The API we're consuming has a base URL we initially hit but are then redirected to version-specific URL where the API request is ultimately made. If we use the base URL with the plugin, we're getting a 401 error, because the authorization header is not being passed to the redirected URL. This is similar to the StackOverflow article here: https://stackoverflow.com/questions/28564961/authorization-header-is-lost-on-redirect

Postman solves this by having a checkbox to explicitly allow this if the user requires. I'm assuming it is slightly less secure, so having it off by default may make sense here as well.

We're using "Next page URL provided" with a path that contains the dot syntax. While we're not encountering an error, we're only getting one page's worth of data. Furthermore, if I pass a completely invalid string into the "Key to next request URL" field ("asdfasdf") the first page's data is returned without an error.

In a copy of the version 1.0.2 plugin, I was able to bypass this issue by commenting out lines 57-68 of pagination.py, though I don't suppose this is the ultimate solution.

I'm getting a response back from an API that looks like this:

{ "Data": { "XRefCode": "Data_Warehouse_Integration", "Rows": [ { "Employee_XRefCode": "5", "Job_ShortName": "Nurse", "Employee_FirstName": "Melissa" }, { "Employee_XRefCode": "7", "Job_ShortName": "Nurse", "Employee_FirstName": "Nick" } ] }, "Paging": { "Next": "https://url-for-next-page-here" } }

If I set the "Data extraction key" as "Data" I get just two values back, XRefCode and Rows, which I'd expect. However, when I try to then drill to the next level to get the row data via "Data.Rows" I get an exception that no data is found for the extraction key. Am I using the wrong syntax for getting down to the nested Data in this format or is there a bug?

This is a great plugin, but it's difficult to use in an enterprise environment since only internal urls can be reached without going through a proxy. Ideally, the underlying requests proxyDict can be set from the UI: http, https, ftp. Then the proxyDict can simply be passed if set for all underlying requests, allowing external urls to be reached.

We've got a requirement to enable an Oauth flow for authentication before data can be retrieved. In order to accomplish this, two API calls needs to be made, first, one using username/password auth to get a Bearer Token, then the second call using the Bearer Token to actually get the data from the API. To accomplish this, we could chain two calls to the plugin, or, make the initial API call to get the Bearer token using a python component. The problem is, there's no way to "pass in" the retrieved Bearer Token into the API Connect plugin at runtime. To enable this, we'd like to be able to set the Bearer Token value in the plugin to ${project_variable_name} and have the Bearer Token be pulled from the project (or global) variables.

Pagination mechanism "per page" starts with 1.

But api may start with page 0.

Would it be possible to set starting index?

I suppose that the paging stop incrementing when result become empty.

But with API that still send first page, event sending "pagingIndex=123...", the paging increment never stop (until server api drop for exceeding quota).

{ "content": "{\"totalCount\": 1, \"pageIndex\": 0, \"limit\": 50, \"isLastPage\": true, \"results\": .... }

Would it be possible to create parameter for "totalCount" or "isLastPage" key to check ?

I'm pulling data from a REST API that correctly returns the boolean values in Json as true and false (starting in lower case). However, when the API Connect plugin finishes with the data and puts them in a DSS Dataset the values have True and False (upper case). This makes parsing the Raw return more difficult. One has to pre-process the results before one can unnest the resulting JSON.
--Tom

The client is calling a GET request instead of a POST or PUT request, even when selecting the POST or PUT methods in the GUI.

Configuration:

Server response ("read", should be "create" or "update"):