/* C language support routines for GDB, the GNU debugger.
Copyright (C) 1992, 1993, 1994, 1995, 1996, 1998, 1999, 2000, 2002, 2003,
- 2004, 2005, 2007, 2008 Free Software Foundation, Inc.
+ 2004, 2005, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
This file is part of GDB.
#include "demangle.h"
#include "cp-abi.h"
#include "cp-support.h"
+#include "gdb_obstack.h"
+#include <ctype.h>
extern void _initialize_c_language (void);
-static void c_emit_char (int c, struct ui_file * stream, int quoter);
-/* Print the character C on STREAM as part of the contents of a literal
- string whose delimiter is QUOTER. Note that that format for printing
- characters and strings is language specific. */
+/* Given a C string type, STR_TYPE, return the corresponding target
+ character set name. */
-static void
-c_emit_char (int c, struct ui_file *stream, int quoter)
+static const char *
+charset_for_string_type (enum c_string_type str_type,
+ struct gdbarch *gdbarch)
{
- const char *escape;
- int host_char;
+ switch (str_type & ~C_CHAR)
+ {
+ case C_STRING:
+ return target_charset (gdbarch);
+ case C_WIDE_STRING:
+ return target_wide_charset (gdbarch);
+ case C_STRING_16:
+ /* FIXME: UTF-16 is not always correct. */
+ if (gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG)
+ return "UTF-16BE";
+ else
+ return "UTF-16LE";
+ case C_STRING_32:
+ /* FIXME: UTF-32 is not always correct. */
+ if (gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG)
+ return "UTF-32BE";
+ else
+ return "UTF-32LE";
+ }
+ internal_error (__FILE__, __LINE__, "unhandled c_string_type");
+}
- c &= 0xFF; /* Avoid sign bit follies */
+/* Classify ELTTYPE according to what kind of character it is. Return
+ the enum constant representing the character type. Also set
+ *ENCODING to the name of the character set to use when converting
+ characters of this type in target BYTE_ORDER to the host character set. */
+
+static enum c_string_type
+classify_type (struct type *elttype, struct gdbarch *gdbarch,
+ const char **encoding)
+{
+ enum c_string_type result;
- escape = c_target_char_has_backslash_escape (c);
- if (escape)
+ /* We loop because ELTTYPE may be a typedef, and we want to
+ successively peel each typedef until we reach a type we
+ understand. We don't use CHECK_TYPEDEF because that will strip
+ all typedefs at once -- but in C, wchar_t is itself a typedef, so
+ that would do the wrong thing. */
+ while (elttype)
{
- if (quoter == '"' && strcmp (escape, "0") == 0)
- /* Print nulls embedded in double quoted strings as \000 to
- prevent ambiguity. */
- fprintf_filtered (stream, "\\000");
+ char *name = TYPE_NAME (elttype);
+
+ if (TYPE_CODE (elttype) == TYPE_CODE_CHAR || !name)
+ {
+ result = C_CHAR;
+ goto done;
+ }
+
+ if (!strcmp (name, "wchar_t"))
+ {
+ result = C_WIDE_CHAR;
+ goto done;
+ }
+
+ if (!strcmp (name, "char16_t"))
+ {
+ result = C_CHAR_16;
+ goto done;
+ }
+
+ if (!strcmp (name, "char32_t"))
+ {
+ result = C_CHAR_32;
+ goto done;
+ }
+
+ if (TYPE_CODE (elttype) != TYPE_CODE_TYPEDEF)
+ break;
+
+ /* Call for side effects. */
+ check_typedef (elttype);
+
+ if (TYPE_TARGET_TYPE (elttype))
+ elttype = TYPE_TARGET_TYPE (elttype);
else
- fprintf_filtered (stream, "\\%s", escape);
+ {
+ /* Perhaps check_typedef did not update the target type. In
+ this case, force the lookup again and hope it works out.
+ It never will for C, but it might for C++. */
+ CHECK_TYPEDEF (elttype);
+ }
}
- else if (target_char_to_host (c, &host_char)
- && host_char_print_literally (host_char))
+
+ /* Punt. */
+ result = C_CHAR;
+
+ done:
+ if (encoding)
+ *encoding = charset_for_string_type (result, gdbarch);
+
+ return result;
+}
+
+/* Return true if print_wchar can display W without resorting to a
+ numeric escape, false otherwise. */
+
+static int
+wchar_printable (gdb_wchar_t w)
+{
+ return (gdb_iswprint (w)
+ || w == LCST ('\a') || w == LCST ('\b')
+ || w == LCST ('\f') || w == LCST ('\n')
+ || w == LCST ('\r') || w == LCST ('\t')
+ || w == LCST ('\v'));
+}
+
+/* A helper function that converts the contents of STRING to wide
+ characters and then appends them to OUTPUT. */
+
+static void
+append_string_as_wide (const char *string, struct obstack *output)
+{
+ for (; *string; ++string)
+ {
+ gdb_wchar_t w = gdb_btowc (*string);
+ obstack_grow (output, &w, sizeof (gdb_wchar_t));
+ }
+}
+
+/* Print a wide character W to OUTPUT. ORIG is a pointer to the
+ original (target) bytes representing the character, ORIG_LEN is the
+ number of valid bytes. WIDTH is the number of bytes in a base
+ characters of the type. OUTPUT is an obstack to which wide
+ characters are emitted. QUOTER is a (narrow) character indicating
+ the style of quotes surrounding the character to be printed.
