1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2012, 2013, 2014 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/any-reader.h"
24 #include "data/attributes.h"
25 #include "data/casereader.h"
26 #include "data/dataset.h"
27 #include "data/dictionary.h"
28 #include "data/file-handle-def.h"
29 #include "data/format.h"
30 #include "data/missing-values.h"
31 #include "data/value-labels.h"
32 #include "data/variable.h"
33 #include "data/vector.h"
34 #include "language/command.h"
35 #include "language/commands/file-handle.h"
36 #include "language/lexer/lexer.h"
37 #include "language/lexer/macro.h"
38 #include "language/lexer/variable-parser.h"
39 #include "libpspp/array.h"
40 #include "libpspp/hash-functions.h"
41 #include "libpspp/i18n.h"
42 #include "libpspp/message.h"
43 #include "libpspp/misc.h"
44 #include "libpspp/pool.h"
45 #include "libpspp/string-array.h"
46 #include "output/pivot-table.h"
47 #include "output/output-item.h"
49 #include "gl/count-one-bits.h"
50 #include "gl/localcharset.h"
51 #include "gl/intprops.h"
52 #include "gl/minmax.h"
53 #include "gl/xalloc.h"
56 #define _(msgid) gettext (msgid)
57 #define N_(msgid) (msgid)
59 /* Information to include in displaying a dictionary. */
66 DF_MEASUREMENT_LEVEL = 1 << 3,
69 DF_ALIGNMENT = 1 << 6,
70 DF_PRINT_FORMAT = 1 << 7,
71 DF_WRITE_FORMAT = 1 << 8,
72 DF_MISSING_VALUES = 1 << 9,
73 #define DF_ALL_VARIABLE ((1 << 10) - 1)
75 /* Value labels table. */
76 DF_VALUE_LABELS = 1 << 10,
78 /* Attribute table. */
79 DF_AT_ATTRIBUTES = 1 << 11, /* Attributes whose names begin with @. */
80 DF_ATTRIBUTES = 1 << 12, /* All other attributes. */
83 static void display_variables (const struct variable **, size_t, int flags);
84 static void display_value_labels (const struct variable **, size_t);
85 static void display_attributes (const struct attrset *,
86 const struct variable **, size_t, int flags);
88 static void report_encodings (const struct file_handle *, struct pool *,
89 char **titles, bool *ids,
90 char **strings, size_t n_strings);
92 static char *get_documents_as_string (const struct dictionary *);
95 add_row (struct pivot_table *table, const char *attribute,
96 struct pivot_value *value)
98 int row = pivot_category_create_leaf (table->dimensions[0]->root,
99 pivot_value_new_text (attribute));
101 pivot_table_put1 (table, row, value);
104 /* SYSFILE INFO utility. */
106 cmd_sysfile_info (struct lexer *lexer, struct dataset *ds)
108 struct any_reader *any_reader;
109 struct file_handle *h;
110 struct dictionary *d;
111 struct casereader *reader;
112 struct any_read_info info;
119 lex_match (lexer, T_SLASH);
121 if (lex_match_id (lexer, "FILE") || lex_is_string (lexer))
123 lex_match (lexer, T_EQUALS);
126 h = fh_parse (lexer, FH_REF_FILE, NULL);
130 else if (lex_match_id (lexer, "ENCODING"))
132 lex_match (lexer, T_EQUALS);
134 if (!lex_force_string (lexer))
138 encoding = ss_xstrdup (lex_tokss (lexer));
148 lex_sbc_missing (lexer, "FILE");
152 any_reader = any_reader_open (h);
159 if (encoding && !strcasecmp (encoding, "detect"))
161 char **titles, **strings;
166 pool = pool_create ();
167 n_strings = any_reader_get_strings (any_reader, pool,
