Seems we are applying the wrong class sometimes to the deepest term in a hierarchy - maybe it has children in the database but none of these are facets? Or?

CommentFileSizeAuthor
20091217-bm1ngmgym4ncm1qa985k9mpg16.png29.15 KBpwolanin

Comments

jpmckinney’s picture

jpmckinney commented
Component: Code » User interface
janusman’s picture

janusman commented

Well there are several ways to approach this, but all have cons IMO:

So... we should style terms as expandable when:

  • Option A) the term_data table says the term has child terms (this is the current behavior). Pros: fast. Cons: misleading when there are no indexed nodes tagged with the child term(s)
  • B) the term_data table AND term_node says there are items marked with child terms. Pros: slower (a JOIN needed). Cons: complex (or impossible to figure out?) JOIN since we'd have to query in a way that takes into account the AS index setting (e.g., we might be keeping certain nodetypes, or there might be hooks altering what's indexed)
  • C) look at facets returned by Solr using the current query only. Pros: fastest. Cons: could also be misleading since we're asking for at most N facets and the children term(s) might be the N+x term(s).

I bet there are more but I just wanted to lay out some of the tradeoffs.

jpmckinney’s picture

jpmckinney commented
Status: Active » Closed (won't fix)

No one seems to care since 2009, and there is no solution that doesn't have cons.