1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2013, 2016 Free Software Foundation, Inc.
3
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
8
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
13
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
16
17 #include <config.h>
18
19 #include <stdlib.h>
20
21 #include "data/data-in.h"
22 #include "data/dictionary.h"
23 #include "data/dataset.h"
24 #include "data/format.h"
25 #include "data/missing-values.h"
26 #include "data/value.h"
27 #include "data/variable.h"
28 #include "language/command.h"
29 #include "language/lexer/lexer.h"
30 #include "language/lexer/value-parser.h"
31 #include "language/lexer/variable-parser.h"
32 #include "libpspp/i18n.h"
33 #include "libpspp/message.h"
34 #include "libpspp/str.h"
35
36 #include "gettext.h"
37 #define _(msgid) gettext (msgid)
38
39 int
cmd_missing_values(struct lexer * lexer,struct dataset * ds)40 cmd_missing_values (struct lexer *lexer, struct dataset *ds)
41 {
42 struct dictionary *dict = dataset_dict (ds);
43 struct variable **v = NULL;
44 size_t nv;
45
46 bool ok = true;
47
48 while (lex_token (lexer) != T_ENDCMD)
49 {
50 size_t i;
51
52 if (!parse_variables (lexer, dict, &v, &nv, PV_NONE))
53 goto error;
54
55 if (!lex_force_match (lexer, T_LPAREN))
56 goto error;
57
58 for (i = 0; i < nv; i++)
59 var_clear_missing_values (v[i]);
60
61 if (!lex_match (lexer, T_RPAREN))
62 {
63 struct missing_values mv;
64
65 for (i = 0; i < nv; i++)
66 if (var_get_type (v[i]) != var_get_type (v[0]))
67 {
68 const struct variable *n = var_is_numeric (v[0]) ? v[0] : v[i];
69 const struct variable *s = var_is_numeric (v[0]) ? v[i] : v[0];
70 msg (SE, _("Cannot mix numeric variables (e.g. %s) and "
71 "string variables (e.g. %s) within a single list."),
72 var_get_name (n), var_get_name (s));
73 goto error;
74 }
75
76 if (var_is_numeric (v[0]))
77 {
78 mv_init (&mv, 0);
79 while (!lex_match (lexer, T_RPAREN))
80 {
81 enum fmt_type type = var_get_print_format (v[0])->type;
82 double x, y;
83
84 if (!parse_num_range (lexer, &x, &y, &type))
85 goto error;
86
87 if (!(x == y
88 ? mv_add_num (&mv, x)
89 : mv_add_range (&mv, x, y)))
90 {
91 msg (SE, _("Too many numeric missing values. At most "
92 "three individual values or one value and "
93 "one range are allowed."));
94 ok = false;
95 }
96
97 lex_match (lexer, T_COMMA);
98 }
99 }
100 else
101 {
102 const char *encoding = dict_get_encoding (dict);
103
104 mv_init (&mv, MV_MAX_STRING);
105 while (!lex_match (lexer, T_RPAREN))
106 {
107 const char *utf8_s;
108 size_t utf8_trunc_len;
109 size_t utf8_len;
110
111 char *raw_s;
112
113 if (!lex_force_string (lexer))
114 {
115 ok = false;
116 break;
117 }
118
119 /* Truncate the string to fit in 8 bytes in the dictionary
120 encoding. */
121 utf8_s = lex_tokcstr (lexer);
122 utf8_len = ss_length (lex_tokss (lexer));
123 utf8_trunc_len = utf8_encoding_trunc_len (utf8_s, encoding,
124 MV_MAX_STRING);
125 if (utf8_trunc_len < utf8_len)
126 msg (SE, _("Truncating missing value to maximum "
127 "acceptable length (%d bytes)."),
128 MV_MAX_STRING);
129
130 /* Recode to dictionary encoding and add. */
131 raw_s = recode_string (encoding, "UTF-8",
132 utf8_s, utf8_trunc_len);
133 if (!mv_add_str (&mv, CHAR_CAST (const uint8_t *, raw_s),
134 strlen (raw_s)))
135 {
136 msg (SE,
137 _("Too many string missing values. "
138 "At most three individual values are allowed."));
139 ok = false;
140 }
141 free (raw_s);
142
143 lex_get (lexer);
144 lex_match (lexer, T_COMMA);
145 }
146 }
147
148 for (i = 0; i < nv; i++)
149 {
150 if (mv_is_resizable (&mv, var_get_width (v[i])))
151 var_set_missing_values (v[i], &mv);
152 else
153 {
154 msg (SE, _("Missing values provided are too long to assign "
155 "to variable of width %d."),
156 var_get_width (v[i]));
157 ok = false;
158 }
159 }
160
161 mv_destroy (&mv);
162 }
163
164 lex_match (lexer, T_SLASH);
165 free (v);
166 v = NULL;
167 }
168
169 free (v);
170 return ok ? CMD_SUCCESS : CMD_FAILURE;
171
172 error:
173 free (v);
174 return CMD_FAILURE;
175 }
176
177