Marks symbols in search-based usages as definitions#63509
Conversation
6105d29 to
d392f39
Compare
There was a problem hiding this comment.
Ah this is because of "typed nil". https://go.dev/play/p/Xn8hLhjQnOb
There was a problem hiding this comment.
Let's attach the counts for syntactic and search-based to the trace as an event (perhaps in the callers directly?)
There was a problem hiding this comment.
As we're doing this once per file wouldn't this be a bit spammy? I'd maybe roll these up into a per-request event?
There was a problem hiding this comment.
Let's log the error instead of dropping it silently?
There was a problem hiding this comment.
That would lead to us double logging any error, as we've run syntactic search already before getting to this. Still in favor of logging here?
d392f39 to
3a739ea
Compare
|
As this PR sits on top of #63464 I've fixed the feedback pertaining to that PR over there and rebased this one. |
3a739ea to
49aff24
Compare
a9baaaa to
02ccd2d
Compare
49aff24 to
62b1b0d
Compare
f3b044f to
dd036d6
Compare
a1abfe1 to
e2af623
Compare
2003fd3 to
e5d6a66
Compare
Closes https://linear.app/sourcegraph/issue/GRAPH-712/determine-usage-kind-for-search-based-usages-by-running-symbol-search
I've decided to run a separate symbol query for simplicity for now. I'll ask the search platform team if it would be more efficient to do it in a single query when I meet with them.
Test plan
Example query for the API console