Dirk Eddelbuettel
Dirk Eddelbuettel
You. Are. The. Man. Thanks so much for unclogging this!
Rebuilt R-devel and can confirm CRAN `nanotime` behaves under the now-amended-and-refined R-devel. Will leave this (now known as 'non-issue') open until the issue cleared at CRAN too and close it...
Thank you @kurthornik . When I checked last night I still had five fails, now down to four so that may be a lagging count. I need to update the...
Thanks for confirming that, @mmaechler. These days I keep an open tab with 'my' report aggregation and I saw it trailing. In a week's time, I hope, we can see...
This got sorted out so the issue is now stale and can be closed.
The intent here is to permit compilation under C++20 with an option to support then-included `fmt`. That is good. But we can test that locally: ```diff modified src/Makevars @@ -1,3...
> The intent here is to permit compilation under C++20 with an option to support then-included fmt. That is good. But we can test that locally Disregard that. The CRAN...
I think there may be two distinct PRs here. Both are welcome. One is the change to src/formatter.cpp which we could also condition on C++20 or greater and which allows...
There may also be a conceptual "gotcha" here. We are really considering two things: - How is the `RcppSpdlog` package installed on the system? For this step _we do not...
I took a quick look at the 'returning reference to temporary' issue. It goes away if we return by value. ```diff modified src/formatter.cpp @@ -24,12 +24,12 @@ #ifdef SPDLOG_USE_STD_FORMAT template...