X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fexpressions%2Fparse.c;h=ed5a07093c712b3b034b82f71d6af1fbdc4db0fd;hb=e63296a3c2bfcf1b40d712e79f17b349c4a6f21d;hp=8c53e200685da64a12067c44347aac8b3e00df24;hpb=f5c108becd49d78f4898cab11352291f5689d24e;p=pspp
diff --git a/src/language/expressions/parse.c b/src/language/expressions/parse.c
index 8c53e20068..ed5a07093c 100644
--- a/src/language/expressions/parse.c
+++ b/src/language/expressions/parse.c
@@ -1,20 +1,18 @@
-/* PSPP - computes sample statistics.
- Copyright (C) 1997-9, 2000, 2006 Free Software Foundation, Inc.
+/* PSPP - a program for statistical analysis.
+ Copyright (C) 1997-9, 2000, 2006, 2010, 2011 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
@@ -25,21 +23,23 @@
#include
#include
-#include "helpers.h"
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
+#include "data/case.h"
+#include "data/dictionary.h"
+#include "data/settings.h"
+#include "data/variable.h"
+#include "language/expressions/helpers.h"
+#include "language/lexer/format-parser.h"
+#include "language/lexer/lexer.h"
+#include "language/lexer/variable-parser.h"
+#include "libpspp/array.h"
+#include "libpspp/assertion.h"
+#include "libpspp/i18n.h"
+#include "libpspp/message.h"
+#include "libpspp/misc.h"
+#include "libpspp/pool.h"
+#include "libpspp/str.h"
+
+#include "gl/xalloc.h"
/* Declarations. */
@@ -345,7 +345,13 @@ type_coercion_core (struct expression *e,
{
/* Convert numeric to boolean. */
if (do_coercion)
- *node = expr_allocate_unary (e, OP_NUM_TO_BOOLEAN, *node);
+ {
+ union any_node *op_name;
+
+ op_name = expr_allocate_string (e, ss_cstr (operator_name));
+ *node = expr_allocate_binary (e, OP_NUM_TO_BOOLEAN, *node,
+ op_name);
+ }
return true;
}
break;
@@ -357,7 +363,7 @@ type_coercion_core (struct expression *e,
msg_disable ();
if ((*node)->type == OP_format
&& fmt_check_input (&(*node)->format.f)
- && fmt_check_type_compat (&(*node)->format.f, VAR_NUMERIC))
+ && fmt_check_type_compat (&(*node)->format.f, VAL_NUMERIC))
{
msg_enable ();
if (do_coercion)
@@ -371,7 +377,7 @@ type_coercion_core (struct expression *e,
msg_disable ();
if ((*node)->type == OP_format
&& fmt_check_output (&(*node)->format.f)
- && fmt_check_type_compat (&(*node)->format.f, VAR_NUMERIC))
+ && fmt_check_type_compat (&(*node)->format.f, VAL_NUMERIC))
{
msg_enable ();
if (do_coercion)
@@ -494,16 +500,14 @@ match_operator (struct lexer *lexer, const struct operator ops[], size_t op_cnt,
const struct operator *op;
for (op = ops; op < ops + op_cnt; op++)
- {
- if (op->token == '-')
- lex_negative_to_dash (lexer);
- if (lex_match (lexer, op->token))
- {
- if (operator != NULL)
- *operator = op;
- return true;
- }
- }
+ if (lex_token (lexer) == op->token)
+ {
+ if (op->token != T_NEG_NUM)
+ lex_get (lexer);
+ if (operator != NULL)
+ *operator = op;
+ return true;
+ }
if (operator != NULL)
*operator = NULL;
return false;
@@ -578,7 +582,7 @@ parse_binary_operators (struct lexer *lexer, struct expression *e, union any_nod
}
if (op_count > 1 && chain_warning != NULL)
- msg (SW, chain_warning);
+ msg (SW, "%s", chain_warning);
return node;
}
@@ -611,7 +615,7 @@ static union any_node *
parse_or (struct lexer *lexer, struct expression *e)
{
static const struct operator op =
- { T_OR, OP_OR, "logical disjunction (\"OR\")" };
+ { T_OR, OP_OR, "logical disjunction (`OR')" };
return parse_binary_operators (lexer, e, parse_and (lexer, e), &op, 1, parse_and, NULL);
}
@@ -621,7 +625,7 @@ static union any_node *
parse_and (struct lexer *lexer, struct expression *e)
{
static const struct operator op =
- { T_AND, OP_AND, "logical conjunction (\"AND\")" };
+ { T_AND, OP_AND, "logical conjunction (`AND')" };
return parse_binary_operators (lexer, e, parse_not (lexer, e),
&op, 1, parse_not, NULL);
@@ -632,7 +636,7 @@ static union any_node *
parse_not (struct lexer *lexer, struct expression *e)
{
static const struct operator op
- = { T_NOT, OP_NOT, "logical negation (\"NOT\")" };
+ = { T_NOT, OP_NOT, "logical negation (`NOT')" };
return parse_inverting_unary_operator (lexer, e, &op, parse_rel);
}
@@ -641,12 +645,12 @@ static union any_node *
parse_rel (struct lexer *lexer, struct expression *e)
{
const char *chain_warning =
- _("Chaining relational operators (e.g. \"a < b < c\") will "
+ _("Chaining relational operators (e.g. `a < b < c') will "
"not produce the mathematically expected result. "
"Use the AND logical operator to fix the problem "
- "(e.g. \"a < b AND b < c\"). "
+ "(e.g. `a < b AND b < c'). "
"If chaining is really intended, parentheses will disable "
- "this warning (e.g. \"(a < b) < c\".)");
+ "this warning (e.g. `(a < b) < c'.)");
union any_node *node = parse_add (lexer, e);
@@ -660,13 +664,13 @@ parse_rel (struct lexer *lexer, struct expression *e)
{
static const struct operator ops[] =
{
- { '=', OP_EQ, "numeric equality (\"=\")" },
- { T_EQ, OP_EQ, "numeric equality (\"EQ\")" },
- { T_GE, OP_GE, "numeric greater-than-or-equal-to (\">=\")" },
- { T_GT, OP_GT, "numeric greater than (\">\")" },
- { T_LE, OP_LE, "numeric less-than-or-equal-to (\"<=\")" },
- { T_LT, OP_LT, "numeric less than (\"<\")" },
- { T_NE, OP_NE, "numeric inequality (\"<>\")" },
+ { T_EQUALS, OP_EQ, "numeric equality (`=')" },
+ { T_EQ, OP_EQ, "numeric equality (`EQ')" },
+ { T_GE, OP_GE, "numeric greater-than-or-equal-to (`>=')" },
+ { T_GT, OP_GT, "numeric greater than (`>')" },
+ { T_LE, OP_LE, "numeric less-than-or-equal-to (`<=')" },
+ { T_LT, OP_LT, "numeric less than (`<')" },
+ { T_NE, OP_NE, "numeric inequality (`<>')" },
};
return parse_binary_operators (lexer, e, node, ops,
@@ -678,13 +682,13 @@ parse_rel (struct lexer *lexer, struct expression *e)
{
static const struct operator ops[] =
{
- { '=', OP_EQ_STRING, "string equality (\"=\")" },
- { T_EQ, OP_EQ_STRING, "string equality (\"EQ\")" },
- { T_GE, OP_GE_STRING, "string greater-than-or-equal-to (\">=\")" },
- { T_GT, OP_GT_STRING, "string greater than (\">\")" },
- { T_LE, OP_LE_STRING, "string less-than-or-equal-to (\"<=\")" },
- { T_LT, OP_LT_STRING, "string less than (\"<\")" },
- { T_NE, OP_NE_STRING, "string inequality (\"<>\")" },
+ { T_EQUALS, OP_EQ_STRING, "string equality (`=')" },
+ { T_EQ, OP_EQ_STRING, "string equality (`EQ')" },
+ { T_GE, OP_GE_STRING, "string greater-than-or-equal-to (`>=')" },
+ { T_GT, OP_GT_STRING, "string greater than (`>')" },
+ { T_LE, OP_LE_STRING, "string less-than-or-equal-to (`<=')" },
+ { T_LT, OP_LT_STRING, "string less than (`<')" },
+ { T_NE, OP_NE_STRING, "string inequality (`<>')" },
};
return parse_binary_operators (lexer, e, node, ops,
@@ -703,8 +707,9 @@ parse_add (struct lexer *lexer, struct expression *e)
{
static const struct operator ops[] =
{
- { '+', OP_ADD, "addition (\"+\")" },
- { '-', OP_SUB, "subtraction (\"-\")" },
+ { T_PLUS, OP_ADD, "addition (`+')" },
+ { T_DASH, OP_SUB, "subtraction (`-')" },
+ { T_NEG_NUM, OP_ADD, "subtraction (`-')" },
};
return parse_binary_operators (lexer, e, parse_mul (lexer, e),
@@ -718,8 +723,8 @@ parse_mul (struct lexer *lexer, struct expression *e)
{
static const struct operator ops[] =
{
- { '*', OP_MUL, "multiplication (\"*\")" },
- { '/', OP_DIV, "division (\"/\")" },
+ { T_ASTERISK, OP_MUL, "multiplication (`*')" },
+ { T_SLASH, OP_DIV, "division (`/')" },
};
return parse_binary_operators (lexer, e, parse_neg (lexer, e),
@@ -731,7 +736,7 @@ parse_mul (struct lexer *lexer, struct expression *e)
static union any_node *
parse_neg (struct lexer *lexer, struct expression *e)
{
- static const struct operator op = { '-', OP_NEG, "negation (\"-\")" };
+ static const struct operator op = { T_DASH, OP_NEG, "negation (`-')" };
return parse_inverting_unary_operator (lexer, e, &op, parse_exp);
}
@@ -739,16 +744,29 @@ static union any_node *
parse_exp (struct lexer *lexer, struct expression *e)
{
static const struct operator op =
- { T_EXP, OP_POW, "exponentiation (\"**\")" };
+ { T_EXP, OP_POW, "exponentiation (`**')" };
const char *chain_warning =
- _("The exponentiation operator (\"**\") is left-associative, "
+ _("The exponentiation operator (`**') is left-associative, "
"even though right-associative semantics are more useful. "
- "That is, \"a**b**c\" equals \"(a**b)**c\", not as \"a**(b**c)\". "
+ "That is, `a**b**c' equals `(a**b)**c', not as `a**(b**c)'. "
"To disable this warning, insert parentheses.");
- return parse_binary_operators (lexer, e, parse_primary (lexer, e), &op, 1,
- parse_primary, chain_warning);
+ union any_node *lhs, *node;
+ bool negative = false;
+
+ if (lex_token (lexer) == T_NEG_NUM)
+ {
+ lhs = expr_allocate_number (e, -lex_tokval (lexer));
+ negative = true;
+ lex_get (lexer);
+ }
+ else
+ lhs = parse_primary (lexer, e);
+
+ node = parse_binary_operators (lexer, e, lhs, &op, 1,
+ parse_primary, chain_warning);
+ return negative ? expr_allocate_unary (e, OP_NEG, node) : node;
}
/* Parses system variables. */
@@ -768,12 +786,14 @@ parse_sysvar (struct lexer *lexer, struct expression *e)
time_t last_proc_time = time_of_last_procedure (e->ds);
struct tm *time;
char temp_buf[10];
+ struct substring s;
time = localtime (&last_proc_time);
sprintf (temp_buf, "%02d %s %02d", abs (time->tm_mday) % 100,
months[abs (time->tm_mon) % 12], abs (time->tm_year) % 100);
- return expr_allocate_string_buffer (e, temp_buf, strlen (temp_buf));
+ ss_alloc_substring (&s, ss_cstr (temp_buf));
+ return expr_allocate_string (e, s);
}
else if (lex_match_id (lexer, "$TRUE"))
return expr_allocate_boolean (e, 1.0);
@@ -802,12 +822,12 @@ parse_sysvar (struct lexer *lexer, struct expression *e)
