head,tail: add the -z,--zero-terminated option

* doc/coreutils.texi: Reference the option description.
* src/head.c: Parameterize the delimiter character.
* src/tail.c: Likewise.
* tests/misc/head.pl: Add test case.
* tests/misc/tail.pl: Likewise.
* NEWS: Mention the new feature.

diff --git a/NEWS b/NEWS
index 192d8fa9e56d8a03bc4fe32855415529f195166a..eb7fc5633c30dc168be7a526c285fd87f55327da 100644 (file)
--- a/NEWS
+++ b/NEWS
@@ -39,6 +39,9 @@ GNU coreutils NEWS                                    -*- outline -*-
   Its status=progress output now uses the same format as ordinary status,
   perhaps with trailing spaces to erase previous progress output.
 
+  head, tail now have -z, --zero-terminated options to work with
+  NUL delimited items.
+
   md5sum now supports the --ignore-missing option to allow
   verifying a subset of files given a larger list of checksums.
   This also affects sha1sum, sha224sum, sha256sum, sha384sum and sha512sum.

diff --git a/doc/coreutils.texi b/doc/coreutils.texi
index 80e9a032dadf55c3038b4afc9d48104e5c1014be..a7a89ad98db9ead7fd27698714b81401b0eaaf8f 100644 (file)
--- a/doc/coreutils.texi
+++ b/doc/coreutils.texi
@@ -2817,6 +2817,8 @@ Never print file name headers.
 @opindex --verbose
 Always print file name headers.
 
+@optZeroTerminated
+
 @end table
 
 For compatibility @command{head} also supports an obsolete option syntax
@@ -3042,6 +3044,8 @@ every @var{number} seconds.
 @opindex --verbose
 Always print file name headers.
 
+@optZeroTerminated
+
 @end table
 
 For compatibility @command{tail} also supports an obsolete usage

diff --git a/src/head.c b/src/head.c
index a5405aae65872fd2cf9d9c7f3d825b38386a76f3..282c2ea8d9278da5c281f22774283820266b85f7 100644 (file)
--- a/src/head.c
+++ b/src/head.c
@@ -58,6 +58,9 @@ static bool presume_input_pipe;
 /* If true, print filename headers. */
 static bool print_headers;
 
+/* Character to split lines by. */
+static char line_end;
+
 /* When to print the filename banners. */
 enum header_mode
 {
@@ -90,6 +93,7 @@ static struct option const long_options[] =
   {"quiet", no_argument, NULL, 'q'},
   {"silent", no_argument, NULL, 'q'},
   {"verbose", no_argument, NULL, 'v'},
+  {"zero-terminated", no_argument, NULL, 'z'},
   {GETOPT_HELP_OPTION_DECL},
   {GETOPT_VERSION_OPTION_DECL},
   {NULL, 0, NULL, 0}
@@ -125,6 +129,9 @@ With more than one FILE, precede each with a header giving the file name.\n\
       fputs (_("\
   -q, --quiet, --silent    never print headers giving file names\n\
   -v, --verbose            always print headers giving file names\n\
+"), stdout);
+      fputs (_("\
+  -z, --zero-terminated    line delimiter is NUL, not newline\n\
 "), stdout);
       fputs (HELP_OPTION_DESCRIPTION, stdout);
       fputs (VERSION_OPTION_DESCRIPTION, stdout);
@@ -532,7 +539,7 @@ elide_tail_lines_pipe (const char *filename, int fd, uintmax_t n_elide,
       {
         char const *buffer_end = tmp->buffer + n_read;
         char const *p = tmp->buffer;
-        while ((p = memchr (p, '\n', buffer_end - p)))
+        while ((p = memchr (p, line_end, buffer_end - p)))
           {
             ++p;
             ++tmp->nlines;
@@ -581,7 +588,7 @@ elide_tail_lines_pipe (const char *filename, int fd, uintmax_t n_elide,
 
   /* If we read any bytes at all, count the incomplete line
      on files that don't end with a newline.  */
-  if (last->nbytes && last->buffer[last->nbytes - 1] != '\n')
+  if (last->nbytes && last->buffer[last->nbytes - 1] != line_end)
     {
       ++last->nlines;
       ++total_lines;
@@ -600,7 +607,7 @@ elide_tail_lines_pipe (const char *filename, int fd, uintmax_t n_elide,
       size_t n = total_lines - n_elide;
       char const *buffer_end = tmp->buffer + tmp->nbytes;
       char const *p = tmp->buffer;
-      while (n && (p = memchr (p, '\n', buffer_end - p)))
+      while (n && (p = memchr (p, line_end, buffer_end - p)))
         {
           ++p;
           ++tmp->nlines;
@@ -664,7 +671,7 @@ elide_tail_lines_seekable (const char *pretty_filename, int fd,
   const bool all_lines = !n_lines;
 