+ NEED_ESCAPE is an in/out flag which is used to track numeric
+ escapes across calls. */
+
+static void
+print_wchar (gdb_wint_t w, const gdb_byte *orig, int orig_len,
+ int width, enum bfd_endian byte_order, struct obstack *output,
+ int quoter, int *need_escapep)
+{
+ int need_escape = *need_escapep;
+
+ *need_escapep = 0;
+ if (gdb_iswprint (w) && (!need_escape || (!gdb_iswdigit (w)
+ && w != LCST ('8')
+ && w != LCST ('9'))))
{
- if (host_char == '\\' || host_char == quoter)
- fputs_filtered ("\\", stream);
- fprintf_filtered (stream, "%c", host_char);
+ gdb_wchar_t wchar = w;
+
+ if (w == gdb_btowc (quoter) || w == LCST ('\\'))
+ obstack_grow_wstr (output, LCST ("\\"));
+ obstack_grow (output, &wchar, sizeof (gdb_wchar_t));
}
else
- fprintf_filtered (stream, "\\%.3o", (unsigned int) c);
+ {
+ switch (w)
+ {
+ case LCST ('\a'):
+ obstack_grow_wstr (output, LCST ("\\a"));
+ break;
+ case LCST ('\b'):
+ obstack_grow_wstr (output, LCST ("\\b"));
+ break;
+ case LCST ('\f'):
+ obstack_grow_wstr (output, LCST ("\\f"));
+ break;
+ case LCST ('\n'):
+ obstack_grow_wstr (output, LCST ("\\n"));
+ break;
+ case LCST ('\r'):
+ obstack_grow_wstr (output, LCST ("\\r"));
+ break;
+ case LCST ('\t'):
+ obstack_grow_wstr (output, LCST ("\\t"));
+ break;
+ case LCST ('\v'):
+ obstack_grow_wstr (output, LCST ("\\v"));
+ break;
+ default:
+ {
+ int i;
+
+ for (i = 0; i + width <= orig_len; i += width)
+ {
+ char octal[30];
+ ULONGEST value;
+
+ value = extract_unsigned_integer (&orig[i], width, byte_order);
+ /* If the value fits in 3 octal digits, print it that
+ way. Otherwise, print it as a hex escape. */
+ if (value <= 0777)
+ sprintf (octal, "\\%.3o", (int) (value & 0777));
+ else
+ sprintf (octal, "\\x%lx", (long) value);
+ append_string_as_wide (octal, output);
+ }
+ /* If we somehow have extra bytes, print them now. */
+ while (i < orig_len)
+ {
+ char octal[5];
+
+ sprintf (octal, "\\%.3o", orig[i] & 0xff);
+ append_string_as_wide (octal, output);
+ ++i;
+ }
+
+ *need_escapep = 1;
+ }
+ break;
+ }
+ }
}
+/* Print the character C on STREAM as part of the contents of a literal
+ string whose delimiter is QUOTER. Note that that format for printing
+ characters and strings is language specific. */
+
void
-c_printchar (int c, struct ui_file *stream)
+c_emit_char (int c, struct type *type,
+ struct ui_file *stream, int quoter)
{
+ enum bfd_endian byte_order = gdbarch_byte_order (get_type_arch (type));
+ struct obstack wchar_buf, output;
+ struct cleanup *cleanups;
+ const char *encoding;
+ gdb_byte *buf;
+ struct wchar_iterator *iter;
+ int need_escape = 0;
+
+ classify_type (type, get_type_arch (type), &encoding);
+
+ buf = alloca (TYPE_LENGTH (type));
+ pack_long (buf, type, c);
+
+ iter = make_wchar_iterator (buf, TYPE_LENGTH (type), encoding,
+ TYPE_LENGTH (type));
+ cleanups = make_cleanup_wchar_iterator (iter);
+
+ /* This holds the printable form of the wchar_t data. */
+ obstack_init (&wchar_buf);
+ make_cleanup_obstack_free (&wchar_buf);
+
+ while (1)
+ {
+ int num_chars;
+ gdb_wchar_t *chars;
+ const gdb_byte *buf;
+ size_t buflen;
+ int print_escape = 1;
+ enum wchar_iterate_result result;
+
+ num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+ if (num_chars < 0)
+ break;
+ if (num_chars > 0)
+ {
+ /* If all characters are printable, print them. Otherwise,
+ we're going to have to print an escape sequence. We
+ check all characters because we want to print the target
+ bytes in the escape sequence, and we don't know character
+ boundaries there. */
+ int i;
+
+ print_escape = 0;
+ for (i = 0; i < num_chars; ++i)
+ if (!wchar_printable (chars[i]))
+ {
+ print_escape = 1;
+ break;
+ }
+
+ if (!print_escape)
+ {
+ for (i = 0; i < num_chars; ++i)
+ print_wchar (chars[i], buf, buflen, TYPE_LENGTH (type),
+ byte_order, &wchar_buf, quoter, &need_escape);
+ }
+ }
+
+ /* This handles the NUM_CHARS == 0 case as well. */
+ if (print_escape)
+ print_wchar (gdb_WEOF, buf, buflen, TYPE_LENGTH (type), byte_order,
+ &wchar_buf, quoter, &need_escape);
+ }
+
+ /* The output in the host encoding. */
+ obstack_init (&output);
+ make_cleanup_obstack_free (&output);
+
+ convert_between_encodings (INTERMEDIATE_ENCODING, host_charset (),
