-
Notifications
You must be signed in to change notification settings - Fork 177
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
FRC-0087: On-chain Deal Aggregation #879
Conversation
Reviewed for language and completeness. No substantial changes to the content or intent of the PR.
FRCs/frc-0070.md
Outdated
|
||
Recommended API interface for the two retrieval methods mentioned above: | ||
|
||
- Retrieval endpoint for Filecoin retrieval: GET /data/<pieceCID> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is there an example of a market implementation on filecoin that supports data at this path?
I believe boost would expose these pieces at /piece/<pieceCID>
, fwiw
We should also specify the expectation for how an aggregator supporting this standard would define where it's HTTPS endpoint for the data layer is located.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
there is no market impl on this today, lighthouse/spheron are supporting retrieval via Ipfs pinning instead of Filecoin, but they want to experiment retrieving from Filecoin in the near future.
Im open to aligning with boost endpoint for consistency (GET at /piece/pieceCID)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm confused about the scope of this PR.
- It's adding two new files (FRC-0067.md and FRC-0070.md)
- FRC-0067.md seems to correspond to existing FRC-0068.md
- FRC-0070.md claims to be FRC-0068 in the metadata, is using an unassigned and conflicting FIP number in filename, and has seemingly the same title as FRC-0068.
I think part of this is due to having been branched out of an older branch, but can we clean this up prior to substantive review? There should only be one RFC here, no unassigned numbers, and no recommitting of existing files. The metadata for the new RFC should match its content.
Co-authored-by: Jorge Soares <547492+jsoares@users.noreply.github.com>
Co-authored-by: Jorge Soares <547492+jsoares@users.noreply.github.com>
FRCs/frc-xxxx.md
Outdated
- Retrieval endpoint for Filecoin retrieval: GET /data/<pieceCID> | ||
- Retrieval gateway for IPFS retrieval: https://gateway/ipfs/<pieceCID> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think this is using the client's piece CID here, rather than the aggregated piece CID, correct? I like the use of commPc
above to differentiate from commPa
, perhaps that could be used here to disambiguate what this is referring to? They look very close to existing boost retrieval endpoints, with /piece/
instead of /data/
today, but those are for stored deal pieces, of commPa
(I don't think we do anything fancy with /piece/
retrievals on PoDSI deals today yet anyway).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes this is referring to client piece CID. We had a call with Will to align on these, so that it is consistent with the piece retrieval gateway FRC.
Note that this PR is not updated (@aashidham will be updating it soon), you can see the latest draft for retrieval section here #866
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed here b2fcc65
FRCs/frc-xxxx.md
Outdated
uint64 providerId; // The storage provider that is storing the data for the deal. | ||
uint64 marketActorId; // The actor ID of the storage market in which the deal representation is stored. dealId is scoped by marketActorId |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
uint64 providerId; // The storage provider that is storing the data for the deal. | |
uint64 marketActorId; // The actor ID of the storage market in which the deal representation is stored. dealId is scoped by marketActorId | |
uint64 providerId; // The storage provider that is storing the data for the deal. | |
uint64 marketActorId; // The actor ID of the storage market in which the deal representation is stored. dealId is scoped by marketActorId |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed here 522c607
FRCs/frc-xxxx.md
Outdated
event SubmitAggregatorRequestWithRaaS(uint256 indexed id, bytes cid, | ||
uint256 _replication_target, uint256 _repair_threshold, | ||
uint256 _renew_threshold); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
event SubmitAggregatorRequestWithRaaS(uint256 indexed id, bytes cid, | |
uint256 _replication_target, uint256 _repair_threshold, | |
uint256 _renew_threshold); | |
event SubmitAggregatorRequestWithRaaS(uint256 indexed id, bytes cid, uint256 _replication_target, uint256 _repair_threshold, uint256 _renew_threshold); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed here 522c607
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@honghaoq doesn't seem to be fixed yet, it's fixed in IOnchainDataAggregator
but not IOffchainDataAggregator
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done 1d850a2 Good catch:)
FRCs/frc-xxxx.md
Outdated
/** | ||
* @notice Function to submit a new file to the aggregator, specifing the raas parameters |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/** | |
* @notice Function to submit a new file to the aggregator, specifing the raas parameters | |
/** | |
* @notice Function to submit a new file to the aggregator, specifing the raas parameters |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed here 522c607
FRCs/frc-xxxx.md
Outdated
*/ | ||
function submitRaaS(bytes memory _cid, bytes memory _fetchLink, | ||
uint256 memory _replication_target, uint256 memory _repair_threshold, | ||
uint256 memory _renew_threshold); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
*/ | |
function submitRaaS(bytes memory _cid, bytes memory _fetchLink, | |
uint256 memory _replication_target, uint256 memory _repair_threshold, | |
uint256 memory _renew_threshold); | |
*/ | |
function submitRaaS(bytes memory _cid, bytes memory _fetchLink, uint256 memory _replication_target, uint256 memory _repair_threshold, uint256 memory _renew_threshold); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed here 522c607
Updated data retrieval section
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks like a reasonable reflection of how data segment indexed data is being structured and made accessible at least in boost. I am in favor of the FIP editors merging this FRC
Thanks @willscott, can you approve in the code review, looks like it needs 3 approvals to be able to merge |
I'm not able to approve - you need two fip editors / code owners, as I understand the process. |
I'm assigning number 0087 to this FRC. Please update the header and filename, and also add an appropriate line to the table in README.md |
Update as per Alex's comment
Updated: fdd0dc7 @anorth can you help approve it? my understanding is we need your approval so @aashidham can merge this PR, thanks |
Please rename the file to frc-0087. If you are able, please also change the README in this same PR so they land together. |
FRCs/frc-xxxx.md
Outdated
created: 2023-10-07 | ||
--- | ||
|
||
# FVM-Enabled Deal Aggregation |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This title doesn't match the title metadata. Please make them consistent, and also with the title listed in the README
changed title metadata
added 0087 to the readme
made authors list consistent
Title metadata, README and file name have all been altered to reflect the latest. @honghaoq your edits were in a separate patch branch, you have privileges to make changes directly on aggregation-frc branch. I do not have privileges to merge this PR Honghao, that is in the CODEOWNERS purview. @anorth have a look and let me know if things are good to merge |
Create aggregation FRC. More edits to come before finalization.