+ tm->tm_sec);
}
else if (lex_match_id (lexer, "$LENGTH"))
- return expr_allocate_number (e, get_viewlength ());
+ return expr_allocate_number (e, settings_get_viewlength ());
else if (lex_match_id (lexer, "$WIDTH"))
- return expr_allocate_number (e, get_viewwidth ());
+ return expr_allocate_number (e, settings_get_viewwidth ());
else
{
- msg (SE, _("Unknown system variable %s."), lex_tokid (lexer));
+ msg (SE, _("Unknown system variable %s."), lex_tokcstr (lexer));
return NULL;
}
}
@@ -819,22 +839,22 @@ parse_primary (struct lexer *lexer, struct expression *e)
switch (lex_token (lexer))
{
case T_ID:
- if (lex_look_ahead (lexer) == '(')
+ if (lex_next_token (lexer, 1) == T_LPAREN)
{
/* An identifier followed by a left parenthesis may be
a vector element reference. If not, it's a function
call. */
- if (e->ds != NULL && dict_lookup_vector (dataset_dict (e->ds), lex_tokid (lexer)) != NULL)
+ if (e->ds != NULL && dict_lookup_vector (dataset_dict (e->ds), lex_tokcstr (lexer)) != NULL)
return parse_vector_element (lexer, e);
else
return parse_function (lexer, e);
}
- else if (lex_tokid (lexer)[0] == '$')
+ else if (lex_tokcstr (lexer)[0] == '$')
{
/* $ at the beginning indicates a system variable. */
return parse_sysvar (lexer, e);
}
- else if (e->ds != NULL && dict_lookup_var (dataset_dict (e->ds), lex_tokid (lexer)))
+ else if (e->ds != NULL && dict_lookup_var (dataset_dict (e->ds), lex_tokcstr (lexer)))
{
/* It looks like a user variable.
(It could be a format specifier, but we'll assume
@@ -855,7 +875,7 @@ parse_primary (struct lexer *lexer, struct expression *e)
return expr_allocate_format (e, &fmt);
/* All attempts failed. */
- msg (SE, _("Unknown identifier %s."), lex_tokid (lexer));
+ msg (SE, _("Unknown identifier %s."), lex_tokcstr (lexer));
return NULL;
}
break;
@@ -870,27 +890,33 @@ parse_primary (struct lexer *lexer, struct expression *e)
case T_STRING:
{
- union any_node *node = expr_allocate_string_buffer (
- e, ds_cstr (lex_tokstr (lexer) ), ds_length (lex_tokstr (lexer) ));
+ const char *dict_encoding;
+ union any_node *node;
+ char *s;
+
+ dict_encoding = (e->ds != NULL
+ ? dict_get_encoding (dataset_dict (e->ds))
+ : "UTF-8");
+ s = recode_string (dict_encoding, "UTF-8", lex_tokcstr (lexer),
+ ss_length (lex_tokss (lexer)));
+ node = expr_allocate_string (e, ss_cstr (s));
+
lex_get (lexer);
return node;
}
- case '(':
+ case T_LPAREN:
{
union any_node *node;
lex_get (lexer);
node = parse_or (lexer, e);
- if (node != NULL && !lex_match (lexer, ')'))
- {
- lex_error (lexer, _("expecting `)'"));
- return NULL;
- }
+ if (node != NULL && !lex_force_match (lexer, T_RPAREN))
+ return NULL;
return node;
}
default:
- lex_error (lexer, _("in expression"));
+ lex_error (lexer, NULL);
return NULL;
}
}
@@ -904,22 +930,22 @@ parse_vector_element (struct lexer *lexer, struct expression *e)
/* Find vector, skip token.
The caller must already have verified that the current token
is the name of a vector. */
- vector = dict_lookup_vector (dataset_dict (e->ds), lex_tokid (lexer));
+ vector = dict_lookup_vector (dataset_dict (e->ds), lex_tokcstr (lexer));
assert (vector != NULL);
lex_get (lexer);
/* Skip left parenthesis token.