+ obstack_base (&wchar_buf),
+ obstack_object_size (&wchar_buf),
+ 1, &output, translit_char);
+ obstack_1grow (&output, '\0');
+
+ fputs_filtered (obstack_base (&output), stream);
+
+ do_cleanups (cleanups);
+}
+
+void
+c_printchar (int c, struct type *type, struct ui_file *stream)
+{
+ enum c_string_type str_type;
+
+ str_type = classify_type (type, get_type_arch (type), NULL);
+ switch (str_type)
+ {
+ case C_CHAR:
+ break;
+ case C_WIDE_CHAR:
+ fputc_filtered ('L', stream);
+ break;
+ case C_CHAR_16:
+ fputc_filtered ('u', stream);
+ break;
+ case C_CHAR_32:
+ fputc_filtered ('U', stream);
+ break;
+ }
+
fputc_filtered ('\'', stream);
- LA_EMIT_CHAR (c, stream, '\'');
+ LA_EMIT_CHAR (c, type, stream, '\'');
fputc_filtered ('\'', stream);
}
printing LENGTH characters, or if FORCE_ELLIPSES. */
void
-c_printstr (struct ui_file *stream, const gdb_byte *string,
- unsigned int length, int width, int force_ellipses,
+c_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
+ unsigned int length, const char *user_encoding, int force_ellipses,
const struct value_print_options *options)
{
+ enum bfd_endian byte_order = gdbarch_byte_order (get_type_arch (type));
unsigned int i;
unsigned int things_printed = 0;
int in_quotes = 0;
int need_comma = 0;
+ int width = TYPE_LENGTH (type);
+ struct obstack wchar_buf, output;
+ struct cleanup *cleanup;
+ enum c_string_type str_type;
+ const char *type_encoding;
+ const char *encoding;
+ struct wchar_iterator *iter;
+ int finished = 0;
+ int need_escape = 0;
+
+ if (length == -1)
+ {
+ unsigned long current_char = 1;
+
+ for (i = 0; current_char; ++i)
+ {
+ QUIT;
+ current_char = extract_unsigned_integer (string + i * width,
+ width, byte_order);
+ }
+ length = i;
+ }
/* If the string was not truncated due to `set print elements', and
the last byte of it is a null, we don't print that, in traditional C
style. */
if (!force_ellipses
&& length > 0
- && (extract_unsigned_integer (string + (length - 1) * width, width)
- == '\0'))
+ && (extract_unsigned_integer (string + (length - 1) * width,
+ width, byte_order) == 0))
length--;
+ str_type = (classify_type (type, get_type_arch (type), &type_encoding)
+ & ~C_CHAR);
+ switch (str_type)
+ {
+ case C_STRING:
+ break;
+ case C_WIDE_STRING:
+ fputs_filtered ("L", stream);
+ break;
+ case C_STRING_16:
+ fputs_filtered ("u", stream);
+ break;
+ case C_STRING_32:
+ fputs_filtered ("U", stream);
+ break;
+ }
+
+ encoding = (user_encoding && *user_encoding) ? user_encoding : type_encoding;
+
if (length == 0)
{
fputs_filtered ("\"\"", stream);
return;
}
- for (i = 0; i < length && things_printed < options->print_max; ++i)
+ /* Arrange to iterate over the characters, in wchar_t form. */
+ iter = make_wchar_iterator (string, length * width, encoding, width);
+ cleanup = make_cleanup_wchar_iterator (iter);
+
+ /* WCHAR_BUF is the obstack we use to represent the string in
+ wchar_t form. */
+ obstack_init (&wchar_buf);
+ make_cleanup_obstack_free (&wchar_buf);
+
+ while (!finished && things_printed < options->print_max)
{
- /* Position of the character we are examining
- to see whether it is repeated. */
- unsigned int rep1;
- /* Number of repetitions we have detected so far. */
- unsigned int reps;
- unsigned long current_char;
+ int num_chars;
+ enum wchar_iterate_result result;
+ gdb_wchar_t *chars;
+ const gdb_byte *buf;
+ size_t buflen;
QUIT;
if (need_comma)
{
- fputs_filtered (", ", stream);
+ obstack_grow_wstr (&wchar_buf, LCST (", "));
need_comma = 0;
}
- current_char = extract_unsigned_integer (string + i * width, width);
+ num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+ /* We only look at repetitions when we were able to convert a
+ single character in isolation. This makes the code simpler
+ and probably does the sensible thing in the majority of
+ cases. */
+ while (num_chars == 1 && things_printed < options->print_max)
+ {
+ /* Count the number of repetitions. */
+ unsigned int reps = 0;
+ gdb_wchar_t current_char = chars[0];
+ const gdb_byte *orig_buf = buf;
+ int orig_len = buflen;
+
+ if (need_comma)
+ {
+ obstack_grow_wstr (&wchar_buf, LCST (", "));
+ need_comma = 0;
+ }
