dd: synchronize output after write errors

[coreutils.git] / src / wc.c

/* wc - print the number of lines, words, and bytes in files
   Copyright (C) 1985-2022 Free Software Foundation, Inc.

   This program is free software: you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation, either version 3 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program.  If not, see <https://www.gnu.org/licenses/>.  */

/* Written by Paul Rubin, phr@ocf.berkeley.edu
   and David MacKenzie, djm@gnu.ai.mit.edu. */

#include <config.h>

#include <stdio.h>
#include <assert.h>
#include <getopt.h>
#include <sys/types.h>
#include <wchar.h>
#include <wctype.h>

#include "system.h"
#include "argv-iter.h"
#include "die.h"
#include "error.h"
#include "fadvise.h"
#include "mbchar.h"
#include "physmem.h"
#include "readtokens0.h"
#include "safe-read.h"
#include "stat-size.h"
#include "xbinary-io.h"
#ifdef USE_AVX2_WC_LINECOUNT
# include <cpuid.h>
#endif

#if !defined iswspace && !HAVE_ISWSPACE
# define iswspace(wc) \
    ((wc) == to_uchar (wc) && isspace (to_uchar (wc)))
#endif

/* The official name of this program (e.g., no 'g' prefix).  */
#define PROGRAM_NAME "wc"

#define AUTHORS \
  proper_name ("Paul Rubin"), \
  proper_name ("David MacKenzie")

/* Size of atomic reads. */
#define BUFFER_SIZE (16 * 1024)

static bool
wc_lines (char const *file, int fd, uintmax_t *lines_out,
          uintmax_t *bytes_out);
#ifdef USE_AVX2_WC_LINECOUNT
/* From wc_avx2.c */
extern bool
wc_lines_avx2 (char const *file, int fd, uintmax_t *lines_out,
               uintmax_t *bytes_out);
#endif
static bool
(*wc_lines_p) (char const *file, int fd, uintmax_t *lines_out,
                uintmax_t *bytes_out) = wc_lines;

static bool debug;

/* Cumulative number of lines, words, chars and bytes in all files so far.
   max_line_length is the maximum over all files processed so far.  */
static uintmax_t total_lines;
static uintmax_t total_words;
static uintmax_t total_chars;
static uintmax_t total_bytes;
static uintmax_t max_line_length;

/* Which counts to print. */
static bool print_lines, print_words, print_chars, print_bytes;
static bool print_linelength;

/* The print width of each count.  */
static int number_width;

/* True if we have ever read the standard input. */
static bool have_read_stdin;

/* Used to determine if file size can be determined without reading.  */
static size_t page_size;

/* Enable to _not_ treat non breaking space as a word separator.  */
static bool posixly_correct;

/* The result of calling fstat or stat on a file descriptor or file.  */
struct fstatus
{
  /* If positive, fstat or stat has not been called yet.  Otherwise,
     this is the value returned from fstat or stat.  */
  int failed;

  /* If FAILED is zero, this is the file's status.  */
  struct stat st;
};

/* For long options that have no equivalent short option, use a
   non-character as a pseudo short option, starting with CHAR_MAX + 1.  */
enum
{
  DEBUG_PROGRAM_OPTION = CHAR_MAX + 1,
  FILES0_FROM_OPTION,
};

static struct option const longopts[] =
{
  {"bytes", no_argument, NULL, 'c'},
  {"chars", no_argument, NULL, 'm'},
  {"lines", no_argument, NULL, 'l'},
  {"words", no_argument, NULL, 'w'},
  {"debug", no_argument, NULL, DEBUG_PROGRAM_OPTION},
  {"files0-from", required_argument, NULL, FILES0_FROM_OPTION},
  {"max-line-length", no_argument, NULL, 'L'},
  {GETOPT_HELP_OPTION_DECL},
  {GETOPT_VERSION_OPTION_DECL},
  {NULL, 0, NULL, 0}
};

