#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>
+#include <limits.h>
#include "calendar.h"
#include "identifier.h"
#include "settings.h"
#include "value.h"
+#include "format.h"
#include <libpspp/assertion.h>
+#include <libpspp/legacy-encoding.h>
#include <libpspp/compiler.h>
#include <libpspp/integer-format.h>
#include <libpspp/message.h>
#include <libpspp/misc.h>
#include <libpspp/str.h>
-
#include "c-ctype.h"
+#include "c-strtod.h"
#include "minmax.h"
-#include "size_max.h"
#include "xalloc.h"
#include "gettext.h"
/* Information about parsing one data field. */
struct data_in
{
+ enum legacy_encoding encoding;/* Encoding of source. */
struct substring input; /* Source. */
enum fmt_type format; /* Input format. */
int implied_decimals; /* Number of implied decimal places. */
int last_column; /* Last column. */
};
-/* Integer format used for IB and PIB input. */
-static enum integer_format input_integer_format = INTEGER_NATIVE;
-/* Floating-point format used for RB and RBHEX input. */
-static enum float_format input_float_format = FLOAT_NATIVE_DOUBLE;
typedef bool data_in_parser_func (struct data_in *);
#define FMT(NAME, METHOD, IMIN, OMIN, IO, CATEGORY) \
static int hexit_value (int c);
\f
-/* Parses the characters in INPUT according to FORMAT. Stores
- the parsed representation in OUTPUT, which has the given WIDTH
- (0 for a numeric field, otherwise the string width).
+/* Parses the characters in INPUT, which are encoded in the given
+ ENCODING, according to FORMAT. Stores the parsed
+ representation in OUTPUT, which has the given WIDTH (0 for
+ a numeric field, otherwise the string width).
If no decimal point is included in a numeric format, then
IMPLIED_DECIMALS decimal places are implied. Specify 0 if no
decimal places should be implied.
- If FIRST_COLUMN is nonzero, then it should be the 1-based
- column number of the first character in INPUT, used in error
- messages. */
+ If FIRST_COLUMN and LAST_COLUMN are nonzero, then they should
+ be the 1-based column number of the first and
+ one-past-the-last-character in INPUT, for use in error
+ messages. (LAST_COLUMN cannot always be calculated from
+ FIRST_COLUMN plus the length of the input because of the
+ possibility of escaped quotes in strings, etc.) */
bool
-data_in (struct substring input,
+data_in (struct substring input, enum legacy_encoding encoding,
enum fmt_type format, int implied_decimals,
- int first_column, union value *output, int width)
+ int first_column, int last_column, union value *output, int width)
{
static data_in_parser_func *const handlers[FMT_NUMBER_OF_FORMATS] =
{
};
struct data_in i;
+ void *copy = NULL;
bool ok;
assert ((width != 0) == fmt_is_string (format));
- i.input = input;
+ if (encoding == LEGACY_NATIVE
+ || fmt_get_category (format) & (FMT_CAT_BINARY | FMT_CAT_STRING))
+ {
+ i.input = input;
+ i.encoding = encoding;
+ }
+ else
+ {
+ ss_alloc_uninit (&i.input, ss_length (input));
+ legacy_recode (encoding, ss_data (input), LEGACY_NATIVE,
+ ss_data (i.input), ss_length (input));
+ i.encoding = LEGACY_NATIVE;
+ copy = ss_data (i.input);
+ }
i.format = format;
i.implied_decimals = implied_decimals;
i.width = width;
i.first_column = first_column;
- i.last_column = first_column + ss_length (input) - 1;
+ i.last_column = last_column;
if (!ss_is_empty (i.input))
{
ok = true;
}
- return ok;
-}
-
-/* Returns the integer format used for IB and PIB input. */
-enum integer_format
-data_in_get_integer_format (void)
-{
- return input_integer_format;
-}
-
-/* Sets the integer format used for IB and PIB input to
- FORMAT. */
-void
-data_in_set_integer_format (enum integer_format format)
-{
- input_integer_format = format;
-}
+ if (copy)
+ free (copy);
-/* Returns the floating-point format used for RB and RBHEX
- input. */
-enum float_format
-data_in_get_float_format (void)
-{
- return input_float_format;
+ return ok;
}
-/* Sets the floating-point format used for RB and RBHEX input to
- FORMAT. */
-void
-data_in_set_float_format (enum float_format format)
-{
- input_float_format = format;
-}
\f
/* Format parsers. */
static bool
parse_number (struct data_in *i)
{
- const struct fmt_number_style *style = fmt_get_style (i->format);
+ const struct fmt_number_style *style =
+ settings_get_style (i->format);
struct string tmp;
int save_errno;
char *tail;
- assert (fmt_get_category (i->format) != FMT_CAT_CUSTOM);
+ if (fmt_get_category (i->format) == FMT_CAT_CUSTOM)