- rep1 = i + 1;
- reps = 1;
- while (rep1 < length
- && extract_unsigned_integer (string + rep1 * width, width)
- == current_char)
+ while (num_chars == 1 && current_char == chars[0])
+ {
+ num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+ ++reps;
+ }
+
+ /* Emit CURRENT_CHAR according to the repetition count and
+ options. */
+ if (reps > options->repeat_count_threshold)
+ {
+ if (in_quotes)
+ {
+ if (options->inspect_it)
+ obstack_grow_wstr (&wchar_buf, LCST ("\\\", "));
+ else
+ obstack_grow_wstr (&wchar_buf, LCST ("\", "));
+ in_quotes = 0;
+ }
+ obstack_grow_wstr (&wchar_buf, LCST ("'"));
+ need_escape = 0;
+ print_wchar (current_char, orig_buf, orig_len, width,
+ byte_order, &wchar_buf, '\'', &need_escape);
+ obstack_grow_wstr (&wchar_buf, LCST ("'"));
+ {
+ /* Painful gyrations. */
+ int j;
+ char *s = xstrprintf (_(" <repeats %u times>"), reps);
+
+ for (j = 0; s[j]; ++j)
+ {
+ gdb_wchar_t w = gdb_btowc (s[j]);
+ obstack_grow (&wchar_buf, &w, sizeof (gdb_wchar_t));
+ }
+ xfree (s);
+ }
+ things_printed += options->repeat_count_threshold;
+ need_comma = 1;
+ }
+ else
+ {
+ /* Saw the character one or more times, but fewer than
+ the repetition threshold. */
+ if (!in_quotes)
+ {
+ if (options->inspect_it)
+ obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
+ else
+ obstack_grow_wstr (&wchar_buf, LCST ("\""));
+ in_quotes = 1;
+ need_escape = 0;
+ }
+
+ while (reps-- > 0)
+ {
+ print_wchar (current_char, orig_buf, orig_len, width,
+ byte_order, &wchar_buf, '"', &need_escape);
+ ++things_printed;
+ }
+ }
+ }
+
+ /* NUM_CHARS and the other outputs from wchar_iterate are valid
+ here regardless of which branch was taken above. */
+ if (num_chars < 0)
{
- ++rep1;
- ++reps;
+ /* Hit EOF. */
+ finished = 1;
+ break;
}
- if (reps > options->repeat_count_threshold)
+ switch (result)
{
- if (in_quotes)
+ case wchar_iterate_invalid:
+ if (!in_quotes)
{
if (options->inspect_it)
- fputs_filtered ("\\\", ", stream);
+ obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
else
- fputs_filtered ("\", ", stream);
- in_quotes = 0;
+ obstack_grow_wstr (&wchar_buf, LCST ("\""));
+ in_quotes = 1;
}
- LA_PRINT_CHAR (current_char, stream);
- fprintf_filtered (stream, _(" <repeats %u times>"), reps);
- i = rep1 - 1;
- things_printed += options->repeat_count_threshold;
- need_comma = 1;
- }
- else
- {
- if (!in_quotes)
+ need_escape = 0;
+ print_wchar (gdb_WEOF, buf, buflen, width, byte_order, &wchar_buf,
+ '"', &need_escape);
+ break;
+
+ case wchar_iterate_incomplete:
+ if (in_quotes)
{
if (options->inspect_it)
- fputs_filtered ("\\\"", stream);
+ obstack_grow_wstr (&wchar_buf, LCST ("\\\","));
else
- fputs_filtered ("\"", stream);
- in_quotes = 1;
+ obstack_grow_wstr (&wchar_buf, LCST ("\","));
+ in_quotes = 0;
}
- LA_EMIT_CHAR (current_char, stream, '"');
- ++things_printed;
+ obstack_grow_wstr (&wchar_buf, LCST (" <incomplete sequence "));
+ print_wchar (gdb_WEOF, buf, buflen, width, byte_order, &wchar_buf,
+ 0, &need_escape);
+ obstack_grow_wstr (&wchar_buf, LCST (">"));
+ finished = 1;
+ break;
}
}
if (in_quotes)
{
if (options->inspect_it)
- fputs_filtered ("\\\"", stream);
+ obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
else
- fputs_filtered ("\"", stream);
+ obstack_grow_wstr (&wchar_buf, LCST ("\""));
}
- if (force_ellipses || i < length)
- fputs_filtered ("...", stream);
+ if (force_ellipses || !finished)
+ obstack_grow_wstr (&wchar_buf, LCST ("..."));
+
+ /* OUTPUT is where we collect `char's for printing. */
+ obstack_init (&output);
+ make_cleanup_obstack_free (&output);
+
+ convert_between_encodings (INTERMEDIATE_ENCODING, host_charset (),
+ obstack_base (&wchar_buf),
+ obstack_object_size (&wchar_buf),
+ 1, &output, translit_char);
+ obstack_1grow (&output, '\0');
+
+ fputs_filtered (obstack_base (&output), stream);
+
+ do_cleanups (cleanup);
}
+
+/* Obtain a C string from the inferior storing it in a newly allocated
+ buffer in BUFFER, which should be freed by the caller. If the
+ in- and out-parameter *LENGTH is specified at -1, the string is read
+ until a null character of the appropriate width is found, otherwise
+ the string is read to the length of characters specified.
