diff.c

这是一个同样来自贝尔实验室的和UNIX有着渊源的操作系统, 其简洁的设计和实现易于我们学习和理解

	case 'H':
	  /* Turn on heuristics that speed processing of large files
	     with a small density of changes.  */
	  heuristic = 1;
	  break;

	case 'i':
	  /* Ignore changes in case.  */
	  ignore_case_flag = 1;
	  ignore_some_changes = 1;
	  ignore_some_line_changes = 1;
	  break;

	case 'I':
	  /* Ignore changes affecting only lines that match the
	     specified regexp.  */
	  add_regexp (&ignore_regexp_list, optarg);
	  ignore_some_changes = 1;
	  break;

	case 'l':
	  /* Pass the output through `pr' to paginate it.  */
	  paginate_flag = 1;
#if !defined(SIGCHLD) && defined(SIGCLD)
#define SIGCHLD SIGCLD
#endif
#ifdef SIGCHLD
	  /* Pagination requires forking and waiting, and
	     System V fork+wait does not work if SIGCHLD is ignored.  */
	  signal (SIGCHLD, SIG_DFL);
#endif
	  break;

	case 'L':
	  /* Specify file labels for `-c' output headers.  */
	  if (!file_label[0])
	    file_label[0] = optarg;
	  else if (!file_label[1])
	    file_label[1] = optarg;
	  else
	    fatal ("too many file label options");
	  break;

	case 'n':
	  /* Output RCS-style diffs, like `-f' except that each command
	     specifies the number of lines affected.  */
	  specify_style (OUTPUT_RCS);
	  break;

	case 'N':
	  /* When comparing directories, if a file appears only in one
	     directory, treat it as present but empty in the other.  */
	  entire_new_file_flag = 1;
	  break;

	case 'o':
	  /* Output in the old tradition style.  */
	  specify_style (OUTPUT_NORMAL);
	  break;

	case 'p':
	  /* Make context-style output and show name of last C function.  */
	  show_c_function = 1;
	  add_regexp (&function_regexp_list, "^[_a-zA-Z$]");
	  break;

	case 'P':
	  /* When comparing directories, if a file appears only in
	     the second directory of the two,
	     treat it as present but empty in the other.  */
	  unidirectional_new_file_flag = 1;
	  break;

	case 'q':
	  no_details_flag = 1;
	  break;

	case 'r':
	  /* When comparing directories,
	     recursively compare any subdirectories found.  */
	  recursive = 1;
	  break;

	case 's':
	  /* Print a message if the files are the same.  */
	  print_file_same_flag = 1;
	  break;

	case 'S':
	  /* When comparing directories, start with the specified
	     file name.  This is used for resuming an aborted comparison.  */
	  dir_start_file = optarg;
	  break;

	case 't':
	  /* Expand tabs to spaces in the output so that it preserves
	     the alignment of the input files.  */
	  tab_expand_flag = 1;
	  break;

	case 'T':
	  /* Use a tab in the output, rather than a space, before the
	     text of an input line, so as to keep the proper alignment
	     in the input line without changing the characters in it.  */
	  tab_align_flag = 1;
	  break;

	case 'u':
	  /* Output the context diff in unidiff format.  */
	  specify_style (OUTPUT_UNIFIED);
	  break;

	case 'v':
	  printf ("diff - GNU diffutils version %s\n", version_string);
	  exit (0);

	case 'w':
	  /* Ignore horizontal white space when comparing lines.  */
	  ignore_all_space_flag = 1;
	  ignore_some_changes = 1;
	  ignore_some_line_changes = 1;
	  break;

	case 'x':
	  add_exclude (optarg);
	  break;

	case 'X':
	  if (add_exclude_file (optarg) != 0)
	    pfatal_with_name (optarg);
	  break;

	case 'y':
	  /* Use side-by-side (sdiff-style) columnar output. */
	  specify_style (OUTPUT_SDIFF);
	  break;

	case 'W':
	  /* Set the line width for OUTPUT_SDIFF.  */
	  if (ck_atoi (optarg, &width) || width <= 0)
	    fatal ("column width must be a positive integer");
	  break;

	case 129:
	  sdiff_left_only = 1;
	  break;

