/* Various declarations for language-independent pretty-print subroutines.
- Copyright (C) 2003, 2004, 2005 Free Software Foundation, Inc.
+ Copyright (C) 2003, 2004, 2005, 2007, 2008, 2009, 2010
+ Free Software Foundation, Inc.
Contributed by Gabriel Dos Reis <gdr@integrable-solutions.net>
This file is part of GCC.
GCC 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
+Software Foundation; either version 3, or (at your option) any later
version.
GCC is distributed in the hope that it will be useful, but WITHOUT ANY
for more details.
You should have received a copy of the GNU General Public License
-along with GCC; see the file COPYING. If not, write to the Free
-Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA
-02110-1301, USA. */
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
#include "config.h"
-#undef FLOAT /* This is for hpux. They should change hpux. */
-#undef FFS /* Some systems define this in param.h. */
#include "system.h"
#include "coretypes.h"
#include "intl.h"
#include "pretty-print.h"
-#include "tree.h"
-#define obstack_chunk_alloc xmalloc
-#define obstack_chunk_free free
+#if HAVE_ICONV
+#include <iconv.h>
+#endif
/* A pointer to the formatted diagnostic message. */
#define pp_formatted_text_data(PP) \
break; \
\
case 2: \
- pp_scalar (PP, "%ll" F, va_arg (ARG, long long T)); \
+ pp_scalar (PP, "%" HOST_LONG_LONG_FORMAT F, va_arg (ARG, long long T)); \
break; \
\
default: \
%.*s: a substring the length of which is specified by an argument
integer.
%Ns: likewise, but length specified as constant in the format string.
- %H: location_t.
- %J: a decl tree, from which DECL_SOURCE_LOCATION will be recorded.
Flag 'q': quote formatted text (must come immediately after '%').
Arguments can be used sequentially, or through %N$ resp. *N$
this point. */
memset (formatters, 0, sizeof formatters);
-
+
for (p = text->format_spec; *p; )
{
while (*p != '\0' && *p != '%')
{
case '\0':
gcc_unreachable ();
-
+
case '%':
obstack_1grow (&buffer->chunk_obstack, '%');
p++;
gcc_assert (chunk < PP_NL_ARGMAX * 2);
args[chunk++] = XOBFINISH (&buffer->chunk_obstack, const char *);
args[chunk] = 0;
-
+
/* Set output to the argument obstack, and switch line-wrapping and
prefixing off. */
buffer->obstack = &buffer->chunk_obstack;
(pp, *text->args_ptr, precision, unsigned, "x");
break;
- case 'H':
- {
- location_t *locus = va_arg (*text->args_ptr, location_t *);
- gcc_assert (text->locus != NULL);
- *text->locus = *locus;
- }
- break;
-
- case 'J':
- {
- tree t = va_arg (*text->args_ptr, tree);
- gcc_assert (text->locus != NULL);
- *text->locus = DECL_SOURCE_LOCATION (t);
- }
- break;
-
case '.':
{
int n;
{
if (pp->prefix != NULL)
{
- free ((char *) pp->prefix);
+ free (CONST_CAST (char *, pp->prefix));
pp->prefix = NULL;
}
}
pp_line_cutoff (pp) = maximum_length;
pp_prefixing_rule (pp) = DIAGNOSTICS_SHOW_PREFIX_ONCE;
pp_set_prefix (pp, prefix);
+ pp_translate_identifiers (pp) = true;
}
/* Append a string delimited by START and END to the output area of
pp_base (pp)->padding = pp_none;
}
}
+\f
+/* The string starting at P has LEN (at least 1) bytes left; if they
+ start with a valid UTF-8 sequence, return the length of that
+ sequence and set *VALUE to the value of that sequence, and
+ otherwise return 0 and set *VALUE to (unsigned int) -1. */
+
+static int
+decode_utf8_char (const unsigned char *p, size_t len, unsigned int *value)
+{
+ unsigned int t = *p;
+
+ if (len == 0)
+ abort ();
+ if (t & 0x80)
+ {
+ size_t utf8_len = 0;
+ unsigned int ch;
+ size_t i;
+ for (t = *p; t & 0x80; t <<= 1)
+ utf8_len++;
+
+ if (utf8_len > len || utf8_len < 2 || utf8_len > 6)
+ {
+ *value = (unsigned int) -1;
+ return 0;
+ }
+ ch = *p & ((1 << (7 - utf8_len)) - 1);
+ for (i = 1; i < utf8_len; i++)
+ {
+ unsigned int u = p[i];
+ if ((u & 0xC0) != 0x80)
+ {
+ *value = (unsigned int) -1;
+ return 0;
+ }
+ ch = (ch << 6) | (u & 0x3F);
+ }
+ if ( (ch <= 0x7F && utf8_len > 1)
+ || (ch <= 0x7FF && utf8_len > 2)
+ || (ch <= 0xFFFF && utf8_len > 3)
+ || (ch <= 0x1FFFFF && utf8_len > 4)
+ || (ch <= 0x3FFFFFF && utf8_len > 5)
+ || (ch >= 0xD800 && ch <= 0xDFFF))