+ The size of a character is determined by the length of the target
+ type of the pointer or array. If VALUE is an array with a known
+ length, the function will not read past the end of the array.
+ On completion, *LENGTH will be set to the size of the string read in
+ characters. (If a length of -1 is specified, the length returned
+ will not include the null character). CHARSET is always set to the
+ target charset. */
+
+void
+c_get_string (struct value *value, gdb_byte **buffer, int *length,
+ struct type **char_type, const char **charset)
+{
+ int err, width;
+ unsigned int fetchlimit;
+ struct type *type = check_typedef (value_type (value));
+ struct type *element_type = TYPE_TARGET_TYPE (type);
+ int req_length = *length;
+ enum bfd_endian byte_order = gdbarch_byte_order (get_type_arch (type));
+ enum c_string_type kind;
+
+ if (element_type == NULL)
+ goto error;
+
+ if (TYPE_CODE (type) == TYPE_CODE_ARRAY)
+ {
+ /* If we know the size of the array, we can use it as a limit on the
+ number of characters to be fetched. */
+ if (TYPE_NFIELDS (type) == 1
+ && TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_RANGE)
+ {
+ LONGEST low_bound, high_bound;
+
+ get_discrete_bounds (TYPE_FIELD_TYPE (type, 0),
+ &low_bound, &high_bound);
+ fetchlimit = high_bound - low_bound + 1;
+ }
+ else
+ fetchlimit = UINT_MAX;
+ }
+ else if (TYPE_CODE (type) == TYPE_CODE_PTR)
+ fetchlimit = UINT_MAX;
+ else
+ /* We work only with arrays and pointers. */
+ goto error;
+
+ if (! c_textual_element_type (element_type, 0))
+ goto error;
+ kind = classify_type (element_type,
+ get_type_arch (element_type),
+ charset);
+ width = TYPE_LENGTH (element_type);
+
+ /* If the string lives in GDB's memory instead of the inferior's, then we
+ just need to copy it to BUFFER. Also, since such strings are arrays
+ with known size, FETCHLIMIT will hold the size of the array. */
+ if ((VALUE_LVAL (value) == not_lval
+ || VALUE_LVAL (value) == lval_internalvar)
+ && fetchlimit != UINT_MAX)
+ {
+ int i;
+ const gdb_byte *contents = value_contents (value);
+
+ /* If a length is specified, use that. */
+ if (*length >= 0)
+ i = *length;
+ else
+ /* Otherwise, look for a null character. */
+ for (i = 0; i < fetchlimit; i++)
+ if (extract_unsigned_integer (contents + i * width, width,
+ byte_order) == 0)
+ break;
+
+ /* I is now either a user-defined length, the number of non-null
+ characters, or FETCHLIMIT. */
+ *length = i * width;
+ *buffer = xmalloc (*length);
+ memcpy (*buffer, contents, *length);
+ err = 0;
+ }
+ else
+ {
+ err = read_string (value_as_address (value), *length, width, fetchlimit,
+ byte_order, buffer, length);
+ if (err)
+ {
+ xfree (*buffer);
+ error (_("Error reading string from inferior: %s"),
+ safe_strerror (err));
+ }
+ }
+
+ /* If the LENGTH is specified at -1, we want to return the string
+ length up to the terminating null character. If an actual length
+ was specified, we want to return the length of exactly what was
+ read. */
+ if (req_length == -1)
+ /* If the last character is null, subtract it from LENGTH. */
+ if (*length > 0
+ && extract_unsigned_integer (*buffer + *length - width, width,
+ byte_order) == 0)
+ *length -= width;
+
+ /* The read_string function will return the number of bytes read.