168 &titles, &ids, &strings);
169 any_reader_close (any_reader);
171 report_encodings (h, pool, titles, ids, strings, n_strings);
179 reader = any_reader_decode (any_reader, encoding, &d, &info);
182 casereader_destroy (reader);
184 struct pivot_table *table = pivot_table_create (N_("File Information"));
185 pivot_dimension_create (table, PIVOT_AXIS_ROW, N_("Attribute"));
187 add_row (table, N_("File"),
188 pivot_value_new_user_text (fh_get_file_name (h), -1));
190 const char *label = dict_get_label (d);
191 add_row (table, N_("Label"),
192 label ? pivot_value_new_user_text (label, -1) : NULL);
194 add_row (table, N_("Created"),
195 pivot_value_new_user_text_nocopy (
196 xasprintf ("%s %s by %s", info.creation_date,
197 info.creation_time, info.product)));
199 if (info.product_ext)
200 add_row (table, N_("Product"),
201 pivot_value_new_user_text (info.product_ext, -1));
203 add_row (table, N_("Integer Format"),
204 pivot_value_new_text (
205 info.integer_format == INTEGER_MSB_FIRST ? N_("Big Endian")
206 : info.integer_format == INTEGER_LSB_FIRST ? N_("Little Endian")
209 add_row (table, N_("Real Format"),
210 pivot_value_new_text (
211 info.float_format == FLOAT_IEEE_DOUBLE_LE ? N_("IEEE 754 LE.")
212 : info.float_format == FLOAT_IEEE_DOUBLE_BE ? N_("IEEE 754 BE.")
213 : info.float_format == FLOAT_VAX_D ? N_("VAX D.")
214 : info.float_format == FLOAT_VAX_G ? N_("VAX G.")
215 : info.float_format == FLOAT_Z_LONG ? N_("IBM 390 Hex Long.")
218 add_row (table, N_("Variables"),
219 pivot_value_new_integer (dict_get_n_vars (d)));
221 add_row (table, N_("Cases"),
223 ? pivot_value_new_text (N_("Unknown"))
224 : pivot_value_new_integer (info.n_cases)));
226 add_row (table, N_("Type"),
227 pivot_value_new_text (info.klass->name));
229 struct variable *weight_var = dict_get_weight (d);
230 add_row (table, N_("Weight"),
232 ? pivot_value_new_variable (weight_var)
233 : pivot_value_new_text (N_("Not weighted"))));
235 add_row (table, N_("Compression"),
236 (info.compression == ANY_COMP_NONE
237 ? pivot_value_new_text (N_("None"))
238 : pivot_value_new_user_text (
239 info.compression == ANY_COMP_SIMPLE ? "SAV" : "ZSAV", -1)));
241 add_row (table, N_("Encoding"),
242 pivot_value_new_user_text (dict_get_encoding (d), -1));
244 if (dict_get_document_n_lines (d) > 0)
245 add_row (table, N_("Documents"),
246 pivot_value_new_user_text_nocopy (get_documents_as_string (d)));
248 pivot_table_submit (table);
250 size_t n_vars = dict_get_n_vars (d);
251 const struct variable **vars = xnmalloc (n_vars, sizeof *vars);
252 for (size_t i = 0; i < dict_get_n_vars (d); i++)
253 vars[i] = dict_get_var (d, i);
254 display_variables (vars, n_vars, DF_ALL_VARIABLE);
255 display_value_labels (vars, n_vars);
256 display_attributes (dict_get_attributes (dataset_dict (ds)),
257 vars, n_vars, DF_ATTRIBUTES);
264 any_read_info_destroy (&info);
273 /* DISPLAY utility. */
275 static void display_documents (const struct dictionary *dict);
276 static void display_vectors (const struct dictionary *dict, int sorted);
279 cmd_display (struct lexer *lexer, struct dataset *ds)
281 /* Variables to display. */
283 const struct variable **vl;
285 if (lex_match_id (lexer, "DOCUMENTS"))
