You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The goal here was originally to support the full gamut of specs for IPFS gateways, but as Lidel points out below, we should probably stick to one since the car variants are SHOULD, until we get feedback from users that different is desired.
if deterministic order is supported, return explicit Content-Type
Note that those extra CAR features are SHOULD and not MUST. In practice gateway implementations may choose to only implement a subset of possible permutations of CAR parameters. for example, today in 2025Q1 we have two presets in the wild:
@SgtPooki If the purpose of this issue is to figure out what CAR should be produced by verified-fetch, probably a good bet is to try to follow what is done in boxo/gateway (deterministic DFS responses without duplicates) and have the same bytes, unless we have a reason to do something different in JS.
The goal here was originally to support the full gamut of specs for IPFS gateways, but as Lidel points out below, we should probably stick to one since the car variants are
SHOULD
, until we get feedback from users that different is desired.see https://specs.ipfs.tech/http-gateways/trustless-gateway/#car-order-request-query-parameter and https://specs.ipfs.tech/http-gateways/trustless-gateway/#car-order-content-type-parameter
The text was updated successfully, but these errors were encountered: