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))
67 token_uninit (&another_token);
68 lex_error (lexer, _("String must contain exactly one token."));
76 macro_tokenize (struct macro *m, const struct substring body, 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 while (state.body.length > 0)
92 struct macro_token mt = {
93 .token = { .type = T_STOP },
94 .representation = { .string = state.body.string },
96 struct token *token = &mt.token;
98 struct scanner scanner;
99 scanner_init (&scanner, token);
103 enum segment_type type;
104 int seg_len = segmenter_push (&state.segmenter, state.body.string,
105 state.body.length, true, &type);
106 assert (seg_len >= 0);
108 struct substring segment = ss_head (state.body, seg_len);
109 ss_advance (&state.body, seg_len);
111 enum scan_result result = scanner_push (&scanner, type, segment, token);
112 if (result == SCAN_SAVE)
114 else if (result == SCAN_BACK)
119 else if (result == SCAN_DONE)
123 /* We have a token in 'token'. */
124 if (is_scan_type (token->type))
126 if (token->type != SCAN_SKIP)
128 /* XXX report error */
133 mt.representation.length = state.body.string - mt.representation.string;
134 macro_tokens_add (&m->body, &mt);
136 token_uninit (token);
141 cmd_define (struct lexer *lexer, struct dataset *ds UNUSED)
143 if (!force_macro_id (lexer))
146 /* Parse macro name. */
147 struct macro *m = xmalloc (sizeof *m);
148 *m = (struct macro) { .name = ss_xstrdup (lex_tokss (lexer)) };
151 if (!lex_force_match (lexer, T_LPAREN))
154 size_t allocated_params = 0;
155 while (!lex_match (lexer, T_RPAREN))
157 if (m->n_params >= allocated_params)
158 m->params = x2nrealloc (m->params, &allocated_params,
161 size_t param_index = m->n_params++;
162 struct macro_param *p = &m->params[param_index];
163 *p = (struct macro_param) { .expand_arg = true };
165 /* Parse parameter name. */
166 if (match_macro_id (lexer, "!POSITIONAL"))
168 if (param_index > 0 && !m->params[param_index - 1].positional)
170 lex_error (lexer, _("Positional parameters must precede "
171 "keyword parameters."));
175 p->positional = true;
176 p->name = xasprintf ("!%zu", param_index + 1);
180 if (!lex_force_id (lexer))
183 p->positional = false;
184 p->name = xasprintf ("!%s", lex_tokcstr (lexer));
187 if (!lex_force_match (lexer, T_EQUALS))
191 /* Parse default value. */
192 if (match_macro_id (lexer, "!DEFAULT"))
194 if (!lex_force_match (lexer, T_LPAREN))
197 /* XXX Should this handle balanced inner parentheses? */
198 while (!lex_match (lexer, T_RPAREN))
200 if (lex_token (lexer) == T_ENDCMD)
202 lex_error_expecting (lexer, ")");
205 const struct macro_token mt = {
206 .token = *lex_next (lexer, 0),
207 .representation = lex_next_representation (lexer, 0, 0),
209 macro_tokens_add (&p->def, &mt);
214 if (match_macro_id (lexer, "!NOEXPAND"))
215 p->expand_arg = false;
217 if (match_macro_id (lexer, "!TOKENS"))
219 if (!lex_force_match (lexer, T_LPAREN)
220 || !lex_force_int_range (lexer, "!TOKENS", 1, INT_MAX))
222 p->arg_type = ARG_N_TOKENS;
223 p->n_tokens = lex_integer (lexer);
225 if (!lex_force_match (lexer, T_RPAREN))
228 else if (match_macro_id (lexer, "!CHAREND"))
230 p->arg_type = ARG_CHAREND;
231 p->charend = (struct token) { .type = T_STOP };
233 if (!lex_force_match (lexer, T_LPAREN)
234 || !parse_quoted_token (lexer, &p->charend)
235 || !lex_force_match (lexer, T_RPAREN))
238 else if (match_macro_id (lexer, "!ENCLOSE"))
240 p->arg_type = ARG_ENCLOSE;
241 p->enclose[0] = p->enclose[1] = (struct token) { .type = T_STOP };
243 if (!lex_force_match (lexer, T_LPAREN)
244 || !parse_quoted_token (lexer, &p->enclose[0])
245 || !lex_force_match (lexer, T_COMMA)
246 || !parse_quoted_token (lexer, &p->enclose[1])
247 || !lex_force_match (lexer, T_RPAREN))
250 else if (match_macro_id (lexer, "!CMDEND"))
251 p->arg_type = ARG_CMDEND;
254 lex_error_expecting (lexer, "!TOKENS", "!CHAREND",
255 "!ENCLOSE", "!CMDEND");
259 if (lex_token (lexer) != T_RPAREN && !lex_force_match (lexer, T_SLASH))
263 struct string body = DS_EMPTY_INITIALIZER;
264 while (!match_macro_id (lexer, "!ENDDEFINE"))
266 if (lex_token (lexer) != T_STRING)
268 lex_error (lexer, _("Expecting macro body or !ENDDEFINE"));
273 ds_put_substring (&body, lex_tokss (lexer));
274 ds_put_byte (&body, '\n');
278 macro_tokenize (m, body.ss, lexer);
281 lex_define_macro (lexer, m);