X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fdata-io%2Fplacement-parser.c;h=ca6f491ee0b7727002523ddf451a18625481c5e2;hb=63ae14f7f555727026138e1668e6cf91a9046bc0;hp=3eb8534f217ac9d99f0aa7afa30e71924d4e3ed4;hpb=f5c108becd49d78f4898cab11352291f5689d24e;p=pspp
diff --git a/src/language/data-io/placement-parser.c b/src/language/data-io/placement-parser.c
index 3eb8534f21..ca6f491ee0 100644
--- a/src/language/data-io/placement-parser.c
+++ b/src/language/data-io/placement-parser.c
@@ -1,35 +1,35 @@
-/* PSPP - computes sample statistics.
- Copyright (C) 2006 Free Software Foundation, Inc.
+/* PSPP - a program for statistical analysis.
+ Copyright (C) 2006, 2010, 2011, 2012 Free Software Foundation, Inc.
- This program 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 of the
- License, or (at your option) any later version.
+ This program 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 3 of the License, or
+ (at your option) any later version.
- This program is distributed in the hope that it will be useful, but
- WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- General Public License for more details.
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
- 02110-1301, USA. */
+ along with this program. If not, see . */
#include
-#include
+#include "language/data-io/placement-parser.h"
#include
-#include
-#include
-#include
-#include
-#include
+#include "data/format.h"
+#include "language/lexer/format-parser.h"
+#include "language/lexer/lexer.h"
+#include "libpspp/message.h"
+#include "libpspp/pool.h"
+#include "libpspp/str.h"
-#include "xalloc.h"
-#include "xsize.h"
+#include "gl/c-strcase.h"
+#include "gl/xalloc.h"
+#include "gl/xsize.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
@@ -43,9 +43,9 @@ enum
PRS_TYPE_NEW_REC /* Next record. */
};
-static bool fixed_parse_columns (struct lexer *, struct pool *, size_t var_cnt, bool for_input,
- struct fmt_spec **, size_t *);
-static bool fixed_parse_fortran (struct lexer *l, struct pool *, bool for_input,
+static bool fixed_parse_columns (struct lexer *, struct pool *, size_t var_cnt,
+ enum fmt_use, struct fmt_spec **, size_t *);
+static bool fixed_parse_fortran (struct lexer *l, struct pool *, enum fmt_use,
struct fmt_spec **, size_t *);
/* Parses Fortran-like or column-based specifications for placing
@@ -53,6 +53,13 @@ static bool fixed_parse_fortran (struct lexer *l, struct pool *, bool for_input,
formats like those parsed by DATA LIST or PRINT. Returns true
only if successful.
+ The formats parsed are either input or output formats, according
+ to USE.
+
+ If USE is FMT_FOR_INPUT, then T, X, and / "formats" are parsed,
+ in addition to regular formats. If USE is FMT_FOR_OUTPUT, then
+ T and X "formats" are parsed but not /.
+
If successful, formats for VAR_CNT variables are stored in
*FORMATS, and the number of formats required is stored in
*FORMAT_CNT. *FORMAT_CNT may be greater than VAR_CNT because
@@ -64,18 +71,20 @@ static bool fixed_parse_fortran (struct lexer *l, struct pool *, bool for_input,
Uses POOL for allocation. When the caller is finished
interpreting *FORMATS, POOL may be destroyed. */
bool
-parse_var_placements (struct lexer *lexer, struct pool *pool, size_t var_cnt, bool for_input,
+parse_var_placements (struct lexer *lexer, struct pool *pool, size_t var_cnt,
+ enum fmt_use use,
struct fmt_spec **formats, size_t *format_cnt)
{
assert (var_cnt > 0);
if (lex_is_number (lexer))
- return fixed_parse_columns (lexer, pool, var_cnt, for_input, formats, format_cnt);
- else if (lex_match (lexer, '('))
+ return fixed_parse_columns (lexer, pool, var_cnt, use,
+ formats, format_cnt);
+ else if (lex_match (lexer, T_LPAREN))
{
size_t assignment_cnt;
size_t i;
- if (!fixed_parse_fortran (lexer, pool, for_input, formats, format_cnt))
+ if (!fixed_parse_fortran (lexer, pool, use, formats, format_cnt))
return false;
assignment_cnt = 0;
@@ -84,9 +93,9 @@ parse_var_placements (struct lexer *lexer, struct pool *pool, size_t var_cnt, bo
if (assignment_cnt != var_cnt)
{
- msg (SE, _("Number of variables specified (%d) "
- "differs from number of variable formats (%d)."),
- (int) var_cnt, (int) assignment_cnt);
+ msg (SE, _("Number of variables specified (%zu) "
+ "differs from number of variable formats (%zu)."),
+ var_cnt, assignment_cnt);
return false;
}
@@ -102,14 +111,15 @@ parse_var_placements (struct lexer *lexer, struct pool *pool, size_t var_cnt, bo