286 display_documents (dataset_dict (ds));
287 else if (lex_match_id (lexer, "FILE"))
289 if (!lex_force_match_id (lexer, "LABEL"))
292 const char *label = dict_get_label (dataset_dict (ds));
294 struct pivot_table *table = pivot_table_create (N_("File Label"));
295 pivot_dimension_create (table, PIVOT_AXIS_ROW, N_("Label"),
297 pivot_table_put1 (table, 0,
298 (label ? pivot_value_new_user_text (label, -1)
299 : pivot_value_new_text (N_("(none)"))));
300 pivot_table_submit (table);
306 bool sorted = lex_match_id (lexer, "SORTED");
308 if (lex_match_id (lexer, "VECTORS"))
310 display_vectors (dataset_dict(ds), sorted);
313 else if (lex_match_id (lexer, "SCRATCH"))
315 dict_get_vars (dataset_dict (ds), &vl, &n, DC_ORDINARY);
325 static const struct subcommand subcommands[] =
327 {"@ATTRIBUTES", DF_ATTRIBUTES | DF_AT_ATTRIBUTES},
328 {"ATTRIBUTES", DF_ATTRIBUTES},
329 {"DICTIONARY", (DF_NAME | DF_POSITION | DF_LABEL
330 | DF_MEASUREMENT_LEVEL | DF_ROLE | DF_WIDTH
331 | DF_ALIGNMENT | DF_PRINT_FORMAT
332 | DF_WRITE_FORMAT | DF_MISSING_VALUES
334 {"INDEX", DF_NAME | DF_POSITION},
335 {"LABELS", DF_NAME | DF_POSITION | DF_LABEL},
337 {"VARIABLES", (DF_NAME | DF_POSITION | DF_PRINT_FORMAT
338 | DF_WRITE_FORMAT | DF_MISSING_VALUES)},
341 const struct subcommand *sbc;
342 struct dictionary *dict = dataset_dict (ds);
345 for (sbc = subcommands; sbc->name != NULL; sbc++)
346 if (lex_match_id (lexer, sbc->name))
352 lex_match (lexer, T_SLASH);
353 lex_match_id (lexer, "VARIABLES");
354 lex_match (lexer, T_EQUALS);
356 if (lex_token (lexer) != T_ENDCMD)
358 if (!parse_variables_const (lexer, dict, &vl, &n, PV_NONE))
365 dict_get_vars (dict, &vl, &n, 0);
370 sort (vl, n, sizeof *vl, (sorted
371 ? compare_var_ptrs_by_name
372 : compare_var_ptrs_by_dict_index), NULL);
374 int variable_flags = flags & DF_ALL_VARIABLE;
376 display_variables (vl, n, variable_flags);
378 if (flags & DF_VALUE_LABELS)
379 display_value_labels (vl, n);
381 int attribute_flags = flags & (DF_ATTRIBUTES | DF_AT_ATTRIBUTES);
383 display_attributes (dict_get_attributes (dataset_dict (ds)),
384 vl, n, attribute_flags);
387 msg (SN, _("No variables to display."));
396 compare_macros_by_name (const void *a_, const void *b_, const void *aux UNUSED)
398 const struct macro *const *ap = a_;
399 const struct macro *const *bp = b_;
400 const struct macro *a = *ap;
401 const struct macro *b = *bp;
403 return utf8_strcasecmp (a->name, b->name);
407 cmd_display_macros (struct lexer *lexer, struct dataset *ds UNUSED)
409 const struct macro_set *set = lex_get_macros (lexer);
411 if (hmap_is_empty (&set->macros))
413 msg (SN, _("No macros to display."));
417 const struct macro **macros = xnmalloc (hmap_count (&set->macros),
420 const struct macro *m;
421 HMAP_FOR_EACH (m, struct macro, hmap_node, &set->macros)
423 assert (n == hmap_count (&set->macros));
424 sort (macros, n, sizeof *macros, compare_macros_by_name, NULL);
426 struct pivot_table *table = pivot_table_create (N_("Macros"));
428 struct pivot_dimension *attributes = pivot_dimension_create (
429 table, PIVOT_AXIS_COLUMN, N_("Attributes"));
430 pivot_category_create_leaf (attributes->root,
431 pivot_value_new_text (N_("Source Location")));
433 struct pivot_dimension *names = pivot_dimension_create (
434 table, PIVOT_AXIS_ROW, N_("Name"));
435 names->root->show_label = true;
437 for (size_t i = 0; i < n; i++)
439 const struct macro *m = macros[i];
441 pivot_category_create_leaf (names->root,
442 pivot_value_new_user_text (m->name, -1));
444 struct string location = DS_EMPTY_INITIALIZER;
445 msg_location_format (m->location, &location);
448 pivot_value_new_user_text_nocopy (ds_steal_cstr (&location)));
451 pivot_table_submit (table);
459 get_documents_as_string (const struct dictionary *dict)
461 const struct string_array *documents = dict_get_documents (dict);
462 struct string s = DS_EMPTY_INITIALIZER;
463 for (size_t i = 0; i < documents->n; i++)
466 ds_put_byte (&s, '\n');
467 ds_put_cstr (&s, documents->strings[i]);
469 return ds_steal_cstr (&s);
473 display_documents (const struct dictionary *dict)
475 struct pivot_table *table = pivot_table_create (N_("Documents"));
476 struct pivot_dimension *d = pivot_dimension_create (
477 table, PIVOT_AXIS_COLUMN, N_("Documents"), N_("Document"));
478 d->hide_all_labels = true;
480 if (!dict_get_documents (dict)->n)
481 pivot_table_put1 (table, 0, pivot_value_new_text (N_("(none)")));
484 char *docs = get_documents_as_string (dict);
485 pivot_table_put1 (table, 0, pivot_value_new_user_text_nocopy (docs));
488 pivot_table_submit (table);
492 display_variables (const struct variable **vl, size_t n, int flags)
494 struct pivot_table *table = pivot_table_create (N_("Variables"));
496 struct pivot_dimension *attributes = pivot_dimension_create (
497 table, PIVOT_AXIS_COLUMN, N_("Attributes"));
504 static const struct heading headings[] = {
505 { DF_POSITION, N_("Position") },
506 { DF_LABEL, N_("Label") },
507 { DF_MEASUREMENT_LEVEL, N_("Measurement Level") },
508 { DF_ROLE, N_("Role") },
509 { DF_WIDTH, N_("Width") },
510 { DF_ALIGNMENT, N_("Alignment") },
511 { DF_PRINT_FORMAT, N_("Print Format") },
512 { DF_WRITE_FORMAT, N_("Write Format") },
513 { DF_MISSING_VALUES, N_("Missing Values") },
515 for (size_t i = 0; i < sizeof headings / sizeof *headings; i++)
516 if (flags & headings[i].flag)
517 pivot_category_create_leaf (attributes->root,
518 pivot_value_new_text (headings[i].title));
520 struct pivot_dimension *names = pivot_dimension_create (
521 table, PIVOT_AXIS_ROW, N_("Name"));
522 names->root->show_label = true;
524 for (size_t i = 0; i < n; i++)
526 const struct variable *v = vl[i];
528 struct pivot_value *name = pivot_value_new_variable (v);
529 name->variable.show = SETTINGS_VALUE_SHOW_VALUE;
530 int row = pivot_category_create_leaf (names->root, name);
533 if (flags & DF_POSITION)
534 pivot_table_put2 (table, x++, row, pivot_value_new_integer (
535 var_get_dict_index (v) + 1));
537 if (flags & DF_LABEL)
539 const char *label = var_get_label (v);
541 pivot_table_put2 (table, x, row,
542 pivot_value_new_user_text (label, -1));
546 if (flags & DF_MEASUREMENT_LEVEL)
549 pivot_value_new_text (measure_to_string (var_get_measure (v))));
554 pivot_value_new_text (var_role_to_string (var_get_role (v))));
556 if (flags & DF_WIDTH)
559 pivot_value_new_integer (var_get_display_width (v)));
