diff --git a/NEWS.md b/NEWS.md index f74eed63a8..f76376b9ca 100644 --- a/NEWS.md +++ b/NEWS.md @@ -118,6 +118,8 @@ 10. `fwrite()` returns a clearer error message when `na = data.frame()` is used, [#7866](https://github.com/Rdatatable/data.table/issues/7866). Thanks @mcol for the report and the fix. +11. `fread()` will now detect and automatically trim [zsv](https://codeberg.org/jangorecki/zsv) metadata embedded in column names of a csv file. [#7963](https://github.com/Rdatatable/data.table/pull/7963) + ## data.table [v1.18.6.1](https://github.com/Rdatatable/data.table/milestone/47) (22 August 2026) diff --git a/R/fread.R b/R/fread.R index 4aa6a01ed0..9e66e19d8a 100644 --- a/R/fread.R +++ b/R/fread.R @@ -370,6 +370,21 @@ yaml=FALSE, tmpdir=tempdir(), tz="UTC") } else { setattr(ans, "class", "data.frame") } + # detect zsv headers + zsv = isTRUE(base::endsWith(names(ans)[1L], "~zsv")) + if (zsv) { + truecolnames = function(x) { + pos = regexpr("~", x, fixed = TRUE) + i = pos > 0L + if (!sum(i)) internal_error("truecolnames should be called only when ~zsv marker was detected in first column name") # nocov + ans = x + ans[i] = substr(x[i], 1L, pos[i] - 1L) + ans + } + if (verbose) + catf('Detected zsv header format: trim zsv to csv\n') + setnames(ans, truecolnames(names(ans))) + } # #1027, make.unique -> make.names as spotted by @DavidArenberg if (check.names) { setattr(ans, 'names', make.names(names(ans), unique=TRUE)) diff --git a/inst/tests/tests.Rraw b/inst/tests/tests.Rraw index 3a446a8136..e8c145cc86 100644 --- a/inst/tests/tests.Rraw +++ b/inst/tests/tests.Rraw @@ -22014,3 +22014,17 @@ test(2388.24, nafill(c("a", NA, NA, "z"), type="nocb", limit=1), c("a", NA, "z", x = seq(as.IDate("1999-01-01"), as.IDate("2027-12-31"), by="day") test(2389.1, isoweek(x), as.integer(format(x, "%V"))) test(2389.2, isoyear(x), as.integer(format(x, "%G"))) + +# zsv format handling +fwrite(fread(text = "id~u8-index~0~zsv,year~u16-partition{2024@0|2025@30|2026@45},city~str,amount~f64 +1,2024,Bern,2.5 +2,2024,Basel, +4,2025,Zug,1.5 +5,2026,Bern,3"), zsv<-tempfile(fileext=".zsv")) +test(2390.01, names(fread(zsv)), c("id","year","city","amount")) +fwrite(fread(text = "id~u8-index~0~csv,year~u16-partition{2024@0|2025@30|2026@45},city~str,amount~f64 +1,2024,Bern,2.5 +2,2024,Basel, +4,2025,Zug,1.5 +5,2026,Bern,3"), csv<-tempfile(fileext=".csv")) +test(2390.02, names(fread(csv)), c("id~u8-index~0~csv","year~u16-partition{2024@0|2025@30|2026@45}","city~str","amount~f64"))