Merge branch 'master' into issue2855

Author: venom1204

NEWS.md

@@ -105,6 +105,8 @@
 
 17. `t1 - t2`, where one is an `IDate` and the other is a `Date`, are now consistent with the case where both are `IDate` or both are `Date`, [#4749](https://github.com/Rdatatable/data.table/issues/4749). Thanks @George9000 for the report and @MichaelChirico for the fix.
 
+18. `fwrite` now allows `dec` to be the same as `sep` for edge cases where only one will be written, e.g. 0-row or 1-column tables. [#7227](https://github.com/Rdatatable/data.table/issues/7227). Thanks @MichaelChirico for the report and @venom1204 for the fix.
+
 ### NOTES
 
 1. The following in-progress deprecations have proceeded:

R/fwrite.R

@@ -44,7 +44,7 @@ fwrite = function(x, file="", append=FALSE, quote="auto",
     is.character(sep) && length(sep)==1L && (nchar(sep) == 1L || identical(sep, "")),
     is.character(sep2) && length(sep2)==3L && nchar(sep2[2L])==1L,
     is.character(dec) && length(dec)==1L && nchar(dec) == 1L,
-    dec != sep,  # sep2!=dec and sep2!=sep checked at C level when we know if list columns are present
+    `dec and sep must be distinct whenever both might be needed` = (!NROW(x) || NCOL(x) <= 1L || dec != sep),  # sep2!=dec and sep2!=sep checked at C level when we know if list columns are present
     is.character(eol) && length(eol)==1L,
     length(qmethod) == 1L && qmethod %chin% c("double", "escape"),
     length(compress) == 1L && compress %chin% c("auto", "none", "gzip"),

inst/tests/tests.Rraw

@@ -10977,9 +10977,9 @@ test(1732.7, fwrite(DT, quote='auto'), output='A,B\n,5\nNA,7\n"",0\nmonty,')
 test(1732.8, fwrite(DT, quote='auto', na="NA"), output='A,B\nNA,5\n"NA",7\n"",0\n"monty",NA')
 
 # dec=","
-test(1733.1, fwrite(data.table(pi),dec=","), error=base_messages$stopifnot("dec != sep"))
+# Test 1733.1 removed, see #7227
 test(1733.2, fwrite(data.table(c(1.2,-8.0,pi,67.99),1:4),dec=",",sep=";"),
-           output="V1;V2\n1,2;1\n-8;2\n3,14159265358979;3\n67,99;4")
+             output="V1;V2\n1,2;1\n-8;2\n3,14159265358979;3\n67,99;4")
 
 # fwrite implied and actual row.names
 DT = data.table(foo=1:3,bar=c(1.2,9.8,-6.0))
@@ -21593,3 +21593,12 @@ test(2336.3, all.equal(as.Date(t1) - t2, t1 - t2))
 test(2336.4, all.equal(as.Date(t2) - t1, t2 - t1))
 test(2336.5, all.equal(t1 - as.Date(t2), t1 - t2))
 test(2336.6, all.equal(t2 - as.Date(t1), t2 - t1))
+
+# fwrite: allow dec=',' with single column, #7227
+test(2337.1, fwrite(data.table(1), dec=","), NULL)
+if (getRversion() >= "4.0.0") { # rely on stopifnot(named = ...) for correct message
+  test(2337.2, fwrite(data.table(0.1, 0.2), dec=",", sep=","), error = "dec and sep must be distinct")
+}
+test(2337.3, is.null(fwrite(data.table(c(0.1, 0.2)), dec=",", sep="\t")))
+test(2337.4, is.null(fwrite(data.table(a=numeric(), b=numeric()), dec=",", sep=",")))
+test(2337.5, is.null(fwrite(data.table(a=numeric()), dec=",", sep=",")))

src/fmelt.c

@@ -103,12 +103,14 @@ static const char *concat(SEXP vec, SEXP idx) {
 // with missing inputs, and -1 in the positions with column names not
 // found. Column names not found will eventually cause error via
 // uniq_diff().
-SEXP chmatch_na(SEXP x, SEXP table){
-  SEXP ans;
-  PROTECT(ans = chmatch(x, table, -1));
-  for(int i=0; i<length(ans); i++){
-    if(STRING_ELT(x, i) == NA_STRING){
-      INTEGER(ans)[i] = NA_INTEGER;
+SEXP chmatch_na(SEXP x, SEXP table)
+{
+  SEXP ans = chmatch(x, table, -1);
+  PROTECT(ans);
+  int *restrict target = INTEGER(ans);
+  for (int i = 0; i < length(ans); i++) {
+    if (STRING_ELT(x, i) == NA_STRING) {
+      target[i] = NA_INTEGER;
     }
   }
   UNPROTECT(1);

src/nqrecreateindices.c

@@ -1,23 +1,24 @@
 #include "data.table.h"
+#include <string.h>
 
 // TODO: Add oxygen style comments and cleanup var names.
 // See other TODOs inside the function.
 SEXP nqRecreateIndices(SEXP xo, SEXP len, SEXP indices, SEXP nArg, SEXP nomatch)
 {
-  R_len_t n = INTEGER(nArg)[0], xn = length(xo);
+  const R_len_t n = INTEGER_RO(nArg)[0], xn = length(xo);
   SEXP ans, newstarts, newlen;
   ans = PROTECT(allocVector(VECSXP, 2));
   SET_VECTOR_ELT(ans, 0, (newstarts = allocVector(INTSXP, n)));
   SET_VECTOR_ELT(ans, 1, (newlen = allocVector(INTSXP, n)));
 
   int *inewlen = INTEGER(newlen);
-  const int *iindices = INTEGER(indices);
-  const int *ilen = INTEGER(len);
-  const int *ixo = INTEGER(xo);
-  const int inomatch = isNull(nomatch) ? 0 : INTEGER(nomatch)[0];
+  const int *iindices = INTEGER_RO(indices);
+  const int *ilen = INTEGER_RO(len);
+  const int *ixo = INTEGER_RO(xo);
+  const int inomatch = isNull(nomatch) ? 0 : INTEGER_RO(nomatch)[0];
   int *inewstarts = INTEGER(newstarts);
 
-  for (int i = 0; i < n; i++) inewlen[i] = 0;
+  memset(inewlen, 0, n * sizeof(int));
 
   // simplifying logic ... also fixes #2275
   for (int i = 0; i < length(indices); i++) {