561 if (flags & DF_ALIGNMENT)
564 pivot_value_new_text (alignment_to_string (
565 var_get_alignment (v))));
567 if (flags & DF_PRINT_FORMAT)
569 struct fmt_spec print = var_get_print_format (v);
570 char s[FMT_STRING_LEN_MAX + 1];
574 pivot_value_new_user_text (fmt_to_string (print, s), -1));
577 if (flags & DF_WRITE_FORMAT)
579 struct fmt_spec write = var_get_write_format (v);
580 char s[FMT_STRING_LEN_MAX + 1];
584 pivot_value_new_user_text (fmt_to_string (write, s), -1));
587 if (flags & DF_MISSING_VALUES)
589 char *s = mv_to_string (var_get_missing_values (v),
590 var_get_encoding (v));
594 pivot_value_new_user_text_nocopy (s));
600 pivot_table_submit (table);
604 any_value_labels (const struct variable **vars, size_t n_vars)
606 for (size_t i = 0; i < n_vars; i++)
607 if (val_labs_count (var_get_value_labels (vars[i])))
613 display_value_labels (const struct variable **vars, size_t n_vars)
615 if (!any_value_labels (vars, n_vars))
618 struct pivot_table *table = pivot_table_create (N_("Value Labels"));
620 pivot_dimension_create (table, PIVOT_AXIS_COLUMN,
621 N_("Label"), N_("Label"));
623 struct pivot_dimension *values = pivot_dimension_create (
624 table, PIVOT_AXIS_ROW, N_("Variable Value"));
625 values->root->show_label = true;
627 struct pivot_footnote *missing_footnote = pivot_table_create_footnote (
628 table, pivot_value_new_text (N_("User-missing value")));
630 for (size_t i = 0; i < n_vars; i++)
632 const struct val_labs *val_labs = var_get_value_labels (vars[i]);
633 size_t n_labels = val_labs_count (val_labs);
637 struct pivot_category *group = pivot_category_create_group__ (
638 values->root, pivot_value_new_variable (vars[i]));
640 const struct val_lab **labels = val_labs_sorted (val_labs);
641 for (size_t j = 0; j < n_labels; j++)
643 const struct val_lab *vl = labels[j];
645 struct pivot_value *value = pivot_value_new_var_value (
646 vars[i], &vl->value);
647 if (value->type == PIVOT_VALUE_NUMERIC)
648 value->numeric.show = SETTINGS_VALUE_SHOW_VALUE;
650 value->string.show = SETTINGS_VALUE_SHOW_VALUE;
651 if (var_is_value_missing (vars[i], &vl->value) == MV_USER)
652 pivot_value_add_footnote (value, missing_footnote);
653 int row = pivot_category_create_leaf (group, value);
655 struct pivot_value *label = pivot_value_new_var_value (
656 vars[i], &vl->value);
657 char *escaped_label = xstrdup (val_lab_get_escaped_label (vl));
658 if (label->type == PIVOT_VALUE_NUMERIC)
660 free (label->numeric.value_label);
661 label->numeric.value_label = escaped_label;
662 label->numeric.show = SETTINGS_VALUE_SHOW_LABEL;
666 free (label->string.value_label);
667 label->string.value_label = escaped_label;
668 label->string.show = SETTINGS_VALUE_SHOW_LABEL;
670 pivot_table_put2 (table, 0, row, label);
674 pivot_table_submit (table);
678 is_at_name (const char *name)
680 return name[0] == '@' || (name[0] == '$' && name[1] == '@');
684 count_attributes (const struct attrset *set, int flags)
686 struct attrset_iterator i;
687 struct attribute *attr;
691 for (attr = attrset_first (set, &i); attr != NULL;
692 attr = attrset_next (set, &i))
693 if (flags & DF_AT_ATTRIBUTES || !is_at_name (attribute_get_name (attr)))