/* Implements parse_var_placements for column-based formats. */
static bool
-fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt, bool for_input,
+fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt,
+ enum fmt_use use,
struct fmt_spec **formats, size_t *format_cnt)
{
struct fmt_spec format;
int fc, lc;
size_t i;
- if ( !parse_column_range (lexer, &fc, &lc, NULL) )
+ if ( !parse_column_range (lexer, 1, &fc, &lc, NULL) )
return false;
/* Divide columns evenly. */
@@ -117,20 +127,20 @@ fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt, boo
if ((lc - fc + 1) % var_cnt)
{
msg (SE, _("The %d columns %d-%d "
- "can't be evenly divided into %u fields."),
- lc - fc + 1, fc, lc, (unsigned int) var_cnt);
+ "can't be evenly divided into %zu fields."),
+ lc - fc + 1, fc, lc, var_cnt);
return false;
}
/* Format specifier. */
- if (lex_match (lexer, '('))
+ if (lex_match (lexer, T_LPAREN))
{
/* Get format type. */
if (lex_token (lexer) == T_ID)
{
if (!parse_format_specifier_name (lexer, &format.type))
return false;
- lex_match (lexer, ',');
+ lex_match (lexer, T_COMMA);
}
else
format.type = FMT_F;
@@ -144,7 +154,7 @@ fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt, boo
else
format.d = 0;
- if (!lex_force_match (lexer, ')'))
+ if (!lex_force_match (lexer, T_RPAREN))
return false;
}
else
@@ -152,7 +162,7 @@ fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt, boo
format.type = FMT_F;
format.d = 0;
}
- if (!fmt_check (&format, for_input))
+ if (!fmt_check (&format, use))
return false;
*formats = pool_nalloc (pool, var_cnt + 1, sizeof **formats);
@@ -166,14 +176,14 @@ fixed_parse_columns (struct lexer *lexer, struct pool *pool, size_t var_cnt, boo
/* Implements parse_var_placements for Fortran-like formats. */
static bool
-fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
+fixed_parse_fortran (struct lexer *lexer, struct pool *pool, enum fmt_use use,
struct fmt_spec **formats, size_t *format_cnt)
{
size_t formats_allocated = 0;
size_t formats_used = 0;
*formats = NULL;
- while (!lex_match (lexer, ')'))
+ while (!lex_match (lexer, T_RPAREN))
{
struct fmt_spec f;
struct fmt_spec *new_formats;
@@ -191,10 +201,10 @@ fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
count = 1;
/* Parse format specifier. */
- if (lex_match (lexer, '('))
+ if (lex_match (lexer, T_LPAREN))
{
/* Call ourselves recursively to handle parentheses. */
- if (!fixed_parse_fortran (lexer, pool, for_input,
+ if (!fixed_parse_fortran (lexer, pool, use,
&new_formats, &new_format_cnt))
return false;
}
@@ -202,7 +212,7 @@ fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
{
new_formats = &f;
new_format_cnt = 1;
- if (lex_match (lexer, '/'))
+ if (use == FMT_FOR_INPUT && lex_match (lexer, T_SLASH))
f.type = PRS_TYPE_NEW_REC;
else
{
@@ -211,9 +221,9 @@ fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
if (!parse_abstract_format_specifier (lexer, type, &f.w, &f.d))
return false;
- if (!strcasecmp (type, "T"))
+ if (!c_strcasecmp (type, "T"))
f.type = PRS_TYPE_T;
- else if (!strcasecmp (type, "X"))
+ else if (!c_strcasecmp (type, "X"))
{
f.type = PRS_TYPE_X;
f.w = count;
@@ -223,10 +233,10 @@ fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
{
if (!fmt_from_name (type, &f.type))
{
- msg (SE, _("Unknown format type \"%s\"."), type);
+ msg (SE, _("Unknown format type `%s'."), type);
return false;
}
- if (!fmt_check (&f, for_input))
+ if (!fmt_check (&f, use))
return false;
}
}
@@ -253,7 +263,7 @@ fixed_parse_fortran (struct lexer *lexer, struct pool *pool, bool for_input,
formats_used += new_format_cnt;
}
- lex_match (lexer, ',');
+ lex_match (lexer, T_COMMA);
}
*format_cnt = formats_used;
@@ -268,7 +278,7 @@ bool
execute_placement_format (const struct fmt_spec *format,
int *record, int *column)
{
- switch (format->type)
+ switch ((int) format->type)
{
case PRS_TYPE_X:
*column += format->w;
@@ -289,42 +299,75 @@ execute_placement_format (const struct fmt_spec *format,
}
}
+static bool
+parse_column__ (int value, int base, int *column)
+{
+ assert (base == 0 || base == 1);
+ *column = value - base + 1;
+ if (*column < 1)
+ {
+ if (base == 1)
+ msg (SE, _("Column positions for fields must be positive."));
+ else
+ msg (SE, _("Column positions for fields must not be negative."));
+ return false;
+ }
+ return true;
+}
+
+/* Parses a BASE-based column using LEXER. Returns true and
+ stores a 1-based column number into *COLUMN if successful,
+ otherwise emits an error message and returns false.