#ifdef USE_AVX2_WC_LINECOUNT
static bool
avx2_supported (void)
{
  unsigned int eax = 0;
  unsigned int ebx = 0;
  unsigned int ecx = 0;
  unsigned int edx = 0;
  bool getcpuid_ok = false;
  bool avx_enabled = false;

  if (__get_cpuid (1, &eax, &ebx, &ecx, &edx))
    {
      getcpuid_ok = true;
      if (ecx & bit_OSXSAVE)
        avx_enabled = true;  /* Support is not disabled.  */
    }


  if (avx_enabled)
    {
      eax = ebx = ecx = edx = 0;
      if (! __get_cpuid_count (7, 0, &eax, &ebx, &ecx, &edx))
        getcpuid_ok = false;
      else
        {
          if (! (ebx & bit_AVX2))
            avx_enabled = false;  /* Hardware doesn't support it.  */
        }
    }


  if (! getcpuid_ok)
    {
      if (debug)
        error (0, 0, "%s", _("failed to get cpuid"));
      return false;
    }
  else if (! avx_enabled)
    {
      if (debug)
        error (0, 0, "%s", _("avx2 support not detected"));
      return false;
    }
  else
    {
      if (debug)
        error (0, 0, "%s", _("using avx2 hardware support"));
      return true;
    }
}
#endif

void
usage (int status)
{
  if (status != EXIT_SUCCESS)
    emit_try_help ();
  else
    {
      printf (_("\
Usage: %s [OPTION]... [FILE]...\n\
  or:  %s [OPTION]... --files0-from=F\n\
"),
              program_name, program_name);
      fputs (_("\
Print newline, word, and byte counts for each FILE, and a total line if\n\
more than one FILE is specified.  A word is a non-zero-length sequence of\n\
printable characters delimited by white space.\n\
"), stdout);

      emit_stdin_note ();

      fputs (_("\
\n\
The options below may be used to select which counts are printed, always in\n\
the following order: newline, word, character, byte, maximum line length.\n\
  -c, --bytes            print the byte counts\n\
  -m, --chars            print the character counts\n\
  -l, --lines            print the newline counts\n\
"), stdout);
      fputs (_("\
      --files0-from=F    read input from the files specified by\n\
                           NUL-terminated names in file F;\n\
                           If F is - then read names from standard input\n\
  -L, --max-line-length  print the maximum display width\n\
  -w, --words            print the word counts\n\
"), stdout);
      fputs (HELP_OPTION_DESCRIPTION, stdout);
      fputs (VERSION_OPTION_DESCRIPTION, stdout);
      emit_ancillary_info (PROGRAM_NAME);
    }
  exit (status);
}

/* Return non zero if a non breaking space.  */
ATTRIBUTE_PURE
static int
iswnbspace (wint_t wc)
{
  return ! posixly_correct
         && (wc == 0x00A0 || wc == 0x2007
             || wc == 0x202F || wc == 0x2060);
}

static int
isnbspace (int c)
{
  return iswnbspace (btowc (c));
}

/* FILE is the name of the file (or NULL for standard input)
   associated with the specified counters.  */
static void
write_counts (uintmax_t lines,
              uintmax_t words,
              uintmax_t chars,
              uintmax_t bytes,
              uintmax_t linelength,
              char const *file)
{
  static char const format_sp_int[] = " %*s";
  char const *format_int = format_sp_int + 1;
  char buf[INT_BUFSIZE_BOUND (uintmax_t)];

  if (print_lines)
    {
      printf (format_int, number_width, umaxtostr (lines, buf));
      format_int = format_sp_int;
    }
  if (print_words)
    {
      printf (format_int, number_width, umaxtostr (words, buf));
      format_int = format_sp_int;
    }
  if (print_chars)
    {
      printf (format_int, number_width, umaxtostr (chars, buf));
      format_int = format_sp_int;
    }
  if (print_bytes)
    {
      printf (format_int, number_width, umaxtostr (bytes, buf));
      format_int = format_sp_int;
    }
  if (print_linelength)
    {
      printf (format_int, number_width, umaxtostr (linelength, buf));
    }
  if (file)
    printf (" %s", strchr (file, '\n') ? quotef (file) : file);
  putchar ('\n');
}

