Project update: Frontend fixes
Sean Colsen
The frontend fixes project is off to a decent start.
So far been working on a draft PR to refactor the cell selection data structure. I’m pretty excited about this refactor because it will lay some groundwork for more easily making improvements that involve the cell selection — one of the buggiest areas of the front end. In the course of this refactoring work, I’ve taken a step back and thought carefully about all the ways that selections of cells can be manipulated with spreadsheets. For example, Ctrl+Down or Shift+Home. Keyboard shortcuts like those have not been on our radar because they are lower priority. But after this refactor they’ll be quite easy to implement. In the course of this refactor, I’ve also been noticing other unreported bugs with our current cell selection logic which I expect my PR to fix.
This PR will probably take at least another week, maybe two, but it should end up fixing the following issues (plus other unreported bugs too) when merged:
- Cell selection partially broken after selecting columns when table is filtered to be empty
- Do not select columns when hovering column headers in a table with no rows
- Weird cell selection behavior when dragging from cell to header
That’ll be the bulk of my project, as I spec’ed it. It’s possible I’ll finish early, but I’m also signed up to assist other projects too, so my pace will probably slow down once that work kicks in.
Sean Colsen
Weekly update on my Frontend fixes project…
Not much has changed since last week.
- Rajat and Pavish have both reviewed and approved the direction I’m heading with the cell selection refactor.
- I’ve made some additional progress on that refactor.
The remaining work within this project is dependent on that refactoring work.
Kriti Godey
Sean Colsen
Weekly update on my Frontend Fixes project.
You may recall that last week I sent two project updates: one for my “frontend fixes” project, and another for my involvement in Dom’s “backend fixes” project. Well for this week, I’ve re-organized the tickets so that my work is contained squarely within the “frontend fixes” project. After all, my work here is entirely fronted work.
As such, the Frontend Fixes project has grown in scope over the past week. My time has been occupied mostly by making column type inference optional, which has proven to be trickier than I originally predicted. I’m getting closer, but still not there yet.
In the past week, I also fixed a small regression Copying FK cells should copy the record summary instead of the PK value.
My work on improving the cell selection behavior is paused. I’m hopeful that we’ll finish the optional type inference work this cycle, but not very hopeful that we’ll finish the cell selection improvements. I’m expecting to have very little time to work on this project at all next week because most of my energy will be focused on reviewing the work of other projects.