- {
- case '\0':
- if (backslash_escapes)
- {
- STORE ('\\');
- STORE ('0');
- STORE ('0');
- c = '0';
- }
- break;
-
- case '?':
- switch (quoting_style)
- {
- case shell_quoting_style:
- goto use_shell_always_quoting_style;
-
- case c_quoting_style:
- if (i + 2 < argsize && arg[i + 1] == '?')
- switch (arg[i + 2])
- {
- case '!': case '\'':
- case '(': case ')': case '-': case '/':
- case '<': case '=': case '>':
- /* Escape the second '?' in what would otherwise be
- a trigraph. */
- c = arg[i + 2];
- i += 2;
- STORE ('?');
- STORE ('\\');
- STORE ('?');
- break;
- }
- break;
-
- default:
- break;
- }
- break;
-
- case '\a': esc = 'a'; goto c_escape;
- case '\b': esc = 'b'; goto c_escape;
- case '\f': esc = 'f'; goto c_escape;
- case '\n': esc = 'n'; goto c_and_shell_escape;
- case '\r': esc = 'r'; goto c_and_shell_escape;
- case '\t': esc = 't'; goto c_and_shell_escape;
- case '\v': esc = 'v'; goto c_escape;
- case '\\': esc = c; goto c_and_shell_escape;
-
- c_and_shell_escape:
- if (quoting_style == shell_quoting_style)
- goto use_shell_always_quoting_style;
- c_escape:
- if (backslash_escapes)
- {
- c = esc;
- goto store_escape;
- }
- break;
-
- case '#': case '~':
- if (i != 0)
- break;
- /* Fall through. */
- case ' ':
- case '!': /* special in bash */
- case '"': case '$': case '&':
- case '(': case ')': case '*': case ';':
- case '<': case '>': case '[':
- case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */
- case '`': case '|':
- /* A shell special character. In theory, '$' and '`' could
- be the first bytes of multibyte characters, which means
- we should check them with mbrtowc, but in practice this
- doesn't happen so it's not worth worrying about. */
- if (quoting_style == shell_quoting_style)
- goto use_shell_always_quoting_style;
- break;
-
- case '\'':
- switch (quoting_style)
- {
- case shell_quoting_style:
- goto use_shell_always_quoting_style;
-
- case shell_always_quoting_style:
- STORE ('\'');
- STORE ('\\');
- STORE ('\'');
- break;
-
- default:
- break;
- }
- break;
-
- case '%': case '+': case ',': case '-': case '.': case '/':
- case '0': case '1': case '2': case '3': case '4': case '5':
- case '6': case '7': case '8': case '9': case ':': case '=':
- case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
- case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
- case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
- case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
- case 'Y': case 'Z': case ']': case '_': case 'a': case 'b':
- case 'c': case 'd': case 'e': case 'f': case 'g': case 'h':
- case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
- case 'o': case 'p': case 'q': case 'r': case 's': case 't':
- case 'u': case 'v': case 'w': case 'x': case 'y': case 'z':
- case '{': case '}':
- /* These characters don't cause problems, no matter what the
- quoting style is. They cannot start multibyte sequences. */
- break;
-
- default:
- /* If we have a multibyte sequence, copy it until we reach
- its end, find an error, or come back to the initial shift
- state. For C-like styles, if the sequence has
- unprintable characters, escape the whole sequence, since
- we can't easily escape single characters within it. */
- {
- /* Length of multibyte sequence found so far. */
- size_t m;
-
- int printable;
-
- if (unibyte_locale)
- {
- m = 1;
- printable = isprint (c);
- }
- else
- {
- mbstate_t mbstate;
- memset (&mbstate, 0, sizeof mbstate);
-
- m = 0;
- printable = 1;
- if (argsize == SIZE_MAX)
- argsize = strlen (arg);
-
- do
- {
- wchar_t w;
- size_t bytes = mbrtowc (&w, &arg[i + m],
- argsize - (i + m), &mbstate);
- if (bytes == 0)
- break;
- else if (bytes == (size_t) -1)
- {
- printable = 0;
- break;
- }
- else if (bytes == (size_t) -2)
- {
- printable = 0;
- while (i + m < argsize && arg[i + m])
- m++;
- break;
- }
- else
- {
- if (! iswprint (w))
- printable = 0;
- m += bytes;
- }
- }
- while (! mbsinit (&mbstate));
- }
-
- if (1 < m || (backslash_escapes && ! printable))
- {
- /* Output a multibyte sequence, or an escaped
- unprintable unibyte character. */
- size_t ilim = i + m;
-
- for (;;)
- {
- if (backslash_escapes && ! printable)
- {
- STORE ('\\');
- STORE ('0' + (c >> 6));
- STORE ('0' + ((c >> 3) & 7));
- c = '0' + (c & 7);
- }
- if (ilim <= i + 1)
