Merge branch 'master' into fix-#7354

Author: sisyphuswastaken

.ci/atime/tests.R

@@ -27,16 +27,14 @@ jobs:
           # jobs (mainly test-coverage) to run on every commit in PRs so as to not slow down dev.
           # GHA does run these jobs concurrently but even so reducing the load seems like a good idea.
           - {os: windows-latest, r: 'devel'}
-          - {os: macos-15-intel, r: 'release'}    
-          - {os: macos-15, r: 'release'}    
+          - {os: macos-15-intel, r: 'release'}
+          - {os: macos-15, r: 'release'}
           # TODO(remotes>2.5.0): Use 24.04[noble?]
-          - {os: ubuntu-22.04, r: 'release', rspm: "https://packagemanager.rstudio.com/cran/__linux__/jammy/latest"}
-          # - {os: ubuntu-22.04,   r: 'devel', rspm: "https://packagemanager.rstudio.com/cran/__linux__/jammy/latest", http-user-agent: "R/4.1.0 (ubuntu-22.04) R (4.1.0 x86_64-pc-linux-gnu x86_64 linux-gnu) on GitHub Actions" }
+          - {os: ubuntu-22.04, r: 'release'}
+          # - {os: ubuntu-22.04,   r: 'devel'}
           #   GLCI covers R-devel; no need to delay contributors in dev due to changes in R-devel in recent days
 
     env:
-      R_REMOTES_NO_ERRORS_FROM_WARNINGS: true
-      RSPM: ${{ matrix.config.rspm }}
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
       _R_CHECK_RD_CHECKRD_MINLEVEL_: -Inf
 
@@ -47,29 +45,6 @@ jobs:
         with:
           r-version: ${{ matrix.config.r }}
 
-
-      - name: Query dependencies
-        run: |
-          install.packages('remotes')
-          saveRDS(remotes::dev_package_deps(dependencies = TRUE), ".github/depends.Rds", version = 2)
-          writeLines(sprintf("R-%i.%i", getRversion()$major, getRversion()$minor), ".github/R-version")
-        shell: Rscript {0}
-
-      - name: Restore R package cache
-        uses: actions/cache@v5
-        with:
-          path: ${{ env.R_LIBS_USER }}
-          key: ${{ runner.os }}-${{ runner.arch }}-${{ hashFiles('.github/R-version') }}-1-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ runner.os }}-${{ runner.arch }}-${{ hashFiles('.github/R-version') }}-1-
-
-      - name: Install system dependencies
-        if: runner.os == 'Linux'
-        run: |
-          while read -r cmd
-          do
-            eval sudo $cmd
-          done < <(Rscript -e 'writeLines(remotes::system_requirements("ubuntu", "22.04"))')
-
       - name: Install R Package Build Dependencies on MacOS, from https://github.com/stan-dev/cmdstanr/pull/1072/files
         if: runner.os == 'macOS'
         uses: r-hub/actions/setup-r-sysreqs@v1
@@ -90,23 +65,8 @@ jobs:
             fi
           fi # otherwise R-bundled runtime is fine
 
-      - name: Install dependencies
-        run: |
-          remotes::install_deps(dependencies = TRUE)
-          remotes::install_cran("rcmdcheck")
-        shell: Rscript {0}
-
-      - name: Check
-        env:
-          _R_CHECK_CRAN_INCOMING_REMOTE_: false
-        run: |
-          options(crayon.enabled = TRUE)
-          rcmdcheck::rcmdcheck(args = c("--no-manual", "--as-cran"), error_on = "warning", check_dir = "check")
-        shell: Rscript {0}
+      - uses: yihui/actions/setup-r-dependencies@HEAD
 
-      - name: Upload check results
-        if: failure()
-        uses: actions/upload-artifact@main
+      - uses: yihui/actions/check-r-package@HEAD
         with:
-          name: ${{ runner.os }}-r${{ matrix.config.r }}-results
-          path: check
+          check-args: "--no-manual --as-cran"

.github/workflows/R-CMD-check.yaml

@@ -58,7 +58,7 @@ jobs:
           Rscript -e 'tools::write_PACKAGES("public/src/contrib", fields="Revision")'
       - name: upload
         if: github.ref == 'refs/heads/master'
-        uses: actions/upload-pages-artifact@v4
+        uses: actions/upload-pages-artifact@v5
         with:
           path: "public"
       - name: deploy

.github/workflows/pkgup.yaml

@@ -30,6 +30,8 @@
 
 5. `tables()` can now optionally report `data.table` objects stored one level deep inside list objects when `depth=1L`, [#2606](https://github.com/Rdatatable/data.table/issues/2606). Thanks @MichaelChirico for the report and @manmita for the PR
 
