Tamer Abdulradi

1. Schema versioning has been implemented to look for both paths. 2. Schema location has been parametrised from schema-engine, but now hardcoded in the macro. 3. Now we have configuration...

Troy has two modes to load the schema: - the single-file mode, it looks for single file `resources/schema.cql` - the versioned mode, it looks for multiple files `resources/schema/01.cql`, `resources/schema/02.cql`, etc...

Ok, I've figured it out. It is a documentation issue. You need [this line](https://github.com/schemasafe/troy/blob/v0.5/examples/demo/build.sbt#L7) in your build.sbt to allow the macro to read the resources folder at compile time.

Before we can pick this task, we need to ensure 100% coverage under cql-ast and cql-parses sub-projects.

This task is on hold now. Please don't pick it up.

Depends on changing the return type of schema engine to include column names. It is not important at the moment, but very beneficial for the shapeless sub-project.

> The following datatype changes are not supported: > > Changing the type of a clustering column > Changing the type of a column on which an index is defined...

Hello @bennythomps, Thanks for reporting this bug. It is `.map(_ => true)` that triggered it. A workaround (for now) is to rewrite your example like the following: ``` class ApiTokenRecord(implicit...

In general avoid writing arbitrary code inside `withSchema` block. We only tested for specific cases, like `setConsistencyLevel`, the current Tree transformation code is not robust. For the specific case of...

Seems that [new documentation](https://cassandra.apache.org/doc/latest/cql/definitions.html#identifiers) allows an Identifier to be a reserved keyword. I can't find any mentions for rejecting keywords as part of cql syntax