   /* Count the incomplete line on files that don't end with a newline.  */
-  if (n_lines && bytes_read && buffer[bytes_read - 1] != '\n')
+  if (n_lines && bytes_read && buffer[bytes_read - 1] != line_end)
     --n_lines;
 
   while (1)
@@ -679,7 +686,7 @@ elide_tail_lines_seekable (const char *pretty_filename, int fd,
           else
             {
               char const *nl;
-              nl = memrchr (buffer, '\n', n);
+              nl = memrchr (buffer, line_end, n);
               if (nl == NULL)
                 break;
               n = nl - buffer;
@@ -804,7 +811,7 @@ head_lines (const char *filename, int fd, uintmax_t lines_to_write)
       if (bytes_read == 0)
         break;
       while (bytes_to_write < bytes_read)
-        if (buffer[bytes_to_write++] == '\n' && --lines_to_write == 0)
+        if (buffer[bytes_to_write++] == line_end && --lines_to_write == 0)
           {
             off_t n_bytes_past_EOL = bytes_read - bytes_to_write;
             /* If we have read more data than that on the specified number
@@ -942,6 +949,8 @@ main (int argc, char **argv)
 
   print_headers = false;
 
+  line_end = '\n';
+
   if (1 < argc && argv[1][0] == '-' && ISDIGIT (argv[1][1]))
     {
       char *a = argv[1];
@@ -986,6 +995,10 @@ main (int argc, char **argv)
               header_mode = always;
               break;
 
+            case 'z':
+              line_end = '\0';
+              break;
+
             default:
               error (0, 0, _("invalid trailing option -- %c"), *a);
               usage (EXIT_FAILURE);
@@ -1006,7 +1019,7 @@ main (int argc, char **argv)
       argc--;
     }
 
-  while ((c = getopt_long (argc, argv, "c:n:qv0123456789", long_options, NULL))
+  while ((c = getopt_long (argc, argv, "c:n:qvz0123456789", long_options, NULL))
          != -1)
     {
       switch (c)
@@ -1039,6 +1052,10 @@ main (int argc, char **argv)
           header_mode = always;
           break;
 
+        case 'z':
+          line_end = '\0';
+          break;
+
         case_GETOPT_HELP_CHAR;
 
         case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);

diff --git a/src/tail.c b/src/tail.c
index 9007888fcd18c8383ec803cbb2e9d61b61bc0321..781adf200df150b295e8a559373fb2698aaa5e44 100644 (file)
--- a/src/tail.c
+++ b/src/tail.c
@@ -180,6 +180,9 @@ static bool from_start;
 /* If true, print filename headers.  */
 static bool print_headers;
 
+/* Character to split lines by. */
+static char line_end;
+
 /* When to print the filename banners.  */
 enum header_mode
 {
@@ -238,6 +241,7 @@ static struct option const long_options[] =
   {"silent", no_argument, NULL, 'q'},
   {"sleep-interval", required_argument, NULL, 's'},
   {"verbose", no_argument, NULL, 'v'},
+  {"zero-terminated", no_argument, NULL, 'z'},
   {GETOPT_HELP_OPTION_DECL},
   {GETOPT_VERSION_OPTION_DECL},
   {NULL, 0, NULL, 0}
@@ -296,6 +300,9 @@ With more than one FILE, precede each with a header giving the file name.\n\
                              with inotify and --pid=P, check process P at\n\
                              least once every N seconds\n\
   -v, --verbose            always output headers giving file names\n\
+"), stdout);
+     fputs (_("\
+  -z, --zero-terminated    line delimiter is NUL, not newline\n\
 "), stdout);
      fputs (HELP_OPTION_DESCRIPTION, stdout);
      fputs (VERSION_OPTION_DESCRIPTION, stdout);
@@ -499,7 +506,7 @@ file_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
   *read_pos = pos + bytes_read;
 
