+1 I double-checked why a JSON object was allowed in the first place. It looks like the original motivation was to support PROJJSON, but we switched to using projjson:<identifier> where <identifier> points to a table property.
On Wed, Mar 19, 2025 at 6:45 AM Eduard Tudenhöfner <etudenhoef...@apache.org> wrote: > +1 > > On Wed, Mar 19, 2025 at 2:30 AM Gang Wu <ust...@gmail.com> wrote: > >> Makes sense. >> >> +1 (non-binding) >> >> On Wed, Mar 19, 2025 at 8:07 AM Jia Yu <ji...@apache.org> wrote: >> >>> +1 (non-binding) >>> >>> Thank you! >>> >>> On 2025/03/19 00:01:00 Szehon Ho wrote: >>> > Hi everyone, >>> > >>> > While working on the reference implementation for Geometry/Geography >>> spec, >>> > we noticed some parts that can be simplified for this first version: >>> > >>> > 1. Default values should always be null (requires WKT serialization >>> > logic, for not many real world use cases) >>> > 2. JSON type serialization can be a json string, not a json object. >>> > >>> > I would like to raise a vote for this minor change to the V3 spec to >>> the >>> > geo types. The PR is: https://github.com/apache/iceberg/pull/12533. >>> > >>> > This vote will be open for at least 72 hours. >>> > >>> > [ ] +1 Add these simplifications to the V3 Geo spec >>> > [ ] +0 >>> > [ ] -1 I have questions and/or concerns >>> > >>> > Thanks, >>> > Szehon >>> > >>> >>
