+static int
+equal_suffix (const struct encoding *encodings, size_t n_encodings,
+ size_t string_idx)
+{
+ const char *s0;
+ size_t s0_len;
+ size_t suffix;
+ size_t i;
+
+ s0 = encodings[0].utf8_strings[string_idx];
+ s0_len = strlen (s0);
+ suffix = s0_len;
+ for (i = 1; i < n_encodings; i++)
+ {
+ const char *si = encodings[i].utf8_strings[string_idx];
+ size_t si_len = strlen (si);
+ size_t j;
+
+ if (si_len < suffix)
+ suffix = si_len;
+ for (j = 0; j < suffix; j++)
+ if (s0[s0_len - j - 1] != si[si_len - j - 1])
+ {
+ suffix = j;
+ if (!suffix)
+ return 0;
+ break;
+ }
+ }
+
+ while (suffix > 0 && s0[s0_len - suffix] != ' ')
+ suffix--;
+ return suffix;
+}
+
+static void
+report_encodings (const struct file_handle *h, struct pool *pool,
+ char **titles, bool *ids, char **strings, size_t n_strings)
+{
+ struct encoding encodings[N_ENCODING_NAMES];
+ size_t n_encodings, n_unique_strings;
+ size_t i, j;
+ struct tab_table *t;
+ size_t row;
+
+ n_encodings = 0;
+ for (i = 0; i < N_ENCODING_NAMES; i++)
+ {
+ char **utf8_strings;
+ struct encoding *e;
+ unsigned int hash;
+
+ utf8_strings = recode_strings (pool, strings, ids, n_strings,
+ encoding_names[i]);
+ if (!utf8_strings)
+ continue;
+
+ /* Hash utf8_strings. */
+ hash = 0;
+ for (j = 0; j < n_strings; j++)
+ hash = hash_string (utf8_strings[j], hash);
+
+ /* If there's a duplicate encoding, just mark it. */
+ e = find_duplicate_encoding (encodings, n_encodings,
+ utf8_strings, n_strings, hash);
+ if (e)
+ {
+ e->encodings |= UINT64_C (1) << i;
+ continue;
+ }
+
+ e = &encodings[n_encodings++];
+ e->encodings = UINT64_C (1) << i;
+ e->utf8_strings = utf8_strings;
+ e->hash = hash;
+ }
+ if (!n_encodings)
+ {
+ msg (SW, _("No valid encodings found."));
+ return;
+ }
+
+ t = tab_create (2, n_encodings + 1);
+ tab_title (t, _("Usable encodings for %s."), fh_get_name (h));
+ tab_caption (t, _("Encodings that can successfully read %s (by specifying "
+ "the encoding name on the GET command's ENCODING "
+ "subcommand). Encodings that yield identical text are "
+ "listed together."), fh_get_name (h));
+ tab_headers (t, 1, 0, 1, 0);
+ tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, 1, n_encodings);
+ tab_hline (t, TAL_1, 0, 1, 1);
+ tab_text (t, 0, 0, TAB_RIGHT, "#");
+ tab_text (t, 1, 0, TAB_LEFT, _("Encodings"));
+ for (i = 0; i < n_encodings; i++)
+ {
+ struct string s;
+
+ ds_init_empty (&s);
+ for (j = 0; j < 64; j++)
+ if (encodings[i].encodings & (UINT64_C (1) << j))
+ ds_put_format (&s, "%s, ", encoding_names[j]);
+ ds_chomp (&s, ss_cstr (", "));
+
+ tab_text_format (t, 0, i + 1, TAB_RIGHT, "%zu", i + 1);
+ tab_text (t, 1, i + 1, TAB_LEFT, ds_cstr (&s));
+ ds_destroy (&s);
+ }
+ tab_submit (t);
+
+ n_unique_strings = 0;
+ for (i = 0; i < n_strings; i++)
+ if (!all_equal (encodings, n_encodings, i))
+ n_unique_strings++;
+ if (!n_unique_strings)
+ return;
+
+ t = tab_create (3, (n_encodings * n_unique_strings) + 1);
+ tab_title (t, _("%s encoded text strings."), fh_get_name (h));
+ tab_caption (t, _("Text strings in the file dictionary that the previously "
+ "listed encodings interpret differently, along with the "
+ "interpretations."));
+ tab_headers (t, 1, 0, 1, 0);
+ tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, 2, n_encodings * n_unique_strings);
+ tab_hline (t, TAL_1, 0, 2, 1);
+
+ tab_text (t, 0, 0, TAB_LEFT, _("Purpose"));
+ tab_text (t, 1, 0, TAB_RIGHT, "#");
+ tab_text (t, 2, 0, TAB_LEFT, _("Text"));
+
+ row = 1;
+ for (i = 0; i < n_strings; i++)
+ if (!all_equal (encodings, n_encodings, i))
+ {
+ int prefix = equal_prefix (encodings, n_encodings, i);
+ int suffix = equal_suffix (encodings, n_encodings, i);
+
+ tab_joint_text (t, 0, row, 0, row + n_encodings - 1,
+ TAB_LEFT, titles[i]);
+ tab_hline (t, TAL_1, 0, 2, row);
+ for (j = 0; j < n_encodings; j++)
+ {
+ const char *s = encodings[j].utf8_strings[i] + prefix;
+
+ tab_text_format (t, 1, row, TAB_RIGHT, "%zu", j + 1);
+ if (prefix || suffix)
+ {
+ size_t len = strlen (s) - suffix;
+ struct string entry;
+
+ ds_init_empty (&entry);
+ if (prefix)
+ ds_put_cstr (&entry, "...");
+ ds_put_substring (&entry, ss_buffer (s, len));
+ if (suffix)
+ ds_put_cstr (&entry, "...");
+ tab_text (t, 2, row, TAB_LEFT, ds_cstr (&entry));
+ }
+ else
+ tab_text (t, 2, row, TAB_LEFT, s);
+ row++;
+ }
+ }
+ tab_submit (t);
+}