694 n_attrs += attribute_get_n_values (attr);
699 display_attrset (struct pivot_table *table, struct pivot_value *set_name,
700 const struct attrset *set, int flags)
702 size_t n_total = count_attributes (set, flags);
705 pivot_value_destroy (set_name);
709 struct pivot_category *group = pivot_category_create_group__ (
710 table->dimensions[1]->root, set_name);
712 size_t n_attrs = attrset_count (set);
713 struct attribute **attrs = attrset_sorted (set);
714 for (size_t i = 0; i < n_attrs; i++)
716 const struct attribute *attr = attrs[i];
717 const char *name = attribute_get_name (attr);
719 if (!(flags & DF_AT_ATTRIBUTES) && is_at_name (name))
722 size_t n_values = attribute_get_n_values (attr);
723 for (size_t j = 0; j < n_values; j++)
725 int row = pivot_category_create_leaf (
728 ? pivot_value_new_user_text_nocopy (xasprintf (
729 "%s[%zu]", name, j + 1))
730 : pivot_value_new_user_text (name, -1)));
731 pivot_table_put2 (table, 0, row,
732 pivot_value_new_user_text (
733 attribute_get_value (attr, j), -1));
740 display_attributes (const struct attrset *dict_attrset,
741 const struct variable **vars, size_t n_vars, int flags)
743 struct pivot_table *table = pivot_table_create (
744 N_("Variable and Dataset Attributes"));
746 pivot_dimension_create (table, PIVOT_AXIS_COLUMN,
747 N_("Value"), N_("Value"));
749 struct pivot_dimension *variables = pivot_dimension_create (
750 table, PIVOT_AXIS_ROW, N_("Variable and Name"));
751 variables->root->show_label = true;
753 display_attrset (table, pivot_value_new_text (N_("(dataset)")),
754 dict_attrset, flags);
755 for (size_t i = 0; i < n_vars; i++)
756 display_attrset (table, pivot_value_new_variable (vars[i]),
757 var_get_attributes (vars[i]), flags);
759 if (pivot_table_is_empty (table))
760 pivot_table_unref (table);
762 pivot_table_submit (table);
765 /* Display a list of vectors. If SORTED is nonzero then they are
766 sorted alphabetically. */
768 display_vectors (const struct dictionary *dict, int sorted)
770 size_t n_vectors = dict_get_n_vectors (dict);
773 msg (SN, _("No vectors defined."));
777 const struct vector **vectors = xnmalloc (n_vectors, sizeof *vectors);
778 for (size_t i = 0; i < n_vectors; i++)
779 vectors[i] = dict_get_vector (dict, i);
781 qsort (vectors, n_vectors, sizeof *vectors, compare_vector_ptrs_by_name);
783 struct pivot_table *table = pivot_table_create (N_("Vectors"));
784 pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Attributes"),
785 N_("Variable"), N_("Print Format"));
786 struct pivot_dimension *vector_dim = pivot_dimension_create (
787 table, PIVOT_AXIS_ROW, N_("Vector and Position"));
788 vector_dim->root->show_label = true;
790 for (size_t i = 0; i < n_vectors; i++)
792 const struct vector *vec = vectors[i];
794 struct pivot_category *group = pivot_category_create_group__ (
795 vector_dim->root, pivot_value_new_user_text (
796 vector_get_name (vectors[i]), -1));
798 for (size_t j = 0; j < vector_get_n_vars (vec); j++)
800 struct variable *var = vector_get_var (vec, j);
802 int row = pivot_category_create_leaf (
803 group, pivot_value_new_integer (j + 1));
805 pivot_table_put2 (table, 0, row, pivot_value_new_variable (var));
