kbendick commented on a change in pull request #3561: URL: https://github.com/apache/iceberg/pull/3561#discussion_r769859433
########## File path: core/src/main/java/org/apache/iceberg/rest/http/IcebergHttpResponse.java ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +package org.apache.iceberg.rest.http; + +import com.fasterxml.jackson.annotation.JsonCreator; +import com.fasterxml.jackson.annotation.JsonInclude; +import com.fasterxml.jackson.annotation.JsonProperty; + +/** + * All responses for the first version of the Iceberg REST API will be JSON objects, a top-level envelope object with two fields, + * `error` and `data`, which are themselves JSON objects. + * + * data: represents the JSON encoded response for a successful path or for a valid / expected failure. JSON representation of various response types + * error: a standardized obect on error, containing the error code, message, type, and additional metadata (an optional JSON object of metadata) as defined below. + * + * All responses for the REST catalog should be wrapped this way, vs using primitives. For example, for a ListTableResponse, listing tables under a namespace "accounting", + * we'd get a JSON object back like the following: + * + * { "data": { "identifiers": [ "accounting.tax", "accounting.currency_conversions"] }, "error", {} } + * + * If the namesapce `accounting` didn't request, the response from that call would have a body like the following, + * where the `code` 40401 is a two-part identifier: + * - HTTP response code: 404 + * - Two digit internal application defined error code for further detail: 01 - Namespace not found. + * + * { "data": {}, "error": { "message": "Failed to list tables. The Namespace 'accounting' does not exist", "type": "NamespaceNotFoundException", "code": 40401 } + * + * We could also embed the HTTP response code plainly by itself, without internally documented codes, as a separate field. I have found having a documented list of internal + * error codes to be very helpful previously, but am open to discussion on this. Review comment: > Re: type values, I personally tend to think that if the field itself is part of the spec, the spec would have to define how to interpret it. I am not sure keeping it "opaque" is going to be that much useful from the spec's perspective. I agree. Since we're using code to just be the HTTP status response code, I would say that `type` should be a well defined set we can look up. We might also choose to allow people to use others that will get ignored by the client but are logged or something. But for now I would say let's plan for it to be a well defined set of values, but not necessarily attempt to enumerate all of them just yet since we don't have about half of the basic Catalog AP in here yet. It's currently mostly just around namespaces and some of the table related ones will be handled next. Sound good? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