static bool
wc_lines (char const *file, int fd, uintmax_t *lines_out, uintmax_t *bytes_out)
{
  size_t bytes_read;
  uintmax_t lines, bytes;
  char buf[BUFFER_SIZE + 1];
  bool long_lines = false;

  if (!lines_out || !bytes_out)
    {
      return false;
    }

  lines = bytes = 0;

  while ((bytes_read = safe_read (fd, buf, BUFFER_SIZE)) > 0)
    {

      if (bytes_read == SAFE_READ_ERROR)
        {
          error (0, errno, "%s", quotef (file));
          return false;
        }

      bytes += bytes_read;

      char *p = buf;
      char *end = buf + bytes_read;
      uintmax_t plines = lines;

      if (! long_lines)
        {
          /* Avoid function call overhead for shorter lines.  */
          while (p != end)
            lines += *p++ == '\n';
        }
      else
        {
          /* rawmemchr is more efficient with longer lines.  */
          *end = '\n';
          while ((p = rawmemchr (p, '\n')) < end)
            {
              ++p;
              ++lines;
            }
        }

      /* If the average line length in the block is >= 15, then use
          memchr for the next block, where system specific optimizations
          may outweigh function call overhead.
          FIXME: This line length was determined in 2015, on both
          x86_64 and ppc64, but it's worth re-evaluating in future with
          newer compilers, CPUs, or memchr() implementations etc.  */
      if (lines - plines <= bytes_read / 15)
        long_lines = true;
      else
        long_lines = false;
    }

  *bytes_out = bytes;
  *lines_out = lines;

  return true;
}

/* Count words.  FILE_X is the name of the file (or NULL for standard
   input) that is open on descriptor FD.  *FSTATUS is its status.
   CURRENT_POS is the current file offset if known, negative if unknown.
   Return true if successful.  */
static bool
wc (int fd, char const *file_x, struct fstatus *fstatus, off_t current_pos)
{
  bool ok = true;
  char buf[BUFFER_SIZE + 1];
  size_t bytes_read;
  uintmax_t lines, words, chars, bytes, linelength;
  bool count_bytes, count_chars, count_complicated;
  char const *file = file_x ? file_x : _("standard input");

  lines = words = chars = bytes = linelength = 0;

  /* If in the current locale, chars are equivalent to bytes, we prefer
     counting bytes, because that's easier.  */
#if MB_LEN_MAX > 1
  if (MB_CUR_MAX > 1)
    {
      count_bytes = print_bytes;
      count_chars = print_chars;
    }
  else
#endif
    {
      count_bytes = print_bytes || print_chars;
      count_chars = false;
    }
  count_complicated = print_words || print_linelength;

  /* Advise the kernel of our access pattern only if we will read().  */
  if (!count_bytes || count_chars || print_lines || count_complicated)
    fdadvise (fd, 0, 0, FADVISE_SEQUENTIAL);

  /* When counting only bytes, save some line- and word-counting
     overhead.  If FD is a 'regular' Unix file, using lseek is enough
     to get its 'size' in bytes.  Otherwise, read blocks of BUFFER_SIZE
     bytes at a time until EOF.  Note that the 'size' (number of bytes)
     that wc reports is smaller than stats.st_size when the file is not
     positioned at its beginning.  That's why the lseek calls below are
     necessary.  For example the command
     '(dd ibs=99k skip=1 count=0; ./wc -c) < /etc/group'
     should make wc report '0' bytes.  */

  if (count_bytes && !count_chars && !print_lines && !count_complicated)
    {
      bool skip_read = false;

      if (0 < fstatus->failed)
        fstatus->failed = fstat (fd, &fstatus->st);

      /* For sized files, seek to one st_blksize before EOF rather than to EOF.
         This works better for files in proc-like file systems where
         the size is only approximate.  */
      if (! fstatus->failed && usable_st_size (&fstatus->st)
          && 0 <= fstatus->st.st_size)
        {
          size_t end_pos = fstatus->st.st_size;
          if (current_pos < 0)
            current_pos = lseek (fd, 0, SEEK_CUR);

          if (end_pos % page_size)
            {
              /* We only need special handling of /proc and /sys files etc.
                 when they're a multiple of PAGE_SIZE.  In the common case
                 for files with st_size not a multiple of PAGE_SIZE,
                 it's more efficient and accurate to use st_size.

                 Be careful here.  The current position may actually be
                 beyond the end of the file.  As in the example above.  */

              bytes = end_pos < current_pos ? 0 : end_pos - current_pos;
              skip_read = true;
            }
          else
            {
              off_t hi_pos = end_pos - end_pos % (ST_BLKSIZE (fstatus->st) + 1);
              if (0 <= current_pos && current_pos < hi_pos
                  && 0 <= lseek (fd, hi_pos, SEEK_CUR))
                bytes = hi_pos - current_pos;
            }
        }

      if (! skip_read)
        {
          fdadvise (fd, 0, 0, FADVISE_SEQUENTIAL);
          while ((bytes_read = safe_read (fd, buf, BUFFER_SIZE)) > 0)
            {
              if (bytes_read == SAFE_READ_ERROR)
                {
                  error (0, errno, "%s", quotef (file));
                  ok = false;
                  break;
                }
              bytes += bytes_read;
            }
        }
    }
  else if (!count_chars && !count_complicated)
    {
#ifdef USE_AVX2_WC_LINECOUNT
      if (avx2_supported ())
        wc_lines_p = wc_lines_avx2;
#endif

      /* Use a separate loop when counting only lines or lines and bytes --
         but not chars or words.  */
      ok = wc_lines_p (file, fd, &lines, &bytes);
    }
#if MB_LEN_MAX > 1
# define SUPPORT_OLD_MBRTOWC 1
  else if (MB_CUR_MAX > 1)
    {
      bool in_word = false;
      uintmax_t linepos = 0;
      mbstate_t state = { 0, };
      bool in_shift = false;
# if SUPPORT_OLD_MBRTOWC
      /* Back-up the state before each multibyte character conversion and
         move the last incomplete character of the buffer to the front
         of the buffer.  This is needed because we don't know whether
         the 'mbrtowc' function updates the state when it returns -2, --
         this is the ISO C 99 and glibc-2.2 behaviour - or not - amended
         ANSI C, glibc-2.1 and Solaris 5.7 behaviour.  We don't have an
         autoconf test for this, yet.  */
      size_t prev = 0; /* number of bytes carried over from previous round */
# else
      const size_t prev = 0;
# endif

      while ((bytes_read = safe_read (fd, buf + prev, BUFFER_SIZE - prev)) > 0)
        {
          char const *p;
# if SUPPORT_OLD_MBRTOWC
          mbstate_t backup_state;
# endif
          if (bytes_read == SAFE_READ_ERROR)
            {
              error (0, errno, "%s", quotef (file));
              ok = false;
              break;
            }

          bytes += bytes_read;
          p = buf;
          bytes_read += prev;
          do
            {
              wchar_t wide_char;
              size_t n;
              bool wide = true;

              if (!in_shift && is_basic (*p))
                {
                  /* Handle most ASCII characters quickly, without calling
                     mbrtowc().  */
                  n = 1;
                  wide_char = *p;
                  wide = false;
                }
              else
                {
                  in_shift = true;
# if SUPPORT_OLD_MBRTOWC
                  backup_state = state;
# endif
                  n = mbrtowc (&wide_char, p, bytes_read, &state);
                  if (n == (size_t) -2)
                    {
# if SUPPORT_OLD_MBRTOWC
                      state = backup_state;
# endif
                      break;
                    }
                  if (n == (size_t) -1)
                    {
                      /* Remember that we read a byte, but don't complain
                         about the error.  Because of the decoding error,
                         this is a considered to be byte but not a
                         character (that is, chars is not incremented).  */
                      p++;
                      bytes_read--;
                      continue;
                    }
                  if (mbsinit (&state))
                    in_shift = false;
                  if (n == 0)
                    {
                      wide_char = 0;
                      n = 1;
                    }
                }

