Closed fedebongio closed 9 years ago
True, I'd say this is by design, more than a bug.
Validating RAMLs is a pretty costly operation, and we only validate before transforming the RAML with resource types, traits, default media types, etc.
to detect these errors, we would have to validate the RAML twice, which would mean a parsing time increase of 10% to 20% (in our past tests). Which for inline use in the api-designer it is quite significant.
The parser isn't validating fully, period. The fact that it isn't fully validating because it would be expensive to validate is perhaps a justified current design limitation, but it's still not doing what it needs to do, and what a user would have expected it to do. I'd like to keep this open until we can provide a better solution: perhaps offer more robust validation, if necessary through a second pass, when the UX allows it (say during idle cycles).
On Wed, Dec 3, 2014 at 12:14 PM, Damian Martinez Gelabert < notifications@github.com> wrote:
Closed #111 https://github.com/raml-org/raml-js-parser/issues/111.
— Reply to this email directly or view it on GitHub https://github.com/raml-org/raml-js-parser/issues/111#event-202191307.
Here is a micro RAML prepared to show the problem: