/* cut - remove parts of lines of files
   Copyright (C) 1984, 1997, 1998, 1999, 2000, 2001, 2002 by David M. Ihnat

   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 2, 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, write to the Free Software Foundation,
   Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */

/* Written by David Ihnat.  */

/* POSIX changes, bug fixes, long-named options, and cleanup
   by David MacKenzie <djm@gnu.ai.mit.edu>.

   Rewrite cut_fields and cut_bytes -- Jim Meyering.  */

#include <config.h>

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

/* Get mbstate_t, mbrtowc().  */
#if HAVE_WCHAR_H
# include <wchar.h>
#endif

#include "system.h"
#include "getstr.h"
#include "closeout.h"
#include "error.h"

/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC
   installation; work around this configuration error.	*/
#if !defined MB_LEN_MAX || MB_LEN_MAX < 2
# undef MB_LEN_MAX
# define MB_LEN_MAX 16
#endif

/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t.  */
#if HAVE_MBRTOWC && defined mbstate_t
# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0)
#endif

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

#define AUTHORS N_ ("David Ihnat, David MacKenzie, and Jim Meyering")

#define FATAL_ERROR(Message)						\
  do									\
    {									\
      error (0, 0, (Message));						\
      usage (2);							\
    }									\
  while (0)

/* Append LOW, HIGH to the list RP of range pairs, allocating additional
   space if necessary.  Update local variable N_RP.  When allocating,
   update global variable N_RP_ALLOCATED.  */

#define ADD_RANGE_PAIR(rp, low, high)					\
  do									\
    {									\
      if (n_rp >= n_rp_allocated)					\
	{								\
	  n_rp_allocated *= 2;						\
	  (rp) = (struct range_pair *) xrealloc ((char *) (rp),		\
				   n_rp_allocated * sizeof (*(rp)));	\
	}								\
      rp[n_rp].lo = (low);						\
      rp[n_rp].hi = (high);						\
      ++n_rp;								\
    }									\
  while (0)

/* Refill the buffer BUF to get a multibyte character. */
#define REFILL_BUFFER(BUF, BUFPOS, BUFLEN, STREAM)			\
  do									\
    {									\
      if (BUFLEN < MB_LEN_MAX && !feof (STREAM) && !ferror (STREAM))	\
	{								\
	  memmove (BUF, BUFPOS, BUFLEN);				\
	  BUFLEN += fread (BUF + BUFLEN, sizeof(char), BUFSIZ, STREAM); \
	  BUFPOS = BUF;							\
	}								\
    }									\
  while (0)

/* Get wide character on BUFPOS. BUFPOS is not included after that.
   If byte sequence is not valid as a character, CONVFAIL is 1. Otherwise 0. */ 
#define GET_NEXT_WC_FROM_BUFFER(WC, BUFPOS, BUFLEN, MBLENGTH, STATE, CONVFAIL) \
  do									\
    {									\
      mbstate_t state_bak;						\
									\
      if (BUFLEN < 1)							\
	{								\
	  WC = WEOF;							\
	  break;							\
	}								\
									\
      /* Get a wide character. */					\
      CONVFAIL = 0;							\
      state_bak = STATE;						\
      MBLENGTH = mbrtowc ((wchar_t *)&WC, BUFPOS, BUFLEN, &STATE);	\
									\
      switch (MBLENGTH)							\
	{								\
	case (size_t)-1:						\
	case (size_t)-2:						\
	  CONVFAIL++;							\
	  STATE = state_bak;						\
	  /* Fall througn. */						\
									\
	case 0:								\
	  MBLENGTH = 1;							\
	  break;							\
	}								\
    }									\
  while (0)

struct range_pair
  {
    unsigned int lo;
    unsigned int hi;
  };

/* This buffer is used to support the semantics of the -s option
   (or lack of same) when the specified field list includes (does
   not include) the first field.  In both of those cases, the entire
   first field must be read into this buffer to determine whether it
   is followed by a delimiter or a newline before any of it may be
   output.  Otherwise, cut_fields can do the job without using this
   buffer.  */
static char *field_1_buffer;