              switch (wide_char)
                {
                case '\n':
                  lines++;
                  FALLTHROUGH;
                case '\r':
                case '\f':
                  if (linepos > linelength)
                    linelength = linepos;
                  linepos = 0;
                  goto mb_word_separator;
                case '\t':
                  linepos += 8 - (linepos % 8);
                  goto mb_word_separator;
                case ' ':
                  linepos++;
                  FALLTHROUGH;
                case '\v':
                mb_word_separator:
                  words += in_word;
                  in_word = false;
                  break;
                default:
                  if (wide && iswprint (wide_char))
                    {
                      /* wcwidth can be expensive on OSX for example,
                         so avoid if uneeded.  */
                      if (print_linelength)
                        {
                          int width = wcwidth (wide_char);
                          if (width > 0)
                            linepos += width;
                        }
                      if (iswspace (wide_char) || iswnbspace (wide_char))
                        goto mb_word_separator;
                      in_word = true;
                    }
                  else if (!wide && isprint (to_uchar (*p)))
                    {
                      linepos++;
                      if (isspace (to_uchar (*p)))
                        goto mb_word_separator;
                      in_word = true;
                    }
                  break;
                }

              p += n;
              bytes_read -= n;
              chars++;
            }
          while (bytes_read > 0);

# if SUPPORT_OLD_MBRTOWC
          if (bytes_read > 0)
            {
              if (bytes_read == BUFFER_SIZE)
                {
                  /* Encountered a very long redundant shift sequence.  */
                  p++;
                  bytes_read--;
                }
              memmove (buf, p, bytes_read);
            }
          prev = bytes_read;
# endif
        }
      if (linepos > linelength)
        linelength = linepos;
      words += in_word;
    }
#endif
  else
    {
      bool in_word = false;
      uintmax_t linepos = 0;

      while ((bytes_read = safe_read (fd, buf, BUFFER_SIZE)) > 0)
        {
          char const *p = buf;
          if (bytes_read == SAFE_READ_ERROR)
            {
              error (0, errno, "%s", quotef (file));
              ok = false;
              break;
            }

          bytes += bytes_read;
          do
            {
              switch (*p++)
                {
                case '\n':
                  lines++;
                  FALLTHROUGH;
                case '\r':
                case '\f':
                  if (linepos > linelength)
                    linelength = linepos;
                  linepos = 0;
                  goto word_separator;
                case '\t':
                  linepos += 8 - (linepos % 8);
                  goto word_separator;
                case ' ':
                  linepos++;
                  FALLTHROUGH;
                case '\v':
                word_separator:
                  words += in_word;
                  in_word = false;
                  break;
                default:
                  if (isprint (to_uchar (p[-1])))
                    {
                      linepos++;
                      if (isspace (to_uchar (p[-1]))
                          || isnbspace (to_uchar (p[-1])))
                        goto word_separator;
                      in_word = true;
                    }
                  break;
                }
            }
          while (--bytes_read);
        }
      if (linepos > linelength)
        linelength = linepos;
      words += in_word;
    }

  if (count_chars < print_chars)
    chars = bytes;

  write_counts (lines, words, chars, bytes, linelength, file_x);
  total_lines += lines;
  total_words += words;
  total_chars += chars;
  total_bytes += bytes;
  if (linelength > max_line_length)
    max_line_length = linelength;

  return ok;
}

static bool
wc_file (char const *file, struct fstatus *fstatus)
{
  if (! file || STREQ (file, "-"))
    {
      have_read_stdin = true;
      xset_binary_mode (STDIN_FILENO, O_BINARY);
      return wc (STDIN_FILENO, file, fstatus, -1);
    }
  else
    {
      int fd = open (file, O_RDONLY | O_BINARY);
      if (fd == -1)
        {
          error (0, errno, "%s", quotef (file));
          return false;
        }
      else
        {
          bool ok = wc (fd, file, fstatus, 0);
          if (close (fd) != 0)
            {
              error (0, errno, "%s", quotef (file));
              return false;
            }
          return ok;
        }
    }
}