- break;
- STORE (c);
- c = arg[++i];
- }
-
- goto store_c;
- }
- }
- }
-
- if (! (backslash_escapes
- && o->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS))))
- goto store_c;
+ {
+ case '\0':
+ if (backslash_escapes)
+ {
+ if (elide_outer_quotes)
+ goto force_outer_quoting_style;
+ STORE ('\\');
+ /* If quote_string were to begin with digits, we'd need to
+ test for the end of the arg as well. However, it's
+ hard to imagine any locale that would use digits in
+ quotes, and set_custom_quoting is documented not to
+ accept them. */
+ if (i + 1 < argsize && '0' <= arg[i + 1] && arg[i + 1] <= '9')
+ {
+ STORE ('0');
+ STORE ('0');
+ }
+ c = '0';
+ /* We don't have to worry that this last '0' will be
+ backslash-escaped because, again, quote_string should
+ not start with it and because quote_these_too is
+ documented as not accepting it. */
+ }
+ else if (flags & QA_ELIDE_NULL_BYTES)
+ continue;
+ break;
+
+ case '?':
+ switch (quoting_style)
+ {
+ case shell_always_quoting_style:
+ if (elide_outer_quotes)
+ goto force_outer_quoting_style;
+ break;
+
+ case c_quoting_style:
+ if ((flags & QA_SPLIT_TRIGRAPHS)
+ && i + 2 < argsize && arg[i + 1] == '?')
+ switch (arg[i + 2])
+ {
+ case '!': case '\'':
+ case '(': case ')': case '-': case '/':
+ case '<': case '=': case '>':
+ /* Escape the second '?' in what would otherwise be
+ a trigraph. */
+ if (elide_outer_quotes)
+ goto force_outer_quoting_style;
+ c = arg[i + 2];
+ i += 2;
+ STORE ('?');
+ STORE ('"');
+ STORE ('"');
+ STORE ('?');
+ break;
+
+ default:
+ break;
+ }
+ break;
+
+ default:
+ break;
+ }
+ break;
+
+ case '\a': esc = 'a'; goto c_escape;
+ case '\b': esc = 'b'; goto c_escape;
+ case '\f': esc = 'f'; goto c_escape;
+ case '\n': esc = 'n'; goto c_and_shell_escape;
+ case '\r': esc = 'r'; goto c_and_shell_escape;
+ case '\t': esc = 't'; goto c_and_shell_escape;
+ case '\v': esc = 'v'; goto c_escape;
+ case '\\': esc = c;
+ /* No need to escape the escape if we are trying to elide
+ outer quotes and nothing else is problematic. */
+ if (backslash_escapes && elide_outer_quotes && quote_string_len)
+ goto store_c;
+
+ c_and_shell_escape:
+ if (quoting_style == shell_always_quoting_style
+ && elide_outer_quotes)
+ goto force_outer_quoting_style;
+ /* Fall through. */
+ c_escape:
+ if (backslash_escapes)
+ {
+ c = esc;
+ goto store_escape;
+ }
+ break;
+
+ case '{': case '}': /* sometimes special if isolated */
+ if (! (argsize == SIZE_MAX ? arg[1] == '\0' : argsize == 1))
+ break;
+ /* Fall through. */
+ case '#': case '~':
+ if (i != 0)
+ break;
+ /* Fall through. */
+ case ' ':
+ case '!': /* special in bash */
+ case '"': case '$': case '&':
+ case '(': case ')': case '*': case ';':
+ case '<':
+ case '=': /* sometimes special in 0th or (with "set -k") later args */
+ case '>': case '[':
+ case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */
+ case '`': case '|':
+ /* A shell special character. In theory, '$' and '`' could
+ be the first bytes of multibyte characters, which means
+ we should check them with mbrtowc, but in practice this
+ doesn't happen so it's not worth worrying about. */
+ if (quoting_style == shell_always_quoting_style
+ && elide_outer_quotes)
+ goto force_outer_quoting_style;
+ break;
+
+ case '\'':
+ if (quoting_style == shell_always_quoting_style)
+ {
+ if (elide_outer_quotes)
+ goto force_outer_quoting_style;
+ STORE ('\'');
+ STORE ('\\');
+ STORE ('\'');
+ }
+ break;
+
+ case '%': case '+': case ',': case '-': case '.': case '/':
+ case '0': case '1': case '2': case '3': case '4': case '5':
+ case '6': case '7': case '8': case '9': case ':':
+ case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
+ case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
+ case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
+ case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
+ case 'Y': case 'Z': case ']': case '_': case 'a': case 'b':
+ case 'c': case 'd': case 'e': case 'f': case 'g': case 'h':
+ case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
+ case 'o': case 'p': case 'q': case 'r': case 's': case 't':
+ case 'u': case 'v': case 'w': case 'x': case 'y': case 'z':
+ /* These characters don't cause problems, no matter what the
+ quoting style is. They cannot start multibyte sequences.
