1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 2007, 2008, 2009, 2010, 2011, 2012, 2013 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/>. */
23 #include "data/dataset.h"
24 #include "data/dictionary.h"
25 #include "data/format.h"
26 #include "data/gnumeric-reader.h"
27 #include "data/ods-reader.h"
28 #include "data/spreadsheet-reader.h"
29 #include "data/psql-reader.h"
30 #include "data/settings.h"
31 #include "language/command.h"
32 #include "language/data-io/data-parser.h"
33 #include "language/data-io/data-reader.h"
34 #include "language/data-io/file-handle.h"
35 #include "language/data-io/placement-parser.h"
36 #include "language/lexer/format-parser.h"
37 #include "language/lexer/lexer.h"
38 #include "libpspp/cast.h"
39 #include "libpspp/i18n.h"
40 #include "libpspp/message.h"
42 #include "gl/xalloc.h"
45 #define _(msgid) gettext (msgid)
46 #define N_(msgid) (msgid)
48 static struct spreadsheet_read_info *parse_spreadsheet (struct lexer *lexer);
49 static void destroy_spreadsheet_read_info (struct spreadsheet_read_info *);
51 static int parse_get_txt (struct lexer *lexer, struct dataset *);
52 static int parse_get_psql (struct lexer *lexer, struct dataset *);
55 cmd_get_data (struct lexer *lexer, struct dataset *ds)
58 lex_force_match (lexer, T_SLASH);
60 if (!lex_force_match_id (lexer, "TYPE"))
63 lex_force_match (lexer, T_EQUALS);
65 tok = strdup (lex_tokcstr (lexer));
66 if (lex_match_id (lexer, "TXT"))
69 return parse_get_txt (lexer, ds);
71 else if (lex_match_id (lexer, "PSQL"))
74 return parse_get_psql (lexer, ds);
76 else if (lex_match_id (lexer, "GNM") ||
77 lex_match_id (lexer, "ODS"))
79 struct casereader *reader = NULL;
80 struct dictionary *dict = NULL;
81 struct spreadsheet_read_info *sri = parse_spreadsheet (lexer);
85 if ( 0 == strncasecmp (tok, "GNM", 3))
86 reader = gnumeric_open_reader (sri, &dict);
87 else if (0 == strncasecmp (tok, "ODS", 3))
88 reader = ods_open_reader (sri, &dict);
92 dataset_set_dict (ds, dict);
93 dataset_set_source (ds, reader);
94 destroy_spreadsheet_read_info (sri);
98 destroy_spreadsheet_read_info (sri);
101 msg (SE, _("Unsupported TYPE %s."), tok);
109 parse_get_psql (struct lexer *lexer, struct dataset *ds)
111 struct psql_read_info psql;
112 psql.allow_clear = false;
113 psql.conninfo = NULL;
116 ds_init_empty (&psql.sql);
118 lex_force_match (lexer, T_SLASH);
120 if (!lex_force_match_id (lexer, "CONNECT"))
123 lex_force_match (lexer, T_EQUALS);
125 if (!lex_force_string (lexer))
128 psql.conninfo = ss_xstrdup (lex_tokss (lexer));
132 while (lex_match (lexer, T_SLASH) )
134 if ( lex_match_id (lexer, "ASSUMEDSTRWIDTH"))
136 lex_match (lexer, T_EQUALS);
137 psql.str_width = lex_integer (lexer);
140 else if ( lex_match_id (lexer, "BSIZE"))
142 lex_match (lexer, T_EQUALS);
143 psql.bsize = lex_integer (lexer);
146 else if ( lex_match_id (lexer, "UNENCRYPTED"))
148 psql.allow_clear = true;
150 else if (lex_match_id (lexer, "SQL"))
152 lex_match (lexer, T_EQUALS);
153 if ( ! lex_force_string (lexer) )
156 ds_put_substring (&psql.sql, lex_tokss (lexer));
161 struct dictionary *dict = NULL;
162 struct casereader *reader = psql_open_reader (&psql, &dict);