+ If length returned from read_string was > 0, return the number of
+ characters read by dividing the number of bytes by width. */
+ if (*length != 0)
+ *length = *length / width;
+
+ *char_type = element_type;
+
+ return;
+
+ error:
+ {
+ char *type_str;
+
+ type_str = type_to_string (type);
+ if (type_str)
+ {
+ make_cleanup (xfree, type_str);
+ error (_("Trying to read string with inappropriate type `%s'."),
+ type_str);
+ }
+ else
+ error (_("Trying to read string with inappropriate type."));
+ }
+}
+
\f
-/* Preprocessing and parsing C and C++ expressions. */
+/* Evaluating C and C++ expressions. */
+
+/* Convert a UCN. The digits of the UCN start at P and extend no
+ farther than LIMIT. DEST_CHARSET is the name of the character set
+ into which the UCN should be converted. The results are written to
+ OUTPUT. LENGTH is the maximum length of the UCN, either 4 or 8.
+ Returns a pointer to just after the final digit of the UCN. */
+static char *
+convert_ucn (char *p, char *limit, const char *dest_charset,
+ struct obstack *output, int length)
+{
+ unsigned long result = 0;
+ gdb_byte data[4];
+ int i;
-/* When we find that lexptr (the global var defined in parse.c) is
- pointing at a macro invocation, we expand the invocation, and call
- scan_macro_expansion to save the old lexptr here and point lexptr
- into the expanded text. When we reach the end of that, we call
- end_macro_expansion to pop back to the value we saved here. The
- macro expansion code promises to return only fully-expanded text,
- so we don't need to "push" more than one level.
+ for (i = 0; i < length && p < limit && isxdigit (*p); ++i, ++p)
+ result = (result << 4) + host_hex_value (*p);
- This is disgusting, of course. It would be cleaner to do all macro
- expansion beforehand, and then hand that to lexptr. But we don't
- really know where the expression ends. Remember, in a command like
+ for (i = 3; i >= 0; --i)
+ {
+ data[i] = result & 0xff;
+ result >>= 8;
+ }
- (gdb) break *ADDRESS if CONDITION
+ convert_between_encodings ("UTF-32BE", dest_charset, data, 4, 4, output,
+ translit_none);
- we evaluate ADDRESS in the scope of the current frame, but we
- evaluate CONDITION in the scope of the breakpoint's location. So
- it's simply wrong to try to macro-expand the whole thing at once. */
-static char *macro_original_text;
-static char *macro_expanded_text;
+ return p;
+}
+/* Emit a character, VALUE, which was specified numerically, to
+ OUTPUT. TYPE is the target character type. */
-void
-scan_macro_expansion (char *expansion)
+static void
+emit_numeric_character (struct type *type, unsigned long value,
+ struct obstack *output)
+{
+ gdb_byte *buffer;
+
+ buffer = alloca (TYPE_LENGTH (type));
+ pack_long (buffer, type, value);
+ obstack_grow (output, buffer, TYPE_LENGTH (type));
+}
+
+/* Convert an octal escape sequence. TYPE is the target character
+ type. The digits of the escape sequence begin at P and extend no
+ farther than LIMIT. The result is written to OUTPUT. Returns a
+ pointer to just after the final digit of the escape sequence. */
+
+static char *
+convert_octal (struct type *type, char *p, char *limit, struct obstack *output)
{
- /* We'd better not be trying to push the stack twice. */
- gdb_assert (! macro_original_text);
- gdb_assert (! macro_expanded_text);
+ int i;
+ unsigned long value = 0;
+
+ for (i = 0;
+ i < 3 && p < limit && isdigit (*p) && *p != '8' && *p != '9';
+ ++i)
+ {
+ value = 8 * value + host_hex_value (*p);
+ ++p;
+ }
- /* Save the old lexptr value, so we can return to it when we're done
- parsing the expanded text. */
- macro_original_text = lexptr;
- lexptr = expansion;
+ emit_numeric_character (type, value, output);
- /* Save the expanded text, so we can free it when we're finished. */
- macro_expanded_text = expansion;
+ return p;
}
+/* Convert a hex escape sequence. TYPE is the target character type.
