I am working on a AWS lambda application that would allow me to backup data from the workbooks of honeycode, using honeycode API. And It is being developed using Java.
i took the AWS Java SDK for Honeycode and, while I made it work, I noticed there are some improvements that could be done here in Honeycode.
First, the SDK needs the workbook ID to connect to it.
It was hard to find it from the user interface.
I had to extract it from one of the urls when opening an app linked to that workbook.
It would make things easier if the workbook id is visible somewhere clear on the user interface of Honeycode.
Second, Honeycode seems to work only on region US_WEST_2.
This was something i did not notice initially. Again, i had to extract that info from the url of an app.
Java SDK client for Honeycode is built like any other client, and so it has a "defaultClient" method that would create the client based on the setup of the AWS lambda.
However, in my case, all my code and services are running in Stockholm region, and so the client was trying to talk to Honeycode API service in Stockholm, and miserably failing.
Once i custom-build the client and setup the region to US_WEST_2, it does work as expected.
Perhaps it would be good to create particular threads on this community that would tackle programming-language-specific issues, like those i discuss above.
Those are some suggestions I noticed while working with the Java SDK.
Hope it helps.
Thanks for your improvement requests here! Great idea too to create more discussions or help around APIs or SDKs; our team can help create spaces for that here.
For finding the workbook ID, you can right-click the screen to quickly grab the ARN and IDs:
In regards to additional regions, this is on radar for our team and we aim to support additional regions in the near future. Thanks for providing your solution on here on setting up the region, as that may be helpful to other users on here as well.
Hope that info is helpful. Know too that I'll pass on your feedback to our team of your experience. It's always helpful to hear thoughts from our users, so please let us know if anything else comes to mind.
Hi Alyssa.
Thank you for the information. I did not notice that option to get ARNs and IDs.
Following on the subject regarding Honeycode Java SDK, i have noticed that to do operations against Honeycode, it is required to have table ids, column ids, etc.
I do see those are available via API but it does require a call to the API to get such values for a particular table, in case of column ids, for example, and then a second call to do the actual operation, such create, update or delete some row, or cell value.
Can developers be sure that IDs such as workbook id, table id or column id, are constant and wont change. If so, it is possible to get those values once and hard-code them on the application, so that API is not called unnecessary to get ids and then again for operations.
Can we developers acknowledge that such ids are static?
Yes, the IDs are static, so you'd only need to retrieve them once either by right-clicking the screen or through API calls.
For reference for others that may see this post, you can retrieve table IDs and column IDs with the following API calls: Table Metadata APIs - Amazon Honeycode.