806 char fmt_string[FMT_STRING_LEN_MAX + 1];
807 fmt_to_string (var_get_print_format (var), fmt_string);
808 pivot_table_put2 (table, 1, row,
809 pivot_value_new_user_text (fmt_string, -1));
813 pivot_table_submit (table);
818 /* Encoding analysis. */
820 static const char *encoding_names[] = {
821 /* These encodings are from http://encoding.spec.whatwg.org/, as retrieved
822 February 2014. Encodings not supported by glibc and encodings relevant
823 only to HTML have been removed. */
857 /* Added by user request. */
861 #define N_ENCODING_NAMES (sizeof encoding_names / sizeof *encoding_names)
871 recode_strings (struct pool *pool,
872 char **strings, bool *ids, size_t n,
873 const char *encoding)
878 utf8_strings = pool_alloc (pool, n * sizeof *utf8_strings);
879 for (i = 0; i < n; i++)
881 struct substring utf8;
884 error = recode_pedantically ("UTF-8", encoding, ss_cstr (strings[i]),
888 ss_rtrim (&utf8, ss_cstr (" "));
889 utf8.string[utf8.length] = '\0';
891 if (ids[i] && !id_is_plausible (utf8.string))
898 utf8_strings[i] = utf8.string;
904 static struct encoding *
905 find_duplicate_encoding (struct encoding *encodings, size_t n_encodings,
906 char **utf8_strings, size_t n_strings,
911 for (e = encodings; e < &encodings[n_encodings]; e++)
918 for (i = 0; i < n_strings; i++)
919 if (strcmp (utf8_strings[i], e->utf8_strings[i]))
930 all_equal (const struct encoding *encodings, size_t n_encodings,
936 s0 = encodings[0].utf8_strings[string_idx];
937 for (i = 1; i < n_encodings; i++)
938 if (strcmp (s0, encodings[i].utf8_strings[string_idx]))
945 equal_prefix (const struct encoding *encodings, size_t n_encodings,
952 s0 = encodings[0].utf8_strings[string_idx];
953 prefix = strlen (s0);
954 for (i = 1; i < n_encodings; i++)
956 const char *si = encodings[i].utf8_strings[string_idx];
959 for (j = 0; j < prefix; j++)
969 while (prefix > 0 && s0[prefix - 1] != ' ')
975 equal_suffix (const struct encoding *encodings, size_t n_encodings,
983 s0 = encodings[0].utf8_strings[string_idx];
984 s0_len = strlen (s0);
986 for (i = 1; i < n_encodings; i++)
988 const char *si = encodings[i].utf8_strings[string_idx];
989 size_t si_len = strlen (si);
994 for (j = 0; j < suffix; j++)
995 if (s0[s0_len - j - 1] != si[si_len - j - 1])
1004 while (suffix > 0 && s0[s0_len - suffix] != ' ')
1010 report_encodings (const struct file_handle *h, struct pool *pool,
1011 char **titles, bool *ids, char **strings, size_t n_strings)
1013 struct encoding encodings[N_ENCODING_NAMES];
1014 size_t n_encodings, n_unique_strings;
1017 for (size_t i = 0; i < N_ENCODING_NAMES; i++)
1019 char **utf8_strings;
1023 utf8_strings = recode_strings (pool, strings, ids, n_strings,
1028 /* Hash utf8_strings. */
1030 for (size_t j = 0; j < n_strings; j++)
1031 hash = hash_string (utf8_strings[j], hash);
1033 /* If there's a duplicate encoding, just mark it. */
1034 e = find_duplicate_encoding (encodings, n_encodings,
1035 utf8_strings, n_strings, hash);
1038 e->encodings |= UINT64_C (1) << i;
1042 e = &encodings[n_encodings++];
1043 e->encodings = UINT64_C (1) << i;
1044 e->utf8_strings = utf8_strings;
1049 msg (SW, _("No valid encodings found."));
