1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 2021 Free Software Foundation, Inc.
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.
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.
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/>. */
21 #include "language/command.h"
22 #include "language/lexer/lexer.h"
23 #include "language/lexer/macro.h"
24 #include "language/lexer/scan.h"
25 #include "language/lexer/token.h"
27 #include "gl/xalloc.h"
30 #define _(msgid) gettext (msgid)
33 force_macro_id (struct lexer *lexer)
35 return lex_token (lexer) == T_MACRO_ID || lex_force_id (lexer);
39 match_macro_id (struct lexer *lexer, const char *id)
42 return lex_match_id (lexer, id);
43 else if (lex_token (lexer) == T_MACRO_ID
44 && ss_equals_case (lex_tokss (lexer), ss_cstr (id)))
54 parse_quoted_token (struct lexer *lexer, struct token *token)
56 if (!lex_force_string (lexer))
59 struct substring s = lex_tokss (lexer);
60 struct string_lexer slex;
61 string_lexer_init (&slex, s.string, s.length, SEG_MODE_INTERACTIVE);
62 struct token another_token;
63 if (!string_lexer_next (&slex, token)
64 || string_lexer_next (&slex, &another_token))
66 token_destroy (token);
67 token_destroy (&another_token);
68 lex_error (lexer, _("String must contain exactly one token."));
76 macro_tokenize (struct macro *m, struct lexer *lexer)
80 struct segmenter segmenter;
81 struct substring body;
84 struct state state = {
85 .segmenter = SEGMENTER_INIT (lex_get_syntax_mode (lexer)),
88 struct state saved = state;
90 struct token token = { .type = T_STOP };
92 while (state.body.length > 0)
94 struct scanner scanner;
95 scanner_init (&scanner, &token);
99 enum segment_type type;
100 int seg_len = segmenter_push (&state.segmenter, state.body.string,
101 state.body.length, true, &type);
102 assert (seg_len >= 0);
104 struct substring segment = ss_head (state.body, seg_len);
105 ss_advance (&state.body, seg_len);
107 enum scan_result result = scanner_push (&scanner, type, segment, &token);
108 if (result == SCAN_SAVE)
110 else if (result == SCAN_BACK)
115 else if (result == SCAN_DONE)
119 /* We have a token in 'token'. */
120 if (is_scan_type (token.type))
122 if (token.type != SCAN_SKIP)
124 /* XXX report error */
128 tokens_add (&m->body_tokens, &token);
129 token_destroy (&token);
134 cmd_define (struct lexer *lexer, struct dataset *ds UNUSED)
136 if (!force_macro_id (lexer))
139 /* Parse macro name. */
140 struct macro *m = xmalloc (sizeof *m);
141 *m = (struct macro) { .name = ss_xstrdup (lex_tokss (lexer)) };
144 if (!lex_force_match (lexer, T_LPAREN))
147 size_t allocated_params = 0;
148 while (!lex_match (lexer, T_RPAREN))
150 if (m->n_params >= allocated_params)
151 m->params = x2nrealloc (m->params, &allocated_params,
154 size_t param_index = m->n_params++;
155 struct macro_param *p = &m->params[param_index];
156 *p = (struct macro_param) { .expand_arg = true };
158 /* Parse parameter name. */
159 if (match_macro_id (lexer, "!POSITIONAL"))
161 if (param_index > 0 && !m->params[param_index - 1].positional)
163 lex_error (lexer, _("Positional parameters must precede "
164 "keyword parameters."));
168 p->positional = true;
169 p->name = xasprintf ("!%zu", param_index + 1);
173 if (!lex_force_id (lexer))
176 p->positional = false;
177 p->name = xasprintf ("!%s", lex_tokcstr (lexer));
180 if (!lex_force_match (lexer, T_EQUALS))
184 /* Parse default value. */
185 if (match_macro_id (lexer, "!DEFAULT"))
187 if (!lex_force_match (lexer, T_LPAREN))
190 /* XXX Should this handle balanced inner parentheses? */
191 while (!lex_match (lexer, T_RPAREN))
193 if (lex_token (lexer) == T_ENDCMD)
195 lex_error_expecting (lexer, ")");
198 tokens_add (&p->def, lex_next (lexer, 0));
203 if (match_macro_id (lexer, "!NOEXPAND"))
204 p->expand_arg = false;
206 if (match_macro_id (lexer, "!TOKENS"))
208 if (!lex_force_match (lexer, T_LPAREN)
209 || !lex_force_int_range (lexer, "!TOKENS", 1, INT_MAX))
211 p->arg_type = ARG_N_TOKENS;
212 p->n_tokens = lex_integer (lexer);
214 if (!lex_force_match (lexer, T_RPAREN))
217 else if (match_macro_id (lexer, "!CHAREND"))
219 p->arg_type = ARG_CHAREND;
220 p->charend = (struct token) { .type = T_STOP };
222 if (!lex_force_match (lexer, T_LPAREN)
223 || !parse_quoted_token (lexer, &p->charend)
224 || !lex_force_match (lexer, T_RPAREN))
227 else if (match_macro_id (lexer, "!ENCLOSE"))
229 p->arg_type = ARG_ENCLOSE;
230 p->enclose[0] = p->enclose[1] = (struct token) { .type = T_STOP };
232 if (!lex_force_match (lexer, T_LPAREN)
233 || !parse_quoted_token (lexer, &p->enclose[0])
234 || !lex_force_match (lexer, T_COMMA)
235 || !parse_quoted_token (lexer, &p->enclose[1])
236 || !lex_force_match (lexer, T_RPAREN))
239 else if (match_macro_id (lexer, "!CMDEND"))
240 p->arg_type = ARG_CMDEND;
243 lex_error_expecting (lexer, "!TOKENS", "!CHAREND",
244 "!ENCLOSE", "!CMDEND");
248 if (lex_token (lexer) != T_RPAREN && !lex_force_match (lexer, T_SLASH))
252 struct string body = DS_EMPTY_INITIALIZER;
253 while (!match_macro_id (lexer, "!ENDDEFINE"))
255 if (lex_token (lexer) != T_STRING)
257 lex_error (lexer, _("Expecting macro body or !ENDDEFINE"));
262 ds_put_substring (&body, lex_tokss (lexer));
263 ds_put_byte (&body, '\n');
266 m->body = ds_ss (&body);
268 macro_tokenize (m, lexer);
270 lex_define_macro (lexer, m);