+ The digits of the escape sequence begin at P and extend no farther
+ than LIMIT. The result is written to OUTPUT. Returns a pointer to
+ just after the final digit of the escape sequence. */
-int
-scanning_macro_expansion (void)
+static char *
+convert_hex (struct type *type, char *p, char *limit, struct obstack *output)
{
- return macro_original_text != 0;
+ unsigned long value = 0;
+
+ while (p < limit && isxdigit (*p))
+ {
+ value = 16 * value + host_hex_value (*p);
+ ++p;
+ }
+
+ emit_numeric_character (type, value, output);
+
+ return p;
}
+#define ADVANCE \
+ do { \
+ ++p; \
+ if (p == limit) \
+ error (_("Malformed escape sequence")); \
+ } while (0)
+
+/* Convert an escape sequence to a target format. TYPE is the target
+ character type to use, and DEST_CHARSET is the name of the target
+ character set. The backslash of the escape sequence is at *P, and
+ the escape sequence will not extend past LIMIT. The results are
+ written to OUTPUT. Returns a pointer to just past the final
+ character of the escape sequence. */
-void
-finished_macro_expansion (void)
+static char *
+convert_escape (struct type *type, const char *dest_charset,
+ char *p, char *limit, struct obstack *output)
{
- /* There'd better be something to pop back to, and we better have
- saved a pointer to the start of the expanded text. */
- gdb_assert (macro_original_text);
- gdb_assert (macro_expanded_text);
-
- /* Pop back to the original text. */
- lexptr = macro_original_text;
- macro_original_text = 0;
-
- /* Free the expanded text. */
- xfree (macro_expanded_text);
- macro_expanded_text = 0;
+ /* Skip the backslash. */
+ ADVANCE;
+
+ switch (*p)
+ {
+ case '\\':
+ obstack_1grow (output, '\\');
+ ++p;
+ break;
+
+ case 'x':
+ ADVANCE;
+ if (!isxdigit (*p))
+ error (_("\\x used with no following hex digits."));
+ p = convert_hex (type, p, limit, output);
+ break;
+
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ p = convert_octal (type, p, limit, output);
+ break;
+
+ case 'u':
+ case 'U':
+ {
+ int length = *p == 'u' ? 4 : 8;
+
+ ADVANCE;
+ if (!isxdigit (*p))
+ error (_("\\u used with no following hex digits"));
+ p = convert_ucn (p, limit, dest_charset, output, length);
+ }
+ }
+
+ return p;
}
+/* Given a single string from a (C-specific) OP_STRING list, convert
+ it to a target string, handling escape sequences specially. The
+ output is written to OUTPUT. DATA is the input string, which has
+ length LEN. DEST_CHARSET is the name of the target character set,
+ and TYPE is the type of target character to use. */
static void
-scan_macro_cleanup (void *dummy)
+parse_one_string (struct obstack *output, char *data, int len,
+ const char *dest_charset, struct type *type)
{
- if (macro_original_text)
- finished_macro_expansion ();
-}
+ char *limit;
+ limit = data + len;
-/* We set these global variables before calling c_parse, to tell it
- how it to find macro definitions for the expression at hand. */
-macro_lookup_ftype *expression_macro_lookup_func;
-void *expression_macro_lookup_baton;
+ while (data < limit)
+ {
+ char *p = data;
+ /* Look for next escape, or the end of the input. */
+ while (p < limit && *p != '\\')
+ ++p;
+ /* If we saw a run of characters, convert them all. */
+ if (p > data)
+ convert_between_encodings (host_charset (), dest_charset,
+ data, p - data, 1, output, translit_none);
+ /* If we saw an escape, convert it. */
+ if (p < limit)
+ p = convert_escape (type, dest_charset, p, limit, output);
+ data = p;
+ }
+}
-static int
-c_preprocess_and_parse (void)
+/* Expression evaluator for the C language family. Most operations
+ are delegated to evaluate_subexp_standard; see that function for a
+ description of the arguments. */
+
+static struct value *
+evaluate_subexp_c (struct type *expect_type, struct expression *exp,
+ int *pos, enum noside noside)
{
- /* Set up a lookup function for the macro expander. */
- struct macro_scope *scope = 0;
- struct cleanup *back_to = make_cleanup (free_current_contents, &scope);
+ enum exp_opcode op = exp->elts[*pos].opcode;
- if (expression_context_block)
- scope = sal_macro_scope (find_pc_line (expression_context_pc, 0));
- else
- scope = default_macro_scope ();
- if (! scope)
- scope = user_macro_scope ();
+ switch (op)
+ {
+ case OP_STRING:
+ {
+ int oplen, limit;
+ struct type *type;
+ struct obstack output;
+ struct cleanup *cleanup;
+ struct value *result;
+ enum c_string_type dest_type;
+ const char *dest_charset;
- expression_macro_lookup_func = standard_macro_lookup;