1053 /* Table of valid encodings. */
1054 struct pivot_table *table = pivot_table_create__ (
1055 pivot_value_new_text_format (N_("Usable encodings for %s."),
1056 fh_get_name (h)), "Usable Encodings");
1057 pivot_table_set_caption (
1058 table, pivot_value_new_text_format (
1059 N_("Encodings that can successfully read %s (by specifying the encoding "
1060 "name on the GET command's ENCODING subcommand). Encodings that "
1061 "yield identical text are listed together."),
1064 pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Encodings"),
1066 struct pivot_dimension *number = pivot_dimension_create__ (
1067 table, PIVOT_AXIS_ROW, pivot_value_new_user_text ("#", -1));
1068 number->root->show_label = true;
1070 for (size_t i = 0; i < n_encodings; i++)
1072 struct string s = DS_EMPTY_INITIALIZER;
1073 for (size_t j = 0; j < 64; j++)
1074 if (encodings[i].encodings & (UINT64_C (1) << j))
1075 ds_put_format (&s, "%s, ", encoding_names[j]);
1076 ds_chomp (&s, ss_cstr (", "));
1078 int row = pivot_category_create_leaf (number->root,
1079 pivot_value_new_integer (i + 1));
1081 table, 0, row, pivot_value_new_user_text_nocopy (ds_steal_cstr (&s)));
1083 pivot_table_submit (table);
1085 n_unique_strings = 0;
1086 for (size_t i = 0; i < n_strings; i++)
1087 if (!all_equal (encodings, n_encodings, i))
1089 if (!n_unique_strings)
1092 /* Table of alternative interpretations. */
1093 table = pivot_table_create__ (
1094 pivot_value_new_text_format (N_("%s Encoded Text Strings"),
1096 "Alternate Encoded Text Strings");
1097 pivot_table_set_caption (
1098 table, pivot_value_new_text (
1099 N_("Text strings in the file dictionary that the previously listed "
1100 "encodings interpret differently, along with the interpretations.")));
1102 pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Text"), N_("Text"));
1104 number = pivot_dimension_create__ (table, PIVOT_AXIS_ROW,
1105 pivot_value_new_user_text ("#", -1));
1106 number->root->show_label = true;
1107 for (size_t i = 0; i < n_encodings; i++)
1108 pivot_category_create_leaf (number->root,
1109 pivot_value_new_integer (i + 1));
1111 struct pivot_dimension *purpose = pivot_dimension_create (
1112 table, PIVOT_AXIS_ROW, N_("Purpose"));
1113 purpose->root->show_label = true;
1115 for (size_t i = 0; i < n_strings; i++)
1116 if (!all_equal (encodings, n_encodings, i))
1118 int prefix = equal_prefix (encodings, n_encodings, i);
1119 int suffix = equal_suffix (encodings, n_encodings, i);
1121 int purpose_idx = pivot_category_create_leaf (
1122 purpose->root, pivot_value_new_user_text (titles[i], -1));
1124 for (size_t j = 0; j < n_encodings; j++)
1126 const char *s = encodings[j].utf8_strings[i] + prefix;
1128 if (prefix || suffix)
1130 size_t len = strlen (s) - suffix;
1131 struct string entry;
1133 ds_init_empty (&entry);
1135 ds_put_cstr (&entry, "...");
1136 ds_put_substring (&entry, ss_buffer (s, len));
1138 ds_put_cstr (&entry, "...");
1140 pivot_table_put3 (table, 0, j, purpose_idx,
1141 pivot_value_new_user_text_nocopy (
1142 ds_steal_cstr (&entry)));
1145 pivot_table_put3 (table, 0, j, purpose_idx,
1146 pivot_value_new_user_text (s, -1));
1150 pivot_table_submit (table);