eaa0d Compare March 23, 2023 22:03

dbcfd force-pushed the feat-document branch from fc8f307 to 233c9f2 Compare March 24, 2023 15:33

dbcfd force-pushed the feat-document branch from 20354cb to f3a92a8 Compare March 28, 2023 00:41

dbcfd changed the title ~~Feat document~~ feat(event): Ability to create and sign events for use with js-ceramic Mar 29, 2023

nathanielc requested changes Mar 31, 2023

View reviewed changes

dbcfd force-pushed the feat-document branch from 0046ed5 to 7e9053b Compare March 31, 2023 21:05

nathanielc reviewed Mar 31, 2023

View reviewed changes

Cargo.toml Show resolved Hide resolved

nathanielc reviewed Mar 31, 2023

View reviewed changes

core/src/lib.rs Outdated Show resolved Hide resolved

nathanielc reviewed Mar 31, 2023

View reviewed changes

core/src/stream_id.rs Outdated Show resolved Hide resolved

dbcfd force-pushed the feat-document branch from 7e9053b to bd58152 Compare April 14, 2023 15:57

dbcfd requested a review from nathanielc April 27, 2023 15:53

nathanielc approved these changes Apr 27, 2023

View reviewed changes

Cargo.toml Outdated Show resolved Hide resolved

nathanielc force-pushed the feat-document branch 3 times, most recently from 32b47bd to b27da17 Compare April 28, 2023 23:21

oed reviewed Apr 29, 2023

View reviewed changes

dbcfd force-pushed the feat-document branch from 77e7f24 to f3b81f0 Compare May 1, 2023 21:11

feat(event): Add event (commit) libraries for creating events for cer…

0bff518

…amic Add event functionality (see https://developers.ceramic.network/protocol/streams/event-log/#events) which is similar to commit functionality currently found in ceramic.

dbcfd force-pushed the feat-document branch from 074dd4e to 0bff518 Compare May 3, 2023 01:11

nathanielc added this pull request to the merge queue May 3, 2023

Merged via the queue into main with commit eff535b May 3, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(event): Ability to create and sign events for use with js-ceramic #3

feat(event): Ability to create and sign events for use with js-ceramic #3

dbcfd commented Mar 14, 2023

oed commented Mar 24, 2023

dbcfd commented Mar 24, 2023

oed commented Mar 24, 2023

dbcfd commented Mar 28, 2023

nathanielc left a comment

nathanielc Mar 31, 2023

dbcfd Mar 31, 2023

nathanielc left a comment

nathanielc commented Apr 28, 2023

oed Apr 29, 2023

oed Apr 29, 2023

dbcfd May 1, 2023

oed May 2, 2023

feat(event): Ability to create and sign events for use with js-ceramic #3

feat(event): Ability to create and sign events for use with js-ceramic #3

Conversation

dbcfd commented Mar 14, 2023

oed commented Mar 24, 2023

dbcfd commented Mar 24, 2023

oed commented Mar 24, 2023

dbcfd commented Mar 28, 2023

nathanielc left a comment

Choose a reason for hiding this comment

nathanielc Mar 31, 2023

Choose a reason for hiding this comment

dbcfd Mar 31, 2023

Choose a reason for hiding this comment

nathanielc left a comment

Choose a reason for hiding this comment

nathanielc commented Apr 28, 2023

oed Apr 29, 2023

Choose a reason for hiding this comment

oed Apr 29, 2023

Choose a reason for hiding this comment

dbcfd May 1, 2023

Choose a reason for hiding this comment

oed May 2, 2023

Choose a reason for hiding this comment