[PATCH] [cut] Treat consecutive delimiters as one in cut binary

[Joe Maimon]

When using the -d option I find it frustrating that arbitrary whitespace
ruins the field count.

witness this script (new option -z)(one line):

dig -x $addr +noall +answer | grep -v '^;' | grep -v '$^' | expand | cut
-d' ' -f5 -z

With the new option in this patch, you can rely on the behavior to
produce a hostname if there is one. Otherwise you cant.

If I am being naive, I apologize.

Thanks,

Joe M





diff -urN coreutils-5.2.0/man/cut.1 coreutils-5.2.0-jm/man/cut.1
--- coreutils-5.2.0/man/cut.1   Sun Feb  8 15:14:24 2004
+++ coreutils-5.2.0-jm/man/cut.1        Fri Feb 20 02:23:40 2004
@@ -35,6 +35,8 @@
 \fB\-\-output\-delimiter\fR=\fISTRING\fR
 use STRING as the output delimiter
 the default is to use the input delimiter
+.HP
+\fB\-z\fR, \fB\-\-squeeze\-delimiters\fR treat consecutive delimiters as one
 .TP
 \fB\-\-help\fR
 display this help and exit
Binary files coreutils-5.2.0/src/.cut.c.swp and 
coreutils-5.2.0-jm/src/.cut.c.swp differ
diff -urN coreutils-5.2.0/src/cut.c coreutils-5.2.0-jm/src/cut.c
--- coreutils-5.2.0/src/cut.c   Wed Jan 21 17:27:02 2004
+++ coreutils-5.2.0-jm/src/cut.c        Fri Feb 20 02:23:35 2004
@@ -129,6 +129,9 @@
 /* The delimeter character for field mode. */
 static unsigned char delim;
 
+/* Whether we treat consecutive delimiters as one (squeeze) */
+static unsigned char delim_squeeze = false;
+
 /* True if the --output-delimiter=STRING option was specified.  */
 static bool output_delimiter_specified;
 
@@ -166,6 +169,7 @@
   {"delimiter", required_argument, 0, 'd'},
   {"only-delimited", no_argument, 0, 's'},
   {"output-delimiter", required_argument, 0, OUTPUT_DELIMITER_OPTION},
+  {"squeeze-delimiters", no_argument, 0, 'z'},
   {GETOPT_HELP_OPTION_DECL},
   {GETOPT_VERSION_OPTION_DECL},
   {0, 0, 0, 0}
@@ -206,6 +210,9 @@
       --output-delimiter=STRING  use STRING as the output delimiter\n\
                             the default is to use the input delimiter\n\
 "), stdout);
+      fputs (_("\
+  -z, --squeeze-delimiters treat consecutive delimiters as one\n\
+"), stdout);
       fputs (HELP_OPTION_DESCRIPTION, stdout);
       fputs (VERSION_OPTION_DESCRIPTION, stdout);
       fputs (_("\
@@ -630,6 +637,12 @@
                }
              found_any_selected_field = true;
 
+              if (delim_squeeze)
+               {
+                 while ((c = getc (stream)) == delim);
+                 ungetc (c, stream);
+               }
+
              while ((c = getc (stream)) != delim && c != '\n' && c != EOF)
                {
                  putchar (c);
@@ -637,6 +650,12 @@
            }
          else
            {
+              if (delim_squeeze)
+               {
+                 while ((c = getc (stream)) == delim);
+                 ungetc (c, stream);
+               }
+
              while ((c = getc (stream)) != delim && c != '\n' && c != EOF)
                {
                  /* Empty.  */
@@ -741,7 +760,7 @@
   delim = '\0';
   have_read_stdin = false;
 
-  while ((optc = getopt_long (argc, argv, "b:c:d:f:ns", longopts, NULL)) != -1)
+  while ((optc = getopt_long (argc, argv, "b:c:d:f:nsz", longopts, NULL)) != 
-1)
     {
       switch (optc)
        {
@@ -788,6 +807,9 @@
 
        case 's':
          suppress_non_delimited = true;
+         break;
+       case 'z':
+         delim_squeeze = true;
          break;
 
        case_GETOPT_HELP_CHAR;