The caller must have verified that the lookahead is a left
parenthesis. */
- assert (lex_token (lexer) == '(');
+ assert (lex_token (lexer) == T_LPAREN);
lex_get (lexer);
element = parse_or (lexer, e);
if (!type_coercion (e, OP_number, &element, "vector indexing")
- || !lex_match (lexer, ')'))
+ || !lex_match (lexer, T_RPAREN))
return NULL;
- return expr_allocate_binary (e, (vector_get_type (vector) == VAR_NUMERIC
+ return expr_allocate_binary (e, (vector_get_type (vector) == VAL_NUMERIC
? OP_VEC_ELEM_NUM : OP_VEC_ELEM_STR),
element, expr_allocate_vector (e, vector));
}
@@ -1019,7 +1045,7 @@ lookup_function (const char *name,
}
static int
-extract_min_valid (char *s)
+extract_min_valid (const char *s)
{
char *p = strrchr (s, '.');
if (p == NULL
@@ -1080,7 +1106,7 @@ validate_function_args (const struct operation *f, int arg_cnt, int min_valid)
&& array_arg_cnt % f->array_granularity != 0)
{
if (f->array_granularity == 2)
- msg (SE, _("%s must have even number of arguments in list."),
+ msg (SE, _("%s must have an even number of arguments in list."),
f->prototype);
else
msg (SE, _("%s must have multiple of %d arguments in list."),
@@ -1147,7 +1173,7 @@ put_invocation (struct string *s,
ds_put_cstr (s, ", ");
ds_put_cstr (s, operations[expr_node_returns (args[i])].prototype);
}
- ds_put_char (s, ')');
+ ds_put_byte (s, ')');
}
static void
@@ -1174,7 +1200,7 @@ no_match (const char *func_name,
for (f = first; f < last; f++)
ds_put_format (&s, "\n%s", f->prototype);
}
- ds_put_char (&s, '.');
+ ds_put_byte (&s, '.');
msg (SE, "%s", ds_cstr (&s));
@@ -1195,17 +1221,17 @@ parse_function (struct lexer *lexer, struct expression *e)
union any_node *n;
- ds_init_string (&func_name, lex_tokstr (lexer));
- min_valid = extract_min_valid (ds_cstr (lex_tokstr (lexer)));
- if (!lookup_function (ds_cstr (lex_tokstr (lexer)), &first, &last))
+ ds_init_substring (&func_name, lex_tokss (lexer));
+ min_valid = extract_min_valid (lex_tokcstr (lexer));
+ if (!lookup_function (lex_tokcstr (lexer), &first, &last))
{
- msg (SE, _("No function or vector named %s."), ds_cstr (lex_tokstr (lexer)));
+ msg (SE, _("No function or vector named %s."), lex_tokcstr (lexer));
ds_destroy (&func_name);
return NULL;
}
lex_get (lexer);
- if (!lex_force_match (lexer, '('))
+ if (!lex_force_match (lexer, T_LPAREN))
{
ds_destroy (&func_name);
return NULL;
@@ -1213,11 +1239,11 @@ parse_function (struct lexer *lexer, struct expression *e)
args = NULL;
arg_cnt = arg_cap = 0;
- if (lex_token (lexer) != ')')
+ if (lex_token (lexer) != T_RPAREN)
for (;;)
{
if (lex_token (lexer) == T_ID
- && toupper (lex_look_ahead (lexer)) == 'T')
+ && lex_next_token (lexer, 1) == T_TO)
{
const struct variable **vars;
size_t var_cnt;
@@ -1238,9 +1264,9 @@ parse_function (struct lexer *lexer, struct expression *e)
add_arg (&args, &arg_cnt, &arg_cap, arg);
}
- if (lex_match (lexer, ')'))
+ if (lex_match (lexer, T_RPAREN))
break;
- else if (!lex_match (lexer, ','))
+ else if (!lex_match (lexer, T_COMMA))
{
lex_error (lexer, _("expecting `,' or `)' invoking %s function"),
first->name);
@@ -1261,7 +1287,7 @@ parse_function (struct lexer *lexer, struct expression *e)
if (!validate_function_args (f, arg_cnt, min_valid))
goto fail;
- if ((f->flags & OPF_EXTENSION) && get_syntax () == COMPATIBLE)
+ if ((f->flags & OPF_EXTENSION) && settings_get_syntax () == COMPATIBLE)
msg (SW, _("%s is a PSPP extension."), f->prototype);
if (f->flags & OPF_UNIMPLEMENTED)
{
@@ -1459,18 +1485,6 @@ expr_allocate_vector (struct expression *e, const struct vector *vector)
return n;
}
-union any_node *
-expr_allocate_string_buffer (struct expression *e,
- const char *string, size_t length)
-{
- union any_node *n = pool_alloc (e->expr_pool, sizeof n->string);
- n->type = OP_string;
- if (length > MAX_STRING)
- length = MAX_STRING;
- n->string.s = copy_string (e, string, length);
- return n;
-}
-
union any_node *
expr_allocate_string (struct expression *e, struct substring s)
{