+ {
+ style = settings_get_style (FMT_F);
+ }
/* Trim spaces and check for missing value representation. */
if (trim_spaces_and_check_missing (i))
return false;
}
- /* Let strtod() do the conversion. */
+ /* Let c_strtod() do the conversion. */
save_errno = errno;
errno = 0;
- i->output->f = strtod (ds_cstr (&tmp), &tail);
+ i->output->f = c_strtod (ds_cstr (&tmp), &tail);
if (*tail != '\0')
{
data_warning (i, _("Invalid numeric syntax."));
return false;
}
- /* Let strtod() do the conversion. */
+ /* Let c_strtod() do the conversion. */
save_errno = errno;
errno = 0;
- i->output->f = strtod (ds_cstr (&tmp), NULL);
+ i->output->f = c_strtod (ds_cstr (&tmp), NULL);
if (errno == ERANGE)
{
if (fabs (i->output->f) > 1)
uint64_t sign_bit;
bytes = MIN (8, ss_length (i->input));
- value = integer_get (input_integer_format, ss_data (i->input), bytes);
+ value = integer_get (settings_get_input_integer_format (), ss_data (i->input), bytes);
sign_bit = UINT64_C(1) << (8 * bytes - 1);
if (!(value & sign_bit))
static bool
parse_PIB (struct data_in *i)
{
- i->output->f = integer_get (input_integer_format, ss_data (i->input),
+ i->output->f = integer_get (settings_get_input_integer_format (), ss_data (i->input),
MIN (8, ss_length (i->input)));
apply_implied_decimals (i);
static bool
parse_RB (struct data_in *i)
{
- size_t size = float_get_size (input_float_format);
+ enum float_format ff = settings_get_input_float_format ();
+ size_t size = float_get_size (ff);
if (ss_length (i->input) >= size)
- float_convert (input_float_format, ss_data (i->input),
+ float_convert (ff, ss_data (i->input),
FLOAT_NATIVE_DOUBLE, &i->output->f);
else
i->output->f = SYSMIS;
static bool
parse_A (struct data_in *i)
{
- buf_copy_rpad (i->output->s, i->width,
- ss_data (i->input), ss_length (i->input));
+ /* This is equivalent to buf_copy_rpad, except that we posibly
+ do a character set recoding in the middle. */
+ char *dst = i->output->s;
+ size_t dst_size = i->width;
+ const char *src = ss_data (i->input);
+ size_t src_size = ss_length (i->input);
+
+ legacy_recode (i->encoding, src, LEGACY_NATIVE, dst, MIN (src_size, dst_size));
+ if (dst_size > src_size)
+ memset (&dst[src_size], ' ', dst_size - src_size);
+
return true;
}
return false;
}
+ if (i->encoding != LEGACY_NATIVE)
+ {
+ hi = legacy_to_native (i->encoding, hi);
+ lo = legacy_to_native (i->encoding, lo);
+ }
if (!c_isxdigit (hi) || !c_isxdigit (lo))
{
data_warning (i, _("Field must contain only hex digits."));
exact matches (except for case) are allowed.
Returns true if successful, false otherwise. */
static bool
-match_name (struct substring token, const char **names, long *output)
+match_name (struct substring token, const char *const *names, long *output)
{
int i;
}
else
{
- static const char *english_names[] =
+ static const char *const english_names[] =
{
"jan", "feb", "mar", "apr", "may", "jun",
"jul", "aug", "sep", "oct", "nov", "dec",
NULL,
};
- static const char *roman_names[] =
+ static const char *const roman_names[] =
{
"i", "ii", "iii", "iv", "v", "vi",
"vii", "viii", "ix", "x", "xi", "xii",
if (*year >= 0 && *year <= 99)
{
- int epoch = get_epoch ();
+ int epoch = settings_get_epoch ();
int epoch_century = ROUND_DOWN (epoch, 100);
int epoch_offset = epoch - epoch_century;
if (*year >= epoch_offset)
cp = buf;
while (c_isdigit (ss_first (i->input)))
*cp++ = ss_get_char (&i->input);
- if (ss_match_char (&i->input, fmt_decimal_char (FMT_F)))
+ if (ss_match_char (&i->input, settings_get_decimal_char (FMT_F)))
*cp++ = '.';
while (c_isdigit (ss_first (i->input)))
*cp++ = ss_get_char (&i->input);
static bool
parse_weekday (struct data_in *i, long *weekday)
{
- static const char *weekday_names[] =
+ static const char *const weekday_names[] =
{
"su", "mo", "tu", "we", "th", "fr", "sa",
NULL,
ds_put_char (&text, '(');
if (i->first_column != 0)
{
- if (i->first_column == i->last_column)
+ if (i->first_column == i->last_column - 1)
ds_put_format (&text, _("column %d"), i->first_column);
else
ds_put_format (&text, _("columns %d-%d"),
- i->first_column, i->last_column);
+ i->first_column, i->last_column - 1);
ds_put_cstr (&text, ", ");
}
ds_put_format (&text, _("%s field) "), fmt_name (i->format));
if (fmt_is_string (i->format))
memset (i->output->s, ' ', i->width);
else
- i->output->f = get_blanks ();
+ i->output->f = settings_get_blanks ();
}
/* Trims leading and trailing spaces from I.