/* The number of bytes allocated for FIELD_1_BUFFER.  */
static size_t field_1_bufsize;

/* The largest byte, character or field index used as an endpoint of a closed
   or degenerate range specification;  this doesn't include the starting
   index of right-open-ended ranges.  For example, with either range spec
   `2-5,9-', `2-3,5,9-' this variable would be set to 5.  */
static unsigned int max_range_endpoint;

/* If nonzero, this is the index of the first field in a range that goes
   to end of line. */
static unsigned int eol_range_start;

/* In byte mode, which bytes to output.
   In character mode, which characters to output.
   In field mode, which DELIM-separated fields to output.
   Bytes, characters and fields are numbered starting with 1,
   so the zeroth element of this array is unused.
   A byte, character or field K has been selected if
   (K <= MAX_RANGE_ENDPOINT and PRINTABLE_FIELD[K])
    || (EOL_RANGE_START > 0 && K >= EOL_RANGE_START).  */
static int *printable_field;

enum operating_mode
  {
    undefined_mode,

    /* Output bytes that are at the given positions. */
    byte_mode,

    /* Output characters that are at the given positions. */
    character_mode,

    /* Output the given delimeter-separated fields. */
    field_mode
  };

/* The name this program was run with. */
char *program_name;

static enum operating_mode operating_mode;

/* If nonzero, when in byte mode, don't split multibyte characters.  */
static int byte_mode_character_aware;

/* If nonzero, the function for single byte locale is work
   if this program runs on multibyte locale. */
static int force_singlebyte_mode;

/* If nonzero do not output lines containing no delimeter characters.
   Otherwise, all such lines are printed.  This option is valid only
   with field mode.  */
static int suppress_non_delimited;

/* The delimeter character for field mode. */
static int delim;
#if HAVE_WCHAR_H
static wchar_t wcdelim;
#endif

/* The length of output_delimiter_string.  */
static size_t output_delimiter_length;

/* The output field separator string.  Defaults to the 1-character
   string consisting of the input delimiter.  */
static char *output_delimiter_string;

/* Nonzero if we have ever read standard input. */
static int have_read_stdin;

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

static struct option const longopts[] =
{
  {"bytes", required_argument, 0, 'b'},
  {"characters", required_argument, 0, 'c'},
  {"fields", required_argument, 0, 'f'},
  {"delimiter", required_argument, 0, 'd'},
  {"only-delimited", no_argument, 0, 's'},
  {"output-delimiter", required_argument, 0, OUTPUT_DELIMITER_OPTION},
  {GETOPT_HELP_OPTION_DECL},
  {GETOPT_VERSION_OPTION_DECL},
  {0, 0, 0, 0}
};

void
usage (int status)
{
  if (status != 0)
    fprintf (stderr, _("Try `%s --help' for more information.\n"),
	     program_name);
  else
    {
      printf (_("\
Usage: %s [OPTION]... [FILE]...\n\
"),
	      program_name);
      fputs (_("\
Print selected parts of lines from each FILE to standard output.\n\
\n\
"), stdout);
      fputs (_("\
Mandatory arguments to long options are mandatory for short options too.\n\
"), stdout);
      fputs (_("\
  -b, --bytes=LIST        output only these bytes\n\
  -c, --characters=LIST   output only these characters\n\
  -d, --delimiter=DELIM   use DELIM instead of TAB for field delimiter\n\
"), stdout);
      fputs (_("\
  -f, --fields=LIST       output only these fields;  also print any line\n\
                            that contains no delimiter character, unless\n\
                            the -s option is specified\n\
  -n                      with -b: don't split multibyte characters\n\
"), stdout);
      fputs (_("\
  -s, --only-delimited    do not print lines not containing delimiters\n\
      --output-delimiter=STRING  use STRING as the output delimiter\n\
                            the default is to use the input delimiter\n\
"), stdout);
      fputs (HELP_OPTION_DESCRIPTION, stdout);
      fputs (VERSION_OPTION_DESCRIPTION, stdout);
      fputs (_("\
\n\
Use one, and only one of -b, -c or -f.  Each LIST is made up of one\n\
range, or many ranges separated by commas.  Each range is one of:\n\
\n\
  N     N'th byte, character or field, counted from 1\n\
  N-    from N'th byte, character or field, to end of line\n\
  N-M   from N'th to M'th (included) byte, character or field\n\
  -M    from first to M'th (included) byte, character or field\n\
\n\
With no FILE, or when FILE is -, read standard input.\n\
"), stdout);
      printf (_("\nReport bugs to <%s>.\n"), PACKAGE_BUGREPORT);
    }
  exit (status == 0 ? EXIT_SUCCESS : EXIT_FAILURE);
}

