Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
MichaelChirico opened an issue on Rdatatable/data.table
Missing 1.16.4 from master (just NEWS?)
I think all of the relevant commits shipped with 1.16.4 are already dual-committed to `master`, but the NEWS items are still not merged to match CRAN.
MichaelChirico created a comment on an issue on Rdatatable/data.table
Could you provide some motivating benchmarks showing how inefficient it is to call `min()`/`max()` separately vs. calculate both in a single loop?
MichaelChirico created a comment on a pull request on github/docs
Take your time / no rush at all. Please prioritize other work first as appropriate.
MichaelChirico created a comment on an issue on Rdatatable/data.table
There is existing #6664 awaiting review, feel free to comment.
MichaelChirico created a comment on a pull request on github/docs
I have no context to offer improvements for the CI errors -- I leave it to the reviewer's judgment whether those should be addressed here or in a separate PR.
MichaelChirico created a review comment on a pull request on github/docs
Looks great, thanks for the refinement!
MichaelChirico created a review comment on a pull request on github/docs
💯 agree. I only chose the note as it was the first mention, didn't feel confident to go so far as to move things around in the article. And no problem at all w the more extensive edits. I raised...
MichaelChirico pushed 1 commit to setdt-s4 Rdatatable/data.table
- NEWS for separately-fixed bug eeeb61c
MichaelChirico opened an issue on Rdatatable/data.table
= assignment to under-allocated S4 slot fails
Another offshoot of #6575, but this issue also affects 1.16.4, so fixing it will be a separate bugfix. ```r setClass("DataTable", slots=c(x="data.table")) options(datatable.alloccol=0L) DT = ...
MichaelChirico pushed 1 commit to setdt-s4 Rdatatable/data.table
- Tweak test so that it would fail on master 8d2e958
MichaelChirico opened a pull request on Rdatatable/data.table
setDT() works on S4 slots
Closes #6701 (and possibly #6575). I'm not totally clear what broke exactly -- something about the change to `.shallow()` broke assignment back to an S4 slot. On a side note, I'm not totally ...
MichaelChirico created a branch on Rdatatable/data.table
setdt-s4 - R's data.table package extends data.frame:
MichaelChirico opened an issue on Rdatatable/data.table
setDT and [,:=] should share code for re-assigning data.tables
These two regions are close to identical: https://github.com/Rdatatable/data.table/blob/70c64ac08c6becae5847cd59ab1efcb4c46437ac/R/data.table.R#L1224-L1236 https://github.com/Rdatatable/data....
MichaelChirico opened an issue on Rdatatable/data.table
setDT fails on S4 slots
Extracted from debugging #6575: ```r library(methods) library(data.table) setClass("tst", slots = c(x = "data.frame")) T = new("tst", x = data.frame(a = 1)) setDT(T@x) class(T@x) # [1] ...
MichaelChirico opened an issue on tdhock/data.table-revdeps
Disable tiny test output colors
Observed in the logs: ``` test_restatapi.R.............. 69 tests [0;32mOK[0m test_restatapi.R.............. 69 tests [0;32mOK[0m test_restatapi.R.............. 69 tests [0...
MichaelChirico created a comment on an issue on markvanderloo/tinytest
> Color can be toggled with tt.pr.color Any consideration for auto-detecting whether color output is supported? {cli} (as well as {crayon}) does so, see [`cli::num_ansi_colors()`](https://cli.r-...
MichaelChirico opened a pull request on kjellpk/dbi.table
Use proper aggregation before passing to message() for reference.test(verbose=TRUE)
Encountered while debugging https://github.com/Rdatatable/data.table/issues/6671: ``` Modes: numeric, listLengths: 1, 4names for current but not for targetAttributes: < target is NULL, current ...
MichaelChirico opened a pull request on Rdatatable/data.table
Retain with=FALSE in j=lit:var cases
For #6671 (not closing because we should wait for the revdep checker to confirm the fix).
MichaelChirico created a branch on Rdatatable/data.table
with-false-num-var - R's data.table package extends data.frame:
MichaelChirico created a comment on an issue on Rdatatable/data.table
OK, here's a reprex: ```r DT = data.table(a=1, b=2:3) DT[, 1:b] ## master # [1] 1 2 # Warning message: # In 1:b : numerical expression has 2 elements: only the first used ## 1.16.4 #...
MichaelChirico opened a pull request on MichaelChirico/dbi.table
Use proper aggregation before passing to message() for reference.test(verbose=TRUE)
Encountered while debugging: ``` Modes: numeric, listLengths: 1, 4names for current but not for targetAttributes: < target is NULL, current is list >target is numeric, current is data.table ``...
MichaelChirico created a comment on an issue on Rdatatable/data.table
Yea, fundamentally you're abusing the S3 dispatch system & the special nature of the base namespace. Taking your example, we can immediately see that the _only_ exported methods are those in {ba...
MichaelChirico closed an issue on Rdatatable/data.table
as.data.table.matrix
In base R `as.data.frame.matrix` is exported and can be used to convert a table or an xtabs object (class `c("xtabs", "table")`) to a data frame with the same shape. In contrast `as.data.frame....
MichaelChirico created a comment on an issue on Rdatatable/data.table
> In base R as.data.frame.matrix is exported just a quick note I think there's some misunderstanding here. {base} namespace has no concept of "exported" (you'll see references to the "special...