+
+ If BASE is 0, zero-based column numbers are parsed; if BASE is
+ 1, 1-based column numbers are parsed. Regardless of BASE, the
+ values stored in *FIRST_COLUMN and *LAST_COLUMN are
+ 1-based. */
+bool
+parse_column (struct lexer *lexer, int base, int *column)
+{
+ assert (base == 0 || base == 1);
+
+ if (!lex_force_int (lexer)
+ || !parse_column__ (lex_integer (lexer), base, column))
+ return false;
+
+ lex_get (lexer);
+ return true;
+}
+
/* Parse a column or a range of columns, specified as a single
- integer or two integer delimited by a dash. Stores the range
+ integer or two integers delimited by a dash. Stores the range
in *FIRST_COLUMN and *LAST_COLUMN. (If only a single integer
is given, it is stored in both.) If RANGE_SPECIFIED is
non-null, then *RANGE_SPECIFIED is set to true if the syntax
contained a dash, false otherwise. Returns true if
successful, false if the syntax was invalid or the values
- specified did not make sense. */
+ specified did not make sense.
+
+ If BASE is 0, zero-based column numbers are parsed; if BASE is
+ 1, 1-based column numbers are parsed. Regardless of BASE, the
+ values stored in *FIRST_COLUMN and *LAST_COLUMN are
+ 1-based. */
bool
-parse_column_range (struct lexer *lexer, int *first_column, int *last_column,
+parse_column_range (struct lexer *lexer, int base,
+ int *first_column, int *last_column,
bool *range_specified)
{
/* First column. */
- if (!lex_force_int (lexer))
+ if (!lex_force_int (lexer)
+ || !parse_column__ (lex_integer (lexer), base, first_column))
return false;
- *first_column = lex_integer (lexer);
- if (*first_column < 1)
- {
- msg (SE, _("Column positions for fields must be positive."));
- return false;
- }
lex_get (lexer);
/* Last column. */
- lex_negative_to_dash (lexer);
- if (lex_match (lexer, '-'))
+ if (lex_is_integer (lexer) && lex_integer (lexer) < 0)
{
- if (!lex_force_int (lexer))
- return false;
- *last_column = lex_integer (lexer);
- if (*last_column < 1)
- {
- msg (SE, _("Column positions for fields must be positive."));
- return false;
- }
- else if (*last_column < *first_column)
+ if (!parse_column__ (-lex_integer (lexer), base, last_column))
+ return false;
+ lex_get (lexer);
+
+ if (*last_column < *first_column)
{
msg (SE, _("The ending column for a field must be "
"greater than the starting column."));
@@ -333,7 +376,6 @@ parse_column_range (struct lexer *lexer, int *first_column, int *last_column,
if (range_specified)
*range_specified = true;
- lex_get (lexer);
}
else
{
@@ -355,7 +397,7 @@ parse_column_range (struct lexer *lexer, int *first_column, int *last_column,
bool
parse_record_placement (struct lexer *lexer, int *record, int *column)
{
- while (lex_match (lexer, '/'))
+ while (lex_match (lexer, T_SLASH))
{
if (lex_is_integer (lexer))
{