- expression_macro_lookup_baton = (void *) scope;
+ obstack_init (&output);
+ cleanup = make_cleanup_obstack_free (&output);
- gdb_assert (! macro_original_text);
- make_cleanup (scan_macro_cleanup, 0);
+ ++*pos;
+ oplen = longest_to_int (exp->elts[*pos].longconst);
- {
- int result = c_parse ();
- do_cleanups (back_to);
- return result;
- }
+ ++*pos;
+ limit = *pos + BYTES_TO_EXP_ELEM (oplen + 1);
+ dest_type
+ = (enum c_string_type) longest_to_int (exp->elts[*pos].longconst);
+ switch (dest_type & ~C_CHAR)
+ {
+ case C_STRING:
+ type = language_string_char_type (exp->language_defn,
+ exp->gdbarch);
+ break;
+ case C_WIDE_STRING:
+ type = lookup_typename (exp->language_defn, exp->gdbarch,
+ "wchar_t", NULL, 0);
+ break;
+ case C_STRING_16:
+ type = lookup_typename (exp->language_defn, exp->gdbarch,
+ "char16_t", NULL, 0);
+ break;
+ case C_STRING_32:
+ type = lookup_typename (exp->language_defn, exp->gdbarch,
+ "char32_t", NULL, 0);
+ break;
+ default:
+ internal_error (__FILE__, __LINE__, "unhandled c_string_type");
+ }
+
+ /* Ensure TYPE_LENGTH is valid for TYPE. */
+ check_typedef (type);
+
+ dest_charset = charset_for_string_type (dest_type, exp->gdbarch);
+
+ ++*pos;
+ while (*pos < limit)
+ {
+ int len;
+
+ len = longest_to_int (exp->elts[*pos].longconst);
+
+ ++*pos;
+ if (noside != EVAL_SKIP)
+ parse_one_string (&output, &exp->elts[*pos].string, len,
+ dest_charset, type);
+ *pos += BYTES_TO_EXP_ELEM (len);
+ }
+
+ /* Skip the trailing length and opcode. */
+ *pos += 2;
+
+ if (noside == EVAL_SKIP)
+ {
+ /* Return a dummy value of the appropriate type. */
+ if ((dest_type & C_CHAR) != 0)
+ result = allocate_value (type);
+ else
+ result = value_cstring ("", 0, type);
+ do_cleanups (cleanup);
+ return result;
+ }
+
+ if ((dest_type & C_CHAR) != 0)
+ {
+ LONGEST value;
+
+ if (obstack_object_size (&output) != TYPE_LENGTH (type))
+ error (_("Could not convert character constant to target character set"));
+ value = unpack_long (type, obstack_base (&output));
+ result = value_from_longest (type, value);
+ }
+ else
+ {
+ int i;
+
+ /* Write the terminating character. */
+ for (i = 0; i < TYPE_LENGTH (type); ++i)
+ obstack_1grow (&output, 0);
+ result = value_cstring (obstack_base (&output),
+ obstack_object_size (&output),
+ type);
+ }
+ do_cleanups (cleanup);
+ return result;
+ }
+ break;
+
+ default:
+ break;
+ }
+ return evaluate_subexp_standard (expect_type, exp, pos, noside);
}
struct language_arch_info *lai)
{
const struct builtin_type *builtin = builtin_type (gdbarch);
+
lai->string_char_type = builtin->builtin_char;
lai->primitive_type_vector
= GDBARCH_OBSTACK_CALLOC (gdbarch, nr_c_primitive_types + 1,
lai->bool_type_default = builtin->builtin_int;
}
+const struct exp_descriptor exp_descriptor_c =
+{
+ print_subexp_standard,
+ operator_length_standard,
+ operator_check_standard,
+ op_name_standard,
+ dump_subexp_body_standard,
+ evaluate_subexp_c
+};
+
const struct language_defn c_language_defn =
{
"c", /* Language name */
case_sensitive_on,
array_row_major,
macro_expansion_c,
- &exp_descriptor_standard,
- c_preprocess_and_parse,
+ &exp_descriptor_c,
+ c_parse,
c_error,
null_post_parser,
c_printchar, /* Print a character constant */
c_language_arch_info,
default_print_array_index,
default_pass_by_reference,
+ c_get_string,
LANG_MAGIC
};
struct language_arch_info *lai)
{
const struct builtin_type *builtin = builtin_type (gdbarch);
+
lai->string_char_type = builtin->builtin_char;
lai->primitive_type_vector
= GDBARCH_OBSTACK_CALLOC (gdbarch, nr_cplus_primitive_types + 1,
case_sensitive_on,
array_row_major,
macro_expansion_c,
- &exp_descriptor_standard,
- c_preprocess_and_parse,
+ &exp_descriptor_c,
+ c_parse,
c_error,
null_post_parser,
c_printchar, /* Print a character constant */
cplus_language_arch_info,
default_print_array_index,
cp_pass_by_reference,
+ c_get_string,
LANG_MAGIC
};
case_sensitive_on,
array_row_major,
macro_expansion_c,
- &exp_descriptor_standard,
- c_preprocess_and_parse,
+ &exp_descriptor_c,
+ c_parse,
c_error,
null_post_parser,
c_printchar, /* Print a character constant */
c_language_arch_info, /* FIXME: la_language_arch_info. */
default_print_array_index,
default_pass_by_reference,
+ c_get_string,
LANG_MAGIC
};
case_sensitive_on,
array_row_major,
macro_expansion_c,
- &exp_descriptor_standard,
- c_preprocess_and_parse,
+ &exp_descriptor_c,
+ c_parse,
c_error,
null_post_parser,
c_printchar, /* Print a character constant */
c_language_arch_info,
default_print_array_index,
default_pass_by_reference,
+ c_get_string,
LANG_MAGIC
};