+6. `yearqtr()` and `yearmon()` now gain an optional format specifier [#7694](https://github.com/Rdatatable/data.table/issues/7694). 'numeric' is the default, which preserves the original behavior, but 'character' formats `yearqtr()` as YYYYQ# (e.g. 2025Q2) and `yearmon()` as YYYYM## (e.g. 2025M02, 2025M10). Thanks to @jan-swissre for the report and @LunaticSage218 for the implementation.
+
 ### BUG FIXES
 
 1. `fread()` with `skip=0` and `(header=TRUE|FALSE)` no longer skips the first row when it has fewer fields than subsequent rows, [#7463](https://github.com/Rdatatable/data.table/issues/7463). Thanks @emayerhofer for the report and @ben-schwen for the fix.
@@ -50,6 +52,8 @@
 
 9. `fread()` no longer replaces a literal header column name `"NA"` with an auto-generated `Vn` name when `na.strings` includes `"NA"`, [#5124](https://github.com/Rdatatable/data.table/issues/5124). Data rows still continue to parse `"NA"` as missing. Thanks @Mashin6 for the report and @shrektan for the fix.
 
+10. `fread()` no longer misreads dates with negative years, [#7704](https://github.com/Rdatatable/data.table/issues/7704). Thanks to @kevinushey for the report and @aitap for the fix.
+
 ### Notes
 
 1. {data.table} now depends on R 3.5.0 (2018).

NEWS.md

@@ -365,8 +365,30 @@ isoyear = function(x) as.integer(format(as.IDate(x), "%G"))
 month   = function(x) convertDate(as.IDate(x), "month")
 quarter = function(x) convertDate(as.IDate(x), "quarter")
 year    = function(x) convertDate(as.IDate(x), "year")
-yearmon = function(x) convertDate(as.IDate(x), "yearmon")
-yearqtr = function(x) convertDate(as.IDate(x), "yearqtr")
+yearmon = function(x, format = c("numeric", "character")) {
+  format = match.arg(format)
+  x_as_idate = as.IDate(x)
+  ymon = convertDate(x_as_idate, "yearmon")
+  if (format == "numeric") return(ymon)
+  ans = rep(NA_character_, length(x_as_idate))
+  ok = !is.na(x_as_idate)
+  yr = floor(ymon[ok])
+  mon = round((ymon[ok] - yr) * 12) + 1L
+  ans[ok] = sprintf("%dM%02d", as.integer(yr), as.integer(mon))
+  ans
+}
+yearqtr = function(x, format = c("numeric", "character")) {
+  format = match.arg(format)
+  x_as_idate = as.IDate(x)
+  yqtr = convertDate(x_as_idate, "yearqtr")
+  if (format == "numeric") return(yqtr)
+  ans = rep(NA_character_, length(x_as_idate))
+  ok = !is.na(x_as_idate)
+  yr = floor(yqtr[ok])
+  qtr = round((yqtr[ok] - yr) * 4) + 1L
+  ans[ok] = sprintf("%dQ%d", as.integer(yr), as.integer(qtr))
+  ans
+}
 
 convertDate = function(x, type) {
   type = match.arg(type, c("yday", "wday", "mday", "week", "month", "quarter", "year", "yearmon", "yearqtr"))

R/IDateTime.R

@@ -21639,3 +21639,23 @@ close(con)
 file.create(f <- tempfile())
 test(2367.6, fread(file(f)), data.table(), warning="Connection has size 0.")
 unlink(f)
+
+# negative years caused UB in leap year calculation, #7704
+x = fread("x\n-1-01-01")$x
+test(2368.1, year(x), -1L)
+test(2368.2, month(x), 1L)
+test(2368.3, mday(x),  1L)
+
+# yearqtr() and yearmon() could optionally output 2025Q4 format #7694
+x = c("1111-11-11", "2019-01-01", "2019-02-28", "2019-03-01", "2019-12-31", "2020-02-29", "2020-03-01", "2020-12-31", "2040-01-01", "2040-12-31", "2100-03-01", NA)
+test(2369.1, yearqtr(x, format="numeric"), c(1111.75, 2019, 2019, 2019, 2019.75, 2020, 2020, 2020.75, 2040, 2040.75, 2100, NA))
+test(2369.2, yearqtr(x, format="numeric"), yearqtr(x))  # numeric is the default, preserves backwards compatibility
+test(2369.3, yearqtr(x, format="character"), c("1111Q4", "2019Q1", "2019Q1", "2019Q1", "2019Q4", "2020Q1", "2020Q1", "2020Q4", "2040Q1", "2040Q4", "2100Q1", NA_character_))
+test(2369.4, yearqtr("2016-08-03 01:02:03.45", format="character"), "2016Q3")
+test(2369.5, yearqtr(NA, format="character"), NA_character_)
+
+test(2370.1, yearmon(x, format="numeric"), c(1111+10/12, 2019, 2019+1/12, 2019+2/12, 2019+11/12, 2020+1/12, 2020+2/12, 2020+11/12, 2040, 2040+11/12, 2100+2/12, NA))
+test(2370.2, yearmon(x, format="numeric"), yearmon(x))  # numeric is the default, preserves backwards compatibility
+test(2370.3, yearmon(x, format="character"), c("1111M11", "2019M01", "2019M02", "2019M03", "2019M12", "2020M02", "2020M03", "2020M12", "2040M01", "2040M12", "2100M03", NA_character_))
+test(2370.4, yearmon("2016-08-03 01:02:03.45", format="character"), "2016M08")
+test(2370.5, yearmon(NA, format="character"), NA_character_)

