David Nowinsky
David Nowinsky
With the introduction of poetry #1348 , the Dockerfile and some of the README steps may not be up-to-date. Could we take some time to make sure it's easy for...
The actual implementation is based on pandas' eval, which is quite permissive (close to any valid python) but doesn't work with some modern dtypes, such as https://pandas.pydata.org/docs/user_guide/integer_na.html My suggestion would...
An SQL translation today is quite costly as it need to make some intermediate requests, mainly to get column names to be added to the select clause. We could remove...
Since the introduction of the `server/` module, weaverbird is becoming more than just a front-end project to translate pipelines to Mongo queries. We now have two different ways of implementing...
The scope of this project is not to support older browsers. Therefore, switching the target to ES6 seems a good idea to me: - it will make the transpilation faster...
A filter like _column "foo" equals "bar"_ will be translated as `{$match: {foo: "bar"}}`. Likewise, a filter like _column "foo" equals "$10"_ will be translated as `{$match: {foo: "$10"}}`. Except...
In step forms, I noticed no fixed convention for CSS classes. In other ToucanToco projects, we use BEM classes. It avoids confusion and name collision. I suggest we apply the...
For small datasets, transformations could be done client-side. This could be handy to: - fetch only some data once from a server and dispatch it across usages in the client...
Reviewing #504 raised the issue that the manner we tie the back-end service to the hosting store and the weaverbird module is quite complicated. I think it could be simplified...