	case 130:
	  sdiff_skip_common_lines = 1;
	  break;

	case 131:
	  /* sdiff-style columns output. */
	  specify_style (OUTPUT_SDIFF);
	  sdiff_help_sdiff = 1;
	  break;

	case 132:
	case 133:
	case 134:
	  specify_style (OUTPUT_IFDEF);
	  if (specify_format (&line_format[c - 132], optarg) != 0)
	    error ("conflicting line format", 0, 0);
	  break;

	case 135:
	  specify_style (OUTPUT_IFDEF);
	  {
	    int i, err = 0;
	    for (i = 0; i < sizeof (line_format) / sizeof (*line_format); i++)
	      err |= specify_format (&line_format[i], optarg);
	    if (err)
	      error ("conflicting line format", 0, 0);
	  }
	  break;

	case 136:
	case 137:
	case 138:
	case 139:
	  specify_style (OUTPUT_IFDEF);
	  if (specify_format (&group_format[c - 136], optarg) != 0)
	    error ("conflicting group format", 0, 0);
	  break;

	case 140:
	  if (ck_atoi (optarg, &horizon_lines) || horizon_lines < 0)
	    fatal ("horizon must be a nonnegative integer");
	  break;

	case 141:
	  usage ();
	  check_stdout ();
	  exit (0);

	case 142:
	  /* Use binary I/O when reading and writing data.
	     On Posix hosts, this has no effect.  */
#if HAVE_SETMODE
	  binary_I_O = 1;
	  setmode (STDOUT_FILENO, O_BINARY);
#endif
	  break;

	default:
	  try_help (0);
	}
      prev = c;
    }

  if (argc - optind != 2)
    try_help (argc - optind < 2 ? "missing operand" : "extra operand");


  {
    /*
     *	We maximize first the half line width, and then the gutter width,
     *	according to the following constraints:
     *	1.  Two half lines plus a gutter must fit in a line.
     *	2.  If the half line width is nonzero:
     *	    a.  The gutter width is at least GUTTER_WIDTH_MINIMUM.
     *	    b.  If tabs are not expanded to spaces,
     *		a half line plus a gutter is an integral number of tabs,
     *		so that tabs in the right column line up.
     */
    int t = tab_expand_flag ? 1 : TAB_WIDTH;
    int off = (width + t + GUTTER_WIDTH_MINIMUM) / (2*t)  *  t;
    sdiff_half_width = max (0, min (off - GUTTER_WIDTH_MINIMUM, width - off)),
    sdiff_column2_offset = sdiff_half_width ? off : width;
  }

  if (show_c_function && output_style != OUTPUT_UNIFIED)
    specify_style (OUTPUT_CONTEXT);

  if (output_style != OUTPUT_CONTEXT && output_style != OUTPUT_UNIFIED)
    context = 0;
  else if (context == -1)
    /* Default amount of context for -c.  */
    context = 3;

  if (output_style == OUTPUT_IFDEF)
    {
      /* Format arrays are char *, not char const *,
	 because integer formats are temporarily modified.
	 But it is safe to assign a constant like "%=" to a format array,
	 since "%=" does not format any integers.  */
      int i;
      for (i = 0; i < sizeof (line_format) / sizeof (*line_format); i++)
	if (!line_format[i])
	  line_format[i] = "%l\n";
      if (!group_format[OLD])
	group_format[OLD]
	  = group_format[UNCHANGED] ? group_format[UNCHANGED] : "%<";
      if (!group_format[NEW])
	group_format[NEW]
	  = group_format[UNCHANGED] ? group_format[UNCHANGED] : "%>";
      if (!group_format[UNCHANGED])
	group_format[UNCHANGED] = "%=";
      if (!group_format[CHANGED])
	group_format[CHANGED] = concat (group_format[OLD],
					group_format[NEW], "");
    }

  no_diff_means_no_output =
    (output_style == OUTPUT_IFDEF ?
      (!*group_format[UNCHANGED]
       || (strcmp (group_format[UNCHANGED], "%=") == 0
	   && !*line_format[UNCHANGED]))
     : output_style == OUTPUT_SDIFF ? sdiff_skip_common_lines : 1);

  switch_string = option_list (argv + 1, optind - 1);

  val = compare_files (0, argv[optind], 0, argv[optind + 1], 0);