inst/tests/tests.Rraw

@@ -97,8 +97,8 @@ isoyear(x)
 month(x)
 quarter(x)
 year(x)
-yearmon(x)
-yearqtr(x)
+yearmon(x, format = c("numeric", "character"))
+yearqtr(x, format = c("numeric", "character"))
 
 }
 
@@ -115,6 +115,7 @@ yearqtr(x)
   the S3 generic.}
   \item{units}{one of the units listed for truncating. May be abbreviated.}
   \item{ms}{ For \code{as.ITime} methods, what should be done with sub-second fractions of input? Valid values are \code{'truncate'} (floor), \code{'nearest'} (round), and \code{'ceil'} (ceiling). See Details. }
+  \item{format}{For \code{yearmon} and \code{yearqtr}, either \code{"numeric"} (default) or \code{"character"}. \code{"character"} formats the result as \code{"2025M04"} for \code{yearmon} and \code{"2025Q2"} for \code{yearqtr}.}
 }
 \details{
 \code{IDate} is a date class derived from \code{Date}. It has the same
@@ -209,7 +210,11 @@ Similarly, \code{isoyear()} returns the ISO 8601 year corresponding to the ISO w
    for second, minute, hour, day of year, day of week,
    day of month, week, month, quarter, and year, respectively.
    \code{yearmon} and \code{yearqtr} return double values representing
-   respectively \code{year + (month-1) / 12} and \code{year + (quarter-1) / 4}.
+   respectively \code{year + (month-1) / 12} and \code{year + (quarter-1) / 4}
+   when \code{format = "numeric"} (the default). When \code{format = "character"},
+   they return character vectors of the form \code{"YYYYM##"} (e.g. \code{"2025M04"},
+   zero-padded for sortability) and \code{"YYYYQN"} (e.g. \code{"2025Q2"}) respectively,
+   with \code{NA} input returned as \code{NA_character_}.
 
    \code{second}, \code{minute}, \code{hour} are taken directly from
    the \code{POSIXlt} representation.
@@ -296,6 +301,11 @@ year(d2)
 isoweek(d2)
 isoyear(d2)
 
+# Character format for yearmon() and yearqtr()
+d3 = as.IDate(c("2019-01-01", "2019-12-31"))
+yearmon(d3, format = "character") # "2019M01" "2019M12"
+yearqtr(d3, format = "character") # "2019Q1"  "2019Q4"
+
 }
 \keyword{utilities}
 

man/IDateTime.Rd

@@ -1087,9 +1087,13 @@ static void parse_iso8601_date_core(const char **pch, int32_t *target)
   if (day == NA_INT32 || day < 1 || (day > (isLeapYear ? leapYearDays[month - 1] : normYearDays[month - 1])))
     return;
 
+  int32_t cycle_year = year % 400;
+  if (cycle_year < 0) cycle_year += 400;
+  int32_t cycle = (year - cycle_year) / 400;
+
   *target =
-    (year / 400 - 4) * cumDaysCycleYears[400] + // days to beginning of 400-year cycle
-    cumDaysCycleYears[year % 400] + // days to beginning of year within 400-year cycle
+    (cycle - 4) * cumDaysCycleYears[400] + // days to beginning of 400-year cycle
+    cumDaysCycleYears[cycle_year] + // days to beginning of year within 400-year cycle
     (isLeapYear ? cumDaysCycleMonthsLeap[month - 1] : cumDaysCycleMonthsNorm[month - 1]) + // days to beginning of month within year
     day - 1; // day within month (subtract 1: 1970-01-01 -> 0)