+ {
+ *value = (unsigned int) -1;
+ return 0;
+ }
+ *value = ch;
+ return utf8_len;
+ }
+ else
+ {
+ *value = t;
+ return 1;
+ }
+}
+
+/* Allocator for identifier_to_locale and corresponding function to
+ free memory. */
+
+void *(*identifier_to_locale_alloc) (size_t) = xmalloc;
+void (*identifier_to_locale_free) (void *) = free;
+
+/* Given IDENT, an identifier in the internal encoding, return a
+ version of IDENT suitable for diagnostics in the locale character
+ set: either IDENT itself, or a string, allocated using
+ identifier_to_locale_alloc, converted to the locale character set
+ and using escape sequences if not representable in the locale
+ character set or containing control characters or invalid byte
+ sequences. Existing backslashes in IDENT are not doubled, so the
+ result may not uniquely specify the contents of an arbitrary byte
+ sequence identifier. */
+
+const char *
+identifier_to_locale (const char *ident)
+{
+ const unsigned char *uid = (const unsigned char *) ident;
+ size_t idlen = strlen (ident);
+ bool valid_printable_utf8 = true;
+ bool all_ascii = true;
+ size_t i;
+
+ for (i = 0; i < idlen;)
+ {
+ unsigned int c;
+ size_t utf8_len = decode_utf8_char (&uid[i], idlen - i, &c);
+ if (utf8_len == 0 || c <= 0x1F || (c >= 0x7F && c <= 0x9F))
+ {
+ valid_printable_utf8 = false;
+ break;
+ }
+ if (utf8_len > 1)
+ all_ascii = false;
+ i += utf8_len;
+ }
+
+ /* If IDENT contains invalid UTF-8 sequences (which may occur with
+ attributes putting arbitrary byte sequences in identifiers), or
+ control characters, we use octal escape sequences for all bytes
+ outside printable ASCII. */
+ if (!valid_printable_utf8)
+ {
+ char *ret = (char *) identifier_to_locale_alloc (4 * idlen + 1);
+ char *p = ret;
+ for (i = 0; i < idlen; i++)
+ {
+ if (uid[i] > 0x1F && uid[i] < 0x7F)
+ *p++ = uid[i];
+ else
+ {
+ sprintf (p, "\\%03o", uid[i]);
+ p += 4;
+ }
+ }
+ *p = 0;
+ return ret;
+ }
+
+ /* Otherwise, if it is valid printable ASCII, or printable UTF-8
+ with the locale character set being UTF-8, IDENT is used. */
+ if (all_ascii || locale_utf8)
+ return ident;
+
+ /* Otherwise IDENT is converted to the locale character set if
+ possible. */
+#if defined ENABLE_NLS && defined HAVE_LANGINFO_CODESET && HAVE_ICONV
+ if (locale_encoding != NULL)
+ {
+ iconv_t cd = iconv_open (locale_encoding, "UTF-8");
+ bool conversion_ok = true;
+ char *ret = NULL;
+ if (cd != (iconv_t) -1)
+ {
+ size_t ret_alloc = 4 * idlen + 1;
+ for (;;)
+ {
+ /* Repeat the whole conversion process as needed with
+ larger buffers so non-reversible transformations can
+ always be detected. */
+ ICONV_CONST char *inbuf = CONST_CAST (char *, ident);
+ char *outbuf;
+ size_t inbytesleft = idlen;
+ size_t outbytesleft = ret_alloc - 1;
+ size_t iconv_ret;
+
+ ret = (char *) identifier_to_locale_alloc (ret_alloc);
+ outbuf = ret;
+
+ if (iconv (cd, 0, 0, 0, 0) == (size_t) -1)
+ {
+ conversion_ok = false;
+ break;
+ }
+
+ iconv_ret = iconv (cd, &inbuf, &inbytesleft,
+ &outbuf, &outbytesleft);
+ if (iconv_ret == (size_t) -1 || inbytesleft != 0)
+ {
+ if (errno == E2BIG)
+ {
+ ret_alloc *= 2;
+ identifier_to_locale_free (ret);
+ ret = NULL;
+ continue;
+ }
+ else
+ {
+ conversion_ok = false;
+ break;
+ }
+ }
+ else if (iconv_ret != 0)
+ {
+ conversion_ok = false;
+ break;
+ }
+ /* Return to initial shift state. */
+ if (iconv (cd, 0, 0, &outbuf, &outbytesleft) == (size_t) -1)
+ {
+ if (errno == E2BIG)
+ {
+ ret_alloc *= 2;
+ identifier_to_locale_free (ret);
+ ret = NULL;
+ continue;
+ }
+ else
+ {
+ conversion_ok = false;
+ break;
+ }
+ }
+ *outbuf = 0;
+ break;
+ }
+ iconv_close (cd);
+ if (conversion_ok)
+ return ret;
+ }
+ }
+#endif
+
+ /* Otherwise, convert non-ASCII characters in IDENT to UCNs. */
+ {
+ char *ret = (char *) identifier_to_locale_alloc (10 * idlen + 1);
+ char *p = ret;
+ for (i = 0; i < idlen;)
+ {
+ unsigned int c;
+ size_t utf8_len = decode_utf8_char (&uid[i], idlen - i, &c);
+ if (utf8_len == 1)
+ *p++ = uid[i];
+ else
+ {
+ sprintf (p, "\\U%08x", c);
+ p += 10;
+ }
+ i += utf8_len;
+ }
+ *p = 0;
+ return ret;
+ }
+}