Repository navigation
GH-20415: [Go] Kernel Input Type for RLE - #14146
Conversation
|
CC @zagto |
westonpace
left a comment
westonpace
left a comment
There was a problem hiding this comment.
Looks fairly straightforward, though I don't truly understand go
| if pErr := recover(); pErr != nil { | ||
| err = fmt.Errorf("arrow/concat: unknown error: %v", pErr) | ||
| switch e := pErr.(type) { | ||
| case error: |
There was a problem hiding this comment.
I don't really know go but why is it important to distinguish between these two error types? Adding unknown error doesn't seem to contribute too much to a user's understanding of the error.
There was a problem hiding this comment.
using %w wraps the error that was panic'd so that someone receiving the error can use errors.Is or errors.As if they desire to react differently based on the type of the error (such as arrow.ErrInvalid vs arrow.ErrIndex vs arrow.ErrNotImplemented etc...) basically it creates an error chain
In the case of the "unknown error" it means that we had a panic with a string rather than a proper error instance. Though I agree that the text "unknown error" doesn't contribute much to a user's understanding here, so i'll just get rid of that.
7bbac16 to
f26c1aa
Compare
563faca to
f38109e
Compare
|
I'll merge this at EOD if no one has any objections/comments |
zeroshade
changed the title
|
|
|
Benchmark runs are scheduled for baseline = d422137 and contender = 27cea43. 27cea43 is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
Uh oh!
There was an error while loading. Please reload this page.