Remove uses of at-async in favor of at-spawn #1039
Conversation
Codecov Report
@@ Coverage Diff @@
## master #1039 +/- ##
==========================================
- Coverage 85.59% 85.24% -0.35%
==========================================
Files 32 32
Lines 3020 3023 +3
==========================================
- Hits 2585 2577 -8
- Misses 435 446 +11
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
quinnj
changed the title
|
Tests currently failing until JuliaServices/ConcurrentUtilities.jl#22 is merged and tagged. |
…on" cooperating tasks to be sticky to the thread that is running them or the parent's task's thread.
Drvi
left a comment
Drvi
left a comment
There was a problem hiding this comment.
Looks good to me:) I think the @async -> @spawn changes are not the most complicated, it's more the switch to OpenSSL and the changes to exception handling which are more difficult to access. But since these are all based on your experiments with large workload, I believe they'll make nice improvements in terms of performance and log deduplication.
| function exceptionlayer(handler) | ||
| return function exceptions(stream; status_exception::Bool=true, logerrors::Bool=false, kw...) | ||
| res = handler(stream; logerrors=logerrors, kw...) | ||
| return function exceptions(stream; status_exception::Bool=true, timedout=nothing, logerrors::Bool=false, logtag=nothing, kw...) |
There was a problem hiding this comment.
IIUC, timedout kwarg is strictly internal thing, the user shouldn't provide their own TimedOut struct to a HTTP.get? If that is the case, we might want to call it _timedout just to make it clearer that this is an internal thing.
#1038