+ A digit or a special letter would cause trouble if it
+ appeared at the beginning of quote_string because we'd then
+ escape by prepending a backslash. However, it's hard to
+ imagine any locale that would use digits or letters as
+ quotes, and set_custom_quoting is documented not to accept
+ them. Also, a digit or a special letter would cause
+ trouble if it appeared in quote_these_too, but that's also
+ documented as not accepting them. */
+ break;
+
+ default:
+ /* If we have a multibyte sequence, copy it until we reach
+ its end, find an error, or come back to the initial shift
+ state. For C-like styles, if the sequence has
+ unprintable characters, escape the whole sequence, since
+ we can't easily escape single characters within it. */
+ {
+ /* Length of multibyte sequence found so far. */
+ size_t m;
+
+ bool printable;
+
+ if (unibyte_locale)
+ {
+ m = 1;
+ printable = isprint (c) != 0;
+ }
+ else
+ {
+ mbstate_t mbstate;
+ memset (&mbstate, 0, sizeof mbstate);
+
+ m = 0;
+ printable = true;
+ if (argsize == SIZE_MAX)
+ argsize = strlen (arg);
+
+ do
+ {
+ wchar_t w;
+ size_t bytes = mbrtowc (&w, &arg[i + m],
+ argsize - (i + m), &mbstate);
+ if (bytes == 0)
+ break;
+ else if (bytes == (size_t) -1)
+ {
+ printable = false;
+ break;
+ }
+ else if (bytes == (size_t) -2)
+ {
+ printable = false;
+ while (i + m < argsize && arg[i + m])
+ m++;
+ break;
+ }
+ else
+ {
+ /* Work around a bug with older shells that "see" a '\'
+ that is really the 2nd byte of a multibyte character.
+ In practice the problem is limited to ASCII
+ chars >= '@' that are shell special chars. */
+ if ('[' == 0x5b && elide_outer_quotes
+ && quoting_style == shell_always_quoting_style)
+ {
+ size_t j;
+ for (j = 1; j < bytes; j++)
+ switch (arg[i + m + j])
+ {
+ case '[': case '\\': case '^':
+ case '`': case '|':
+ goto force_outer_quoting_style;
+
+ default:
+ break;
+ }
+ }
+
+ if (! iswprint (w))
+ printable = false;
+ m += bytes;
+ }
+ }
+ while (! mbsinit (&mbstate));
+ }
+
+ if (1 < m || (backslash_escapes && ! printable))
+ {
+ /* Output a multibyte sequence, or an escaped
+ unprintable unibyte character. */
+ size_t ilim = i + m;
+
+ for (;;)
+ {
+ if (backslash_escapes && ! printable)
+ {
+ if (elide_outer_quotes)
+ goto force_outer_quoting_style;
+ STORE ('\\');
+ STORE ('0' + (c >> 6));
+ STORE ('0' + ((c >> 3) & 7));
+ c = '0' + (c & 7);
+ }
+ else if (is_right_quote)
+ {
+ STORE ('\\');
+ is_right_quote = false;
+ }
+ if (ilim <= i + 1)
+ break;
+ STORE (c);
+ c = arg[++i];
+ }
+
+ goto store_c;
+ }
+ }
+ }
+
+ if (! ((backslash_escapes || elide_outer_quotes)
+ && quote_these_too
+ && quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS)))
+ && !is_right_quote)
+ goto store_c;