  /* Print any messages that were saved up for last.  */
  print_message_queue ();

  check_stdout ();
  exit (val);
  return val;
}

/* Add the compiled form of regexp PATTERN to REGLIST.  */

static void
add_regexp (reglist, pattern)
     struct regexp_list **reglist;
     char const *pattern;
{
  struct regexp_list *r;
  char const *m;

  r = (struct regexp_list *) xmalloc (sizeof (*r));
  bzero (r, sizeof (*r));
  r->buf.fastmap = xmalloc (256);
  m = re_compile_pattern (pattern, strlen (pattern), &r->buf);
  if (m != 0)
    error ("%s: %s", pattern, m);

  /* Add to the start of the list, since it's easier than the end.  */
  r->next = *reglist;
  *reglist = r;
}

static void
try_help (reason)
     char const *reason;
{
  if (reason)
    error ("%s", reason, 0);
  error ("Try `%s --help' for more information.", program_name, 0);
  exit (2);
}

static void
check_stdout ()
{
  if (ferror (stdout) || fclose (stdout) != 0)
    fatal ("write error");
}

static char const * const option_help[] = {
"-i  --ignore-case  Consider upper- and lower-case to be the same.",
"-w  --ignore-all-space  Ignore all white space.",
"-b  --ignore-space-change  Ignore changes in the amount of white space.",
"-B  --ignore-blank-lines  Ignore changes whose lines are all blank.",
"-I RE  --ignore-matching-lines=RE  Ignore changes whose lines all match RE.",
#if HAVE_SETMODE
"--binary  Read and write data in binary mode.",
#endif
"-a  --text  Treat all files as text.\n",
"-c  -C NUM  --context[=NUM]  Output NUM (default 2) lines of copied context.",
"-u  -U NUM  --unified[=NUM]  Output NUM (default 2) lines of unified context.",
"  -NUM  Use NUM context lines.",
"  -L LABEL  --label LABEL  Use LABEL instead of file name.",
"  -p  --show-c-function  Show which C function each change is in.",
"  -F RE  --show-function-line=RE  Show the most recent line matching RE.",
"-q  --brief  Output only whether files differ.",
"-e  --ed  Output an ed script.",
"-n  --rcs  Output an RCS format diff.",
"-y  --side-by-side  Output in two columns.",
"  -w NUM  --width=NUM  Output at most NUM (default 130) characters per line.",
"  --left-column  Output only the left column of common lines.",
"  --suppress-common-lines  Do not output common lines.",
"-DNAME  --ifdef=NAME  Output merged file to show `#ifdef NAME' diffs.",
"--GTYPE-group-format=GFMT  Similar, but format GTYPE input groups with GFMT.",
"--line-format=LFMT  Similar, but format all input lines with LFMT.",
"--LTYPE-line-format=LFMT  Similar, but format LTYPE input lines with LFMT.",
"  LTYPE is `old', `new', or `unchanged'.  GTYPE is LTYPE or `changed'.",
"  GFMT may contain:",
"    %<  lines from FILE1",
"    %>  lines from FILE2",
"    %=  lines common to FILE1 and FILE2",
"    %[-][WIDTH][.[PREC]]{doxX}LETTER  printf-style spec for LETTER",
"      LETTERs are as follows for new group, lower case for old group:",
"        F  first line number",
"        L  last line number",
"        N  number of lines = L-F+1",
"        E  F-1",
"        M  L+1",
"  LFMT may contain:",
"    %L  contents of line",
"    %l  contents of line, excluding any trailing newline",
"    %[-][WIDTH][.[PREC]]{doxX}n  printf-style spec for input line number",
"  Either GFMT or LFMT may contain:",
"    %%  %",
"    %c'C'  the single character C",
"    %c'\\OOO'  the character with octal code OOO\n",
"-l  --paginate  Pass the output through `pr' to paginate it.",
"-t  --expand-tabs  Expand tabs to spaces in output.",
"-T  --initial-tab  Make tabs line up by prepending a tab.\n",
"-r  --recursive  Recursively compare any subdirectories found.",
"-N  --new-file  Treat absent files as empty.",
"-P  --unidirectional-new-file  Treat absent first files as empty.",