8000 Improve availability of completions for Data Catalog nodes · Issue #752 · tconbeer/harlequin · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
Improve availability of completions for Data Catalog nodes #752
Open
@tconbeer

Description

@tconbeer

Since #491, we lazy-load the catalog.

However, the major downside is that completions are only available for the catalog nodes that have been loaded already. Typically this would only include databases and schemas, not tables or columns (until their grandparent is expanded in the data table).

A couple of ideas for improving this:

  1. When triggering a member autocomplete, fetch the children. So if I type schema., fetch schema's tables. This will take a bit of doing, as right now I believe the Completions don't have a formal reference back to their nodes (and in fact, a single deduplicated completion can refer to multiple nodes). So figuring out what to fetch when we get the string schema could be hard.
  2. Cache the catalog (again), possibly only for the purposes of building completions. Given the above issues, I think it makes sense to cache the catalog tree, since that will be easier to invalidate when the nodes are loaded. (Though still not trivial).

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions

      0