-
Notifications
You must be signed in to change notification settings - Fork 2.1k
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
QueryInput struct broken after copying for pagination #309
Comments
I've pushed two related fixes to master which should resolve this issue. I fixed the copy semantics on nil slices and maps, and I made a similar fix on our path value operations which were causing the subsequent pagination to run forever because the nil slices in the output were being treated as valid tokens. This operation should now correctly paginate. Feel free to re-open if you're still running into this issue. Thanks for reporting! |
Thanks for promptly fix the case. It works well for me at least one week. :) |
Services --- * Synced the V2 SDK with latest AWS service API definitions. * Fixes aws#304 * Fixes aws#295 SDK Breaking changes --- This update includes multiple breaking changes to the SDK. These updates improve the SDK's usability, consistency. Client type name --- The API client type is renamed to `Client` for consistency, and remove stutter between package and client type name. Using Amazon S3 API client as an example, the `s3.S3` type is renamed to `s3.Client`. New API operation response type --- API operations' `Request.Send` method now returns a Response type for the specific operation. The Response type wraps the operation's Output parameter, and includes a method for the response's metadata such as RequestID. The Output type is an anonymous embedded field within the Output type. If your application was passing the Output value around you'll need to extract it directly, or pass the Response type instead. New API operation paginator utility --- This change removes the `Paginate` method from API operation Request types, (e.g. ListObjectsRequest). A new Paginator constructor is added that can be used to page these operations. To update your application to use the new pattern, where `Paginate` was being called, replace this with the Paginator type's constructor. The usage of the returned Paginator type is unchanged. ```go req := svc.ListObjectsRequest(params) p := req.Paginate() ``` Is updated to to use the Paginator constructor instead of Paginate method. ```go req := svc.ListObjectsRequest(params) p := s3.NewListObjectsPaginator(req) ``` Other changes --- * Standardizes API client package name to be based on the API model's `ServiceID`. * Standardizes API client operation input and output type names. * Removes `endpoints` package's service identifier constants. These values were unstable. Each API client package contains an `EndpointsID` constant that can be used for service specific endpoint lookup. * Fix API endpoint lookups to use the API's modeled `EndpointsID` (aka `enpdointPrefix`). Searching for API endpoints in the `endpoints` package should use the API client package's, `EndpointsID`. SDK Enhancements --- * Update CI tests to ensure all codegen changes are accounted for in PR (aws#183) * Updates the CI tests to ensure that any code generation changes are accounted for in the PR, and that there were no mistaken changes made without also running code generation. This change should also help ensure that code generation order is stable, and there are no ordering issues with the SDK's codegen. * Related aws#1966 * `service/dynamodb/expression`: Fix Builder with KeyCondition example (aws#306) * Fixes the ExampleBuilder_WithKeyCondition example to include the ExpressionAttributeNames member being set. * Fixes aws#285 * `aws/defaults`: Fix UserAgent execution environment key (aws#307) * Fixes the SDK's UserAgent key for the execution environment. * Fixes aws#276 * `private/model/api`: Improve SDK API reference doc generation (aws#309) * Improves the SDK's generated documentation for API client, operation, and types. This fixes several bugs in the doc generation causing poor formatting, an difficult to read reference documentation. * Fix aws#308 * Related aws#2617
I used dynamodb API, QueryPages, to get COUNT from table. I just found the content of QueryInput being injected some dummy field for second or further request. It is because the QueryInput of the further request is created by the following code in aws/request.go
awsutil.CopyOf makes any pointer field of the structure, QueryInput, have corresponding instance. It introduces DDB returns the following error:
Here is the printing result of QueryInput I used
The QueryInput of first page:
The QueryInput of second page:
Obviously, the second page uses the QueryInput object with many dummy field. The result of awsutil. CopyOf, here, is not what DynamoDB expected.
The text was updated successfully, but these errors were encountered: