Also, I noticed something peculiar. Here’s a response of 10 vs 20 paginated items returned by an API for an app that we’re working on. The 10 items are automatically formatted, but the 20 aren’t. Wondering why.
As for the pretty printing, we stopped pretty printer when the data is too large, due to several reasons:
it’s hard to know a good format for large data. For example, vec { 1;2;3 } .vs. vec {\n 1;\n2;\n3;\n };
The pretty printer we used is unfortunately not linear time, so for large data, it’s not very efficient;
Even with the same type, we may need different format for different cases. For example, both account_id and asset_content can be of type blob, but we may prefer to display account_id in hex, and asset_content as printable ascii. Ideally, we need a way to specify how to pretty print each field.
Ok. I’ll try to use idl2json for now. Thank you for the workaround.
P.S. I think it’s helpful to provide prebuilt binaries for idl2json. Figuring out another build toolchain is cumbersome and deters adoption
Created an issue here