static int
print_kth (unsigned int k)
{
  return ((0 < eol_range_start && eol_range_start <= k)
	  || (k <= max_range_endpoint && printable_field[k]));
}

/* Given the list of field or byte range specifications FIELDSTR, set
   MAX_RANGE_ENDPOINT and allocate and initialize the PRINTABLE_FIELD
   array.  If there is a right-open-ended range, set EOL_RANGE_START
   to its starting index.  FIELDSTR should be composed of one or more
   numbers or ranges of numbers, separated by blanks or commas.
   Incomplete ranges may be given: `-m' means `1-m'; `n-' means `n'
   through end of line.  Return nonzero if FIELDSTR contains at least
   one field specification, zero otherwise.  */

/* FIXME-someday:  What if the user wants to cut out the 1,000,000-th field
   of some huge input file?  This function shouldn't have to alloate a table
   of a million ints just so we can test every field < 10^6 with an array
   dereference.  Instead, consider using a dynamic hash table.  It would be
   simpler and nearly as good a solution to use a 32K x 4-byte table with
   one bit per field index instead of a whole `int' per index.  */

static int
set_fields (const char *fieldstr)
{
  unsigned int initial = 1;	/* Value of first number in a range.  */
  unsigned int value = 0;	/* If nonzero, a number being accumulated.  */
  int dash_found = 0;		/* Nonzero if a '-' is found in this field.  */
  int field_found = 0;		/* Non-zero if at least one field spec
				   has been processed.  */

  struct range_pair *rp;
  unsigned int n_rp;
  unsigned int n_rp_allocated;
  unsigned int i;

  n_rp = 0;
  n_rp_allocated = 16;
  rp = (struct range_pair *) xmalloc (n_rp_allocated * sizeof (*rp));

  /* Collect and store in RP the range end points.
     It also sets EOL_RANGE_START if appropriate.  */

  for (;;)
    {
      if (*fieldstr == '-')
	{
	  /* Starting a range. */
	  if (dash_found)
	    FATAL_ERROR (_("invalid byte, character or field list"));
	  dash_found++;
	  fieldstr++;

	  if (value)
	    {
	      initial = value;
	      value = 0;
	    }
	  else
	    initial = 1;
	}
      else if (*fieldstr == ',' || ISBLANK (*fieldstr) || *fieldstr == '\0')
	{
	  /* Ending the string, or this field/byte sublist. */
	  if (dash_found)
	    {
	      dash_found = 0;

	      /* A range.  Possibilites: -n, m-n, n-.
		 In any case, `initial' contains the start of the range. */
	      if (value == 0)
		{
		  /* `n-'.  From `initial' to end of line. */
		  if(eol_range_start == 0 ||
		     (eol_range_start != 0 && eol_range_start > initial))
		    eol_range_start = initial;
		  field_found = 1;
		}
	      else
		{
		  /* `m-n' or `-n' (1-n). */
		  if (value < initial)
		    FATAL_ERROR (_("invalid byte, character or field list"));

		  /* Is there already a range going to end of line? */
		  if (eol_range_start != 0)
		    {
		      /* Yes.  Is the new sequence already contained
			 in the old one?  If so, no processing is
			 necessary. */
		      if (initial < eol_range_start)
			{
			  /* No, the new sequence starts before the
			     old.  Does the old range going to end of line
			     extend into the new range?  */
			  if (value + 1 >= eol_range_start)
			    {
			      /* Yes.  Simply move the end of line marker. */
			      eol_range_start = initial;
			    }
			  else
			    {
			      /* No.  A simple range, before and disjoint from
				 the range going to end of line.  Fill it. */
			      ADD_RANGE_PAIR (rp, initial, value);
			    }