/* Return the file status for the NFILES files addressed by FILE.
   Optimize the case where only one number is printed, for just one
   file; in that case we can use a print width of 1, so we don't need
   to stat the file.  Handle the case of (nfiles == 0) in the same way;
   that happens when we don't know how long the list of file names will be.  */

static struct fstatus *
get_input_fstatus (size_t nfiles, char *const *file)
{
  struct fstatus *fstatus = xnmalloc (nfiles ? nfiles : 1, sizeof *fstatus);

  if (nfiles == 0
      || (nfiles == 1
          && ((print_lines + print_words + print_chars
               + print_bytes + print_linelength)
              == 1)))
    fstatus[0].failed = 1;
  else
    {
      for (size_t i = 0; i < nfiles; i++)
        fstatus[i].failed = (! file[i] || STREQ (file[i], "-")
                             ? fstat (STDIN_FILENO, &fstatus[i].st)
                             : stat (file[i], &fstatus[i].st));
    }

  return fstatus;
}

/* Return a print width suitable for the NFILES files whose status is
   recorded in FSTATUS.  Optimize the same special case that
   get_input_fstatus optimizes.  */

ATTRIBUTE_PURE
static int
compute_number_width (size_t nfiles, struct fstatus const *fstatus)
{
  int width = 1;

  if (0 < nfiles && fstatus[0].failed <= 0)
    {
      int minimum_width = 1;
      uintmax_t regular_total = 0;

      for (size_t i = 0; i < nfiles; i++)
        if (! fstatus[i].failed)
          {
            if (S_ISREG (fstatus[i].st.st_mode))
              regular_total += fstatus[i].st.st_size;
            else
              minimum_width = 7;
          }

      for (; 10 <= regular_total; regular_total /= 10)
        width++;
      if (width < minimum_width)
        width = minimum_width;
    }

  return width;
}


int
main (int argc, char **argv)
{
  bool ok;
  int optc;
  size_t nfiles;
  char **files;
  char *files_from = NULL;
  struct fstatus *fstatus;
  struct Tokens tok;

  initialize_main (&argc, &argv);
  set_program_name (argv[0]);
  setlocale (LC_ALL, "");
  bindtextdomain (PACKAGE, LOCALEDIR);
  textdomain (PACKAGE);

  atexit (close_stdout);

  page_size = getpagesize ();
  /* Line buffer stdout to ensure lines are written atomically and immediately
     so that processes running in parallel do not intersperse their output.  */
  setvbuf (stdout, NULL, _IOLBF, 0);

  posixly_correct = (getenv ("POSIXLY_CORRECT") != NULL);

  print_lines = print_words = print_chars = print_bytes = false;
  print_linelength = false;
  total_lines = total_words = total_chars = total_bytes = max_line_length = 0;

  while ((optc = getopt_long (argc, argv, "clLmw", longopts, NULL)) != -1)
    switch (optc)
      {
      case 'c':
        print_bytes = true;
        break;

      case 'm':
        print_chars = true;
        break;

      case 'l':
        print_lines = true;
        break;

      case 'w':
        print_words = true;
        break;

      case 'L':
        print_linelength = true;
        break;

      case DEBUG_PROGRAM_OPTION:
        debug = true;
        break;

      case FILES0_FROM_OPTION:
        files_from = optarg;
        break;

      case_GETOPT_HELP_CHAR;

      case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);

      default:
        usage (EXIT_FAILURE);
      }

  if (! (print_lines || print_words || print_chars || print_bytes
         || print_linelength))
    print_lines = print_words = print_bytes = true;

  bool read_tokens = false;
  struct argv_iterator *ai;
  if (files_from)
    {
      FILE *stream;

      /* When using --files0-from=F, you may not specify any files
         on the command-line.  */
      if (optind < argc)
        {
          error (0, 0, _("extra operand %s"), quoteaf (argv[optind]));
          fprintf (stderr, "%s\n",
                   _("file operands cannot be combined with --files0-from"));
          usage (EXIT_FAILURE);
        }

      if (STREQ (files_from, "-"))
        stream = stdin;
      else
        {
          stream = fopen (files_from, "r");
          if (stream == NULL)
            die (EXIT_FAILURE, errno, _("cannot open %s for reading"),
                 quoteaf (files_from));
        }