166 dataset_set_dict (ds, dict);
167 dataset_set_source (ds, reader);
171 ds_destroy (&psql.sql);
172 free (psql.conninfo);
178 ds_destroy (&psql.sql);
179 free (psql.conninfo);
184 static struct spreadsheet_read_info *
185 parse_spreadsheet (struct lexer *lexer)
187 struct spreadsheet_read_info *sri = xzalloc (sizeof *sri);
188 sri->sheet_index = 1;
189 sri->read_names = true;
192 lex_force_match (lexer, T_SLASH);
194 if (!lex_force_match_id (lexer, "FILE"))
197 lex_force_match (lexer, T_EQUALS);
199 if (!lex_force_string (lexer))
202 sri->file_name = utf8_to_filename (lex_tokcstr (lexer));
206 while (lex_match (lexer, T_SLASH) )
208 if ( lex_match_id (lexer, "ASSUMEDSTRWIDTH"))
210 lex_match (lexer, T_EQUALS);
211 sri->asw = lex_integer (lexer);
214 else if (lex_match_id (lexer, "SHEET"))
216 lex_match (lexer, T_EQUALS);
217 if (lex_match_id (lexer, "NAME"))
219 if ( ! lex_force_string (lexer) )
222 sri->sheet_name = ss_xstrdup (lex_tokss (lexer));
223 sri->sheet_index = -1;
227 else if (lex_match_id (lexer, "INDEX"))
229 sri->sheet_index = lex_integer (lexer);
234 msg (SE, _("%s must be followed by either \"%s\" or \"%s\"."),
235 "/SHEET", "NAME", "INDEX");
239 else if (lex_match_id (lexer, "CELLRANGE"))
241 lex_match (lexer, T_EQUALS);
243 if (lex_match_id (lexer, "FULL"))
245 sri->cell_range = NULL;
247 else if (lex_match_id (lexer, "RANGE"))
249 if ( ! lex_force_string (lexer) )
252 sri->cell_range = ss_xstrdup (lex_tokss (lexer));
257 msg (SE, _("%s must be followed by either \"%s\" or \"%s\"."),
258 "/CELLRANGE", "FULL", "RANGE");
262 else if (lex_match_id (lexer, "READNAMES"))
264 lex_match (lexer, T_EQUALS);
266 if ( lex_match_id (lexer, "ON"))
268 sri->read_names = true;
270 else if (lex_match_id (lexer, "OFF"))
272 sri->read_names = false;
276 msg (SE, _("%s must be followed by either \"%s\" or \"%s\"."),
277 "/READNAMES", "ON", "OFF");
283 lex_error (lexer, NULL);
291 destroy_spreadsheet_read_info (sri);
297 set_type (struct data_parser *parser, const char *subcommand,
298 enum data_parser_type type, bool *has_type)
302 data_parser_set_type (parser, type);
305 else if (type != data_parser_get_type (parser))
307 msg (SE, _("%s is allowed only with %s arrangement, but %s arrangement "
308 "was stated or implied earlier in this command."),
310 type == DP_FIXED ? "FIXED" : "DELIMITED",
311 type == DP_FIXED ? "DELIMITED" : "FIXED");
318 parse_get_txt (struct lexer *lexer, struct dataset *ds)
320 struct data_parser *parser = NULL;
321 struct dictionary *dict = dict_create (get_default_encoding ());
322 struct file_handle *fh = NULL;
323 struct dfm_reader *reader = NULL;
324 char *encoding = NULL;
328 enum data_parser_type type;
331 lex_force_match (lexer, T_SLASH);
333 if (!lex_force_match_id (lexer, "FILE"))
335 lex_force_match (lexer, T_EQUALS);
336 fh = fh_parse (lexer, FH_REF_FILE | FH_REF_INLINE, NULL);
340 parser = data_parser_create (dict);
342 data_parser_set_type (parser, DP_DELIMITED);
343 data_parser_set_span (parser, false);
344 data_parser_set_quotes (parser, ss_empty ());
345 data_parser_set_empty_line_has_field (parser, true);