			  /* In any case, some fields were selected. */
			  field_found = 1;
			}
		    }
		  else
		    {
		      /* There is no range going to end of line. */
		      ADD_RANGE_PAIR (rp, initial, value);
		      field_found = 1;
		    }
		  value = 0;
		}
	    }
	  else if (value != 0)
	    {
	      /* A simple field number, not a range. */
	      ADD_RANGE_PAIR (rp, value, value);
	      value = 0;
	      field_found = 1;
	    }

	  if (*fieldstr == '\0')
	    {
	      break;
	    }

	  fieldstr++;
	}
      else if (ISDIGIT (*fieldstr))
	{
	  /* detect overflow. */
	  {
	    double check_overflow;

	   check_overflow = value + (*fieldstr - '0') / 10.0;
	   if(check_overflow > UINT_MAX / 10.0)
	      FATAL_ERROR (_("too large number"));
	  }
	  value = 10 * value + *fieldstr - '0';
	  fieldstr++;
	}
      else
	FATAL_ERROR (_("invalid byte, character or field list"));
    }

  max_range_endpoint = 0;
  for (i = 0; i < n_rp; i++)
    {
      if (rp[i].hi > max_range_endpoint)
	max_range_endpoint = rp[i].hi;
    }

  /* Allocate an array large enough so that it may be indexed by
     the field numbers corresponding to all finite ranges
     (i.e. `2-6' or `-4', but not `5-') in FIELDSTR.  */

  printable_field = (int *) xmalloc ((max_range_endpoint + 1) * sizeof (int));
  memset (printable_field, 0, (max_range_endpoint + 1) * sizeof (int));

  /* Set the array entries corresponding to integers in the ranges of RP.  */
  for (i = 0; i < n_rp; i++)
    {
      unsigned int j;
      for (j = rp[i].lo; j <= rp[i].hi; j++)
	{
	  printable_field[j] = 1;
	}
    }

  free (rp);

  return field_found;
}

/* Read from stream STREAM, printing to standard output any selected bytes.  */
static void
cut_bytes (FILE *stream)
{
  int byte_idx;			/* number of bytes in the line so far. */

  byte_idx = 0;
  while (1)
    {
      int c;			/* each byte from the file. */

      c = getc (stream);

      if (c == '\n')
	{
	  putchar ('\n');
	  byte_idx = 0;
	}
      else if (c == EOF)
	{
	  if (byte_idx > 0)
	    putchar ('\n');
	  break;
	}
      else
	{
	  ++byte_idx;
	  if (print_kth (byte_idx))
	    {
	      putchar (c);
	    }
	}
    }
}

#if HAVE_MBRTOWC
/* This function is in use for the following case.

   1. Read from the stream STREAM, printing to standard output any selected
   characters. 

   2. Read from stream STREAM, printing to standard output any selected bytes,
   without splitting multibyte characters.  */
 
static void
cut_characters_or_cut_bytes_no_split (FILE *stream)
{
  int idx;		/* number of bytes or characters in the line so far. */
  char buf[MB_LEN_MAX + BUFSIZ];  /* For spooling a read byte sequence. */
  char *bufpos;		/* Next read position of BUF. */
  size_t buflen;	/* The length of the byte sequence in buf. */
  wint_t wc;		/* A gotten wide character. */
  size_t mblength;	/* The byte size of a multibyte character which shows
			   as same character as WC. */
  mbstate_t state;	/* State of the stream. */
  int convfail;		/* 1, when conversion is failed. Otherwise 0. */

  idx = 0;
  buflen = 0;
  bufpos = buf;
  memset (&state, '\0', sizeof(mbstate_t));

  while (1)
    {
      REFILL_BUFFER (buf, bufpos, buflen, stream);

      GET_NEXT_WC_FROM_BUFFER (wc, bufpos, buflen, mblength, state, convfail);

      if (wc == WEOF)
	{
	  if (idx > 0)
	    putchar ('\n');