      /* Read the file list into RAM if we can detect its size and that
         size is reasonable.  Otherwise, we'll read a name at a time.  */
      struct stat st;
      if (fstat (fileno (stream), &st) == 0
          && S_ISREG (st.st_mode)
          && st.st_size <= MIN (10 * 1024 * 1024, physmem_available () / 2))
        {
          read_tokens = true;
          readtokens0_init (&tok);
          if (! readtokens0 (stream, &tok) || fclose (stream) != 0)
            die (EXIT_FAILURE, 0, _("cannot read file names from %s"),
                 quoteaf (files_from));
          files = tok.tok;
          nfiles = tok.n_tok;
          ai = argv_iter_init_argv (files);
        }
      else
        {
          files = NULL;
          nfiles = 0;
          ai = argv_iter_init_stream (stream);
        }
    }
  else
    {
      static char *stdin_only[] = { NULL };
      files = (optind < argc ? argv + optind : stdin_only);
      nfiles = (optind < argc ? argc - optind : 1);
      ai = argv_iter_init_argv (files);
    }

  if (!ai)
    xalloc_die ();

  fstatus = get_input_fstatus (nfiles, files);
  number_width = compute_number_width (nfiles, fstatus);

  ok = true;
  for (int i = 0; /* */; i++)
    {
      bool skip_file = false;
      enum argv_iter_err ai_err;
      char *file_name = argv_iter (ai, &ai_err);
      if (!file_name)
        {
          switch (ai_err)
            {
            case AI_ERR_EOF:
              goto argv_iter_done;
            case AI_ERR_READ:
              error (0, errno, _("%s: read error"),
                     quotef (files_from));
              ok = false;
              goto argv_iter_done;
            case AI_ERR_MEM:
              xalloc_die ();
            default:
              assert (!"unexpected error code from argv_iter");
            }
        }
      if (files_from && STREQ (files_from, "-") && STREQ (file_name, "-"))
        {
          /* Give a better diagnostic in an unusual case:
             printf - | wc --files0-from=- */
          error (0, 0, _("when reading file names from stdin, "
                         "no file name of %s allowed"),
                 quoteaf (file_name));
          skip_file = true;
        }

      if (!file_name[0])
        {
          /* Diagnose a zero-length file name.  When it's one
             among many, knowing the record number may help.
             FIXME: currently print the record number only with
             --files0-from=FILE.  Maybe do it for argv, too?  */
          if (files_from == NULL)
            error (0, 0, "%s", _("invalid zero-length file name"));
          else
            {
              /* Using the standard 'filename:line-number:' prefix here is
                 not totally appropriate, since NUL is the separator, not NL,
                 but it might be better than nothing.  */
              unsigned long int file_number = argv_iter_n_args (ai);
              error (0, 0, "%s:%lu: %s", quotef (files_from),
                     file_number, _("invalid zero-length file name"));
            }
          skip_file = true;
        }

      if (skip_file)
        ok = false;
      else
        ok &= wc_file (file_name, &fstatus[nfiles ? i : 0]);

      if (! nfiles)
        fstatus[0].failed = 1;
    }
 argv_iter_done:

  /* No arguments on the command line is fine.  That means read from stdin.
     However, no arguments on the --files0-from input stream is an error
     means don't read anything.  */
  if (ok && !files_from && argv_iter_n_args (ai) == 0)
    ok &= wc_file (NULL, &fstatus[0]);

  if (read_tokens)
    readtokens0_free (&tok);

  if (1 < argv_iter_n_args (ai))
    write_counts (total_lines, total_words, total_chars, total_bytes,
                  max_line_length, _("total"));

  argv_iter_free (ai);

  free (fstatus);

  if (have_read_stdin && close (STDIN_FILENO) != 0)
    die (EXIT_FAILURE, errno, "-");

  return ok ? EXIT_SUCCESS : EXIT_FAILURE;
}