join,sort,uniq: with -z, treat '\n' as a field separator

* NEWS: Mention the change in behavior.
* doc/coreutils.texi (newlineFieldSeparator): A new description,
referenced from ({join,sort,uniq} invocation).
* src/system.h (field_sep): A new inline function to determine
if a character is a field separator.
* src/join.c (usage): s/whitespace/blank/ to be more accurate
wrt which characters are field separators.
(xfields): s/isblank/field_sep/.
* src/sort.c (inittables): Likewise.
* src/uniq.c (find_field): Likewise.
* tests/misc/join.pl: Adjust -z test, and add a test/example
for processing the whole record with field processing.
* tests/misc/sort.pl: Add -z test cases, including case with '\n'.
* tests/misc/uniq.pl: Add -z -f test case with \n.

4 files changed, 15 insertions, 8 deletions

diff --git a/src/join.c b/src/join.c
index 8686428fb..9b25da667 100644
--- a/src/join.c
+++ b/src/join.c
@@ -194,7 +194,7 @@ Usage: %s [OPTION]... FILE1 FILE2\n\
               program_name);
       fputs (_("\
 For each pair of input lines with identical join fields, write a line to\n\
-standard output.  The default join field is the first, delimited by whitespace.\
+standard output.  The default join field is the first, delimited by blanks.\
 \n\
 "), stdout);
       fputs (_("\
@@ -284,19 +284,19 @@ xfields (struct line *line)
   else if (tab < 0)
     {
       /* Skip leading blanks before the first field.  */
-      while (isblank (to_uchar (*ptr)))
+      while (field_sep (*ptr))
         if (++ptr == lim)
           return;
 
       do
         {
           char *sep;
-          for (sep = ptr + 1; sep != lim && ! isblank (to_uchar (*sep)); sep++)
+          for (sep = ptr + 1; sep != lim && ! field_sep (*sep); sep++)
             continue;
           extract_field (line, ptr, sep - ptr);
           if (sep == lim)
             return;
-          for (ptr = sep + 1; ptr != lim && isblank (to_uchar (*ptr)); ptr++)
+          for (ptr = sep + 1; ptr != lim && field_sep (*ptr); ptr++)
             continue;
         }
       while (ptr != lim);
diff --git a/src/sort.c b/src/sort.c
index aca3b4231..575877d22 100644
--- a/src/sort.c
+++ b/src/sort.c
@@ -1275,9 +1275,9 @@ inittables (void)
 
   for (i = 0; i < UCHAR_LIM; ++i)
     {
-      blanks[i] = !! isblank (i);
+      blanks[i] = field_sep (i);
       nonprinting[i] = ! isprint (i);
-      nondictionary[i] = ! isalnum (i) && ! isblank (i);
+      nondictionary[i] = ! isalnum (i) && ! field_sep (i);
       fold_toupper[i] = toupper (i);
     }
 
diff --git a/src/system.h b/src/system.h
index c1c4a18a3..9898bc79c 100644
--- a/src/system.h
+++ b/src/system.h
@@ -155,6 +155,13 @@ enum
    errors that the cast doesn't.  */
 static inline unsigned char to_uchar (char ch) { return ch; }
 
+/* '\n' is considered a field separator with  --zero-terminated.  */
+static inline bool
+field_sep (unsigned char ch)
+{
+  return isblank (ch) || ch == '\n';
+}
+
 #include <locale.h>
 
 /* Take care of NLS matters.  */
diff --git a/src/uniq.c b/src/uniq.c
index 6f8cd4a70..0e118da9d 100644
--- a/src/uniq.c
+++ b/src/uniq.c
@@ -261,9 +261,9 @@ find_field (struct linebuffer const *line)
 
   for (count = 0; count < skip_fields && i < size; count++)
     {
-      while (i < size && isblank (to_uchar (lp[i])))
+      while (i < size && field_sep (lp[i]))
         i++;
-      while (i < size && !isblank (to_uchar (lp[i])))
+      while (i < size && !field_sep (lp[i]))
         i++;
     }