349 if (!lex_force_match (lexer, T_SLASH))
352 if (lex_match_id (lexer, "ENCODING"))
354 lex_match (lexer, T_EQUALS);
355 if (!lex_force_string (lexer))
359 encoding = ss_xstrdup (lex_tokss (lexer));
363 else if (lex_match_id (lexer, "ARRANGEMENT"))
367 lex_match (lexer, T_EQUALS);
368 if (lex_match_id (lexer, "FIXED"))
369 ok = set_type (parser, "ARRANGEMENT=FIXED", DP_FIXED, &has_type);
370 else if (lex_match_id (lexer, "DELIMITED"))
371 ok = set_type (parser, "ARRANGEMENT=DELIMITED",
372 DP_DELIMITED, &has_type);
375 lex_error_expecting (lexer, "FIXED", "DELIMITED", NULL_SENTINEL);
381 else if (lex_match_id (lexer, "FIRSTCASE"))
383 lex_match (lexer, T_EQUALS);
384 if (!lex_force_int (lexer))
386 if (lex_integer (lexer) < 1)
388 msg (SE, _("Value of FIRSTCASE must be 1 or greater."));
391 data_parser_set_skip (parser, lex_integer (lexer) - 1);
394 else if (lex_match_id_n (lexer, "DELCASE", 4))
396 if (!set_type (parser, "DELCASE", DP_DELIMITED, &has_type))
398 lex_match (lexer, T_EQUALS);
399 if (lex_match_id (lexer, "LINE"))
400 data_parser_set_span (parser, false);
401 else if (lex_match_id (lexer, "VARIABLES"))
403 data_parser_set_span (parser, true);
405 /* VARIABLES takes an integer argument, but for no
406 good reason. We just ignore it. */
407 if (!lex_force_int (lexer))
413 lex_error_expecting (lexer, "LINE", "VARIABLES", NULL_SENTINEL);
417 else if (lex_match_id (lexer, "FIXCASE"))
419 if (!set_type (parser, "FIXCASE", DP_FIXED, &has_type))
421 lex_match (lexer, T_EQUALS);
422 if (!lex_force_int (lexer))
424 if (lex_integer (lexer) < 1)
426 msg (SE, _("Value of FIXCASE must be at least 1."));
429 data_parser_set_records (parser, lex_integer (lexer));
432 else if (lex_match_id (lexer, "IMPORTCASES"))
434 lex_match (lexer, T_EQUALS);
435 if (lex_match (lexer, T_ALL))
437 data_parser_set_case_limit (parser, -1);
438 data_parser_set_case_percent (parser, 100);
440 else if (lex_match_id (lexer, "FIRST"))
442 if (!lex_force_int (lexer))
444 if (lex_integer (lexer) < 1)
446 msg (SE, _("Value of FIRST must be at least 1."));
449 data_parser_set_case_limit (parser, lex_integer (lexer));
452 else if (lex_match_id (lexer, "PERCENT"))
454 if (!lex_force_int (lexer))
456 if (lex_integer (lexer) < 1 || lex_integer (lexer) > 100)
458 msg (SE, _("Value of PERCENT must be between 1 and 100."));
461 data_parser_set_case_percent (parser, lex_integer (lexer));
465 else if (lex_match_id_n (lexer, "DELIMITERS", 4))
467 struct string hard_seps = DS_EMPTY_INITIALIZER;
468 const char *soft_seps = "";
472 if (!set_type (parser, "DELIMITERS", DP_DELIMITED, &has_type))
474 lex_match (lexer, T_EQUALS);
476 if (!lex_force_string (lexer))
479 /* XXX should support multibyte UTF-8 characters */
480 s = lex_tokss (lexer);
481 if (ss_match_string (&s, ss_cstr ("\\t")))
482 ds_put_cstr (&hard_seps, "\t");
483 if (ss_match_string (&s, ss_cstr ("\\\\")))
484 ds_put_cstr (&hard_seps, "\\");
485 while ((c = ss_get_byte (&s)) != EOF)
489 ds_put_byte (&hard_seps, c);
490 data_parser_set_soft_delimiters (parser, ss_cstr (soft_seps));
491 data_parser_set_hard_delimiters (parser, ds_ss (&hard_seps));
492 ds_destroy (&hard_seps);
496 else if (lex_match_id (lexer, "QUALIFIERS"))
498 if (!set_type (parser, "QUALIFIERS", DP_DELIMITED, &has_type))
500 lex_match (lexer, T_EQUALS);
502 if (!lex_force_string (lexer))
505 /* XXX should support multibyte UTF-8 characters */
506 if (settings_get_syntax () == COMPATIBLE
507 && ss_length (lex_tokss (lexer)) != 1)
509 msg (SE, _("In compatible syntax mode, the QUALIFIER string "
510 "must contain exactly one character."));
514 data_parser_set_quotes (parser, lex_tokss (lexer));
517 else if (settings_get_syntax () == ENHANCED
518 && lex_match_id (lexer, "ESCAPE"))
519 data_parser_set_quote_escape (parser, true);
520 else if (lex_match_id (lexer, "VARIABLES"))
524 lex_error_expecting (lexer, "VARIABLES", NULL_SENTINEL);
528 lex_match (lexer, T_EQUALS);
531 type = data_parser_get_type (parser);
534 struct fmt_spec input, output;
538 while (type == DP_FIXED && lex_match (lexer, T_SLASH))
540 if (!lex_force_int (lexer))
542 if (lex_integer (lexer) < record)
544 msg (SE, _("The record number specified, %ld, is at or "
545 "before the previous record, %d. Data "
546 "fields must be listed in order of "
547 "increasing record number."),
548 lex_integer (lexer), record);
551 if (lex_integer (lexer) > data_parser_get_records (parser))
553 msg (SE, _("The record number specified, %ld, exceeds "
554 "the number of records per case specified "
556 lex_integer (lexer), data_parser_get_records (parser));
559 record = lex_integer (lexer);
563 if (!lex_force_id (lexer)
564 || !dict_id_is_valid (dict, lex_tokcstr (lexer), true))
566 name = xstrdup (lex_tokcstr (lexer));
569 if (type == DP_DELIMITED)
571 if (!parse_format_specifier (lexer, &input)
572 || !fmt_check_input (&input))
575 output = fmt_for_output_from_input (&input);
579 char fmt_type_name[FMT_TYPE_LEN_MAX + 1];
580 enum fmt_type fmt_type;
583 if (!parse_column_range (lexer, 0, &fc, &lc, NULL))
586 /* Accept a format (e.g. F8.2) or just a type name (e.g. DOLLAR). */
587 if (!parse_abstract_format_specifier (lexer, fmt_type_name, &w, &d))
589 if (!fmt_from_name (fmt_type_name, &fmt_type))
591 msg (SE, _("Unknown format type `%s'."), fmt_type_name);
595 /* Compose input format. */
596 input.type = fmt_type;
597 input.w = lc - fc + 1;
599 if (!fmt_check_input (&input))
602 /* Compose output format. */
605 output.type = fmt_type;
608 if (!fmt_check_output (&output))
612 output = fmt_for_output_from_input (&input);
615 v = dict_create_var (dict, name, fmt_var_width (&input));
618 msg (SE, _("%s is a duplicate variable name."), name);
621 var_set_both_formats (v, &output);
623 if (type == DP_DELIMITED)
624 data_parser_add_delimited_field (parser, &input,
625 var_get_case_index (v),
628 data_parser_add_fixed_field (parser, &input, var_get_case_index (v),
633 while (lex_token (lexer) != T_ENDCMD);
635 reader = dfm_open_reader (fh, lexer, encoding);
639 data_parser_make_active_file (parser, ds, reader, dict);
645 data_parser_destroy (parser);
650 return CMD_CASCADING_FAILURE;
655 destroy_spreadsheet_read_info (struct spreadsheet_read_info *sri)
660 free (sri->sheet_name);
661 free (sri->cell_range);
662 free (sri->file_name);