   /* Count the incomplete line on files that don't end with a newline.  */
-  if (bytes_read && buffer[bytes_read - 1] != '\n')
+  if (bytes_read && buffer[bytes_read - 1] != line_end)
     --n_lines;
 
   do
@@ -510,7 +517,7 @@ file_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
       while (n)
         {
           char const *nl;
-          nl = memrchr (buffer, '\n', n);
+          nl = memrchr (buffer, line_end, n);
           if (nl == NULL)
             break;
           n = nl - buffer;
@@ -595,7 +602,7 @@ pipe_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
       {
         char const *buffer_end = tmp->buffer + n_read;
         char const *p = tmp->buffer;
-        while ((p = memchr (p, '\n', buffer_end - p)))
+        while ((p = memchr (p, line_end, buffer_end - p)))
           {
             ++p;
             ++tmp->nlines;
@@ -649,7 +656,7 @@ pipe_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
     goto free_lbuffers;
 
   /* Count the incomplete line on files that don't end with a newline.  */
-  if (last->buffer[last->nbytes - 1] != '\n')
+  if (last->buffer[last->nbytes - 1] != line_end)
     {
       ++last->nlines;
       ++total_lines;
@@ -671,7 +678,7 @@ pipe_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
         size_t j;
         for (j = total_lines - n_lines; j; --j)
           {
-            beg = memchr (beg, '\n', buffer_end - beg);
+            beg = memchr (beg, line_end, buffer_end - beg);
             assert (beg);
             ++beg;
           }
@@ -857,7 +864,7 @@ start_lines (const char *pretty_filename, int fd, uintmax_t n_lines,
       *read_pos += bytes_read;
 
       char *p = buffer;
-      while ((p = memchr (p, '\n', buffer_end - p)))
+      while ((p = memchr (p, line_end, buffer_end - p)))
         {
           ++p;
           if (--n_lines == 0)
@@ -2047,7 +2054,7 @@ parse_options (int argc, char **argv,
 {
   int c;
 
-  while ((c = getopt_long (argc, argv, "c:n:fFqs:v0123456789",
+  while ((c = getopt_long (argc, argv, "c:n:fFqs:vz0123456789",
                            long_options, NULL))
          != -1)
     {
@@ -2124,6 +2131,10 @@ parse_options (int argc, char **argv,
           *header_mode = always;
           break;
 
+        case 'z':
+          line_end = '\0';
+          break;
+
         case_GETOPT_HELP_CHAR;
 
         case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);
@@ -2221,6 +2232,7 @@ main (int argc, char **argv)
 
   count_lines = true;
   forever = from_start = print_headers = false;
+  line_end = '\n';
   obsolete_option = parse_obsolete_option (argc, argv, &n_units);
   argc -= obsolete_option;
   argv += obsolete_option;

diff --git a/tests/misc/head.pl b/tests/misc/head.pl
index 78644f2a5dac65fd39649f04c6a7e98e5c5d9ecc..1f565cf16748beb2065ff942e31ff20afa88fbd2 100755 (executable)
--- a/tests/misc/head.pl
+++ b/tests/misc/head.pl
@@ -72,6 +72,10 @@ my @Tests =
   ['no-oct-2', '-010', {IN=>"\n"x12}, {OUT=>"\n"x10}],
   ['no-oct-3', '-n 08', {IN=>"\n"x12}, {OUT=>"\n"x8}],
   ['no-oct-4', '-c 08', {IN=>"\n"x12}, {OUT=>"\n"x8}],
+
+  # --zero-terminated
+  ['zero-1', '-z -n 1',  {IN=>"x\0y"}, {OUT=>"x\0"}],
+  ['zero-2', '-z -n 2',  {IN=>"x\0y"}, {OUT=>"x\0y"}],
 );
 
 @Tests = triple_test \@Tests;

diff --git a/tests/misc/tail.pl b/tests/misc/tail.pl
index c23102fe839258a16b955aa07488fdde2dac19fd..0d9bc48348acb66c3c1f95469669792701689e95 100755 (executable)
--- a/tests/misc/tail.pl
+++ b/tests/misc/tail.pl
@@ -101,6 +101,10 @@ my @tv = (
 
 # With textutils-1.22, this failed.
 ['f-pipe-1', '-f -n 1', "a\nb\n", "b\n", 0],
+
+# --zero-terminated
+['zero-1', '-z -n 1', "x\0y", "y", 0],
+['zero-2', '-z -n 2', "x\0y", "x\0y", 0],
 );
 
 my @Tests;