Mostly the fact that it's based on Markdown, which makes the raw specification far easier to read with a text editor than OpenAPI/Swagger. Markdown also permits styling in the various descriptive portions, which makes for superior documentation.
I ran into some deficiencies, though, at least with the parser I was using with Node/TS -- IIRC (and it's been a few years), I wasn't able to specify a wide variety of disparate responses (e.g., an HTTP 200 with an application/json Content-Type header, an HTTP 200 with a text/plain Content-Type header, and an HTTP 400 response with an X-Error-Code header). Since API Blueprint was introduced, the tooling around OpenAPI has improved dramatically and it's become a de facto standard, so I'd probably avoid API Blueprint for anything serious.
It's unfortunate, though, because I really liked the idea.
GitHub repo is marked as archived by the owner. Is this project still alive?
The company behind it got bought by Oracle, so all the public stuff was archived. Not sure if there is still any community around it after that.
Confused why this is posted now - the project has been dead for years it seems..
I may have missed it, but wondering what the advantages are over OpenAPI/Swagger?
Mostly the fact that it's based on Markdown, which makes the raw specification far easier to read with a text editor than OpenAPI/Swagger. Markdown also permits styling in the various descriptive portions, which makes for superior documentation.
I ran into some deficiencies, though, at least with the parser I was using with Node/TS -- IIRC (and it's been a few years), I wasn't able to specify a wide variety of disparate responses (e.g., an HTTP 200 with an application/json Content-Type header, an HTTP 200 with a text/plain Content-Type header, and an HTTP 400 response with an X-Error-Code header). Since API Blueprint was introduced, the tooling around OpenAPI has improved dramatically and it's become a de facto standard, so I'd probably avoid API Blueprint for anything serious.
It's unfortunate, though, because I really liked the idea.
It’s interesting (and of course understandable) how all the effort goes into specifying RMM level 2 but not level 3 APIs.
Looks to be an abandoned project based on the last Github update being 3 years ago.