combined: Added the option "combine" to terminal and screen driver options.

Note that, combining characters work only on XTerm with UTF-8 locale setings.

src/config/options.inc

@@ -864,6 +864,11 @@ static struct option_info config_options_info[] = {
 		"only the subset of UTF-8 according to terminal codepage is used.\n"
 		"ELinks ignores this option if the terminal codepage is UTF-8.")),
 
+	INIT_OPT_BOOL("terminal._template_", N_("Combining characters"),
+		"combine", 0, 0,
+		N_("Enable combining characters. It works only with "
+		"the xterm in UTF-8 mode.")),
+
 	INIT_OPT_BOOL("terminal._template_", N_("Restrict frames in cp850/852"),
 		"restrict_852", 0, 0,
 		N_("Restrict the characters used when drawing lines. Makes sense\n"

src/dialogs/options.c

@@ -91,6 +91,7 @@ enum termopt {
 	TERM_OPT_UTF_8_IO,
 	TERM_OPT_TRANSPARENCY,
 	TERM_OPT_UNDERLINE,
+	TERM_OPT_COMBINE,
 
 	TERM_OPTIONS,
 };
@@ -104,6 +105,7 @@ static struct option_resolver resolvers[] = {
 	{ TERM_OPT_TRANSPARENCY, "transparency"	},
 	{ TERM_OPT_UTF_8_IO,	 "utf_8_io"	},
 	{ TERM_OPT_UNDERLINE,	 "underline"	},
+	{ TERM_OPT_COMBINE,	 "combine"	},
 };
 
 static widget_handler_status_T
@@ -149,7 +151,7 @@ push_save_button(struct dialog_data *dlg_data, struct widget_data *button)
 #define	RADIO_TRUE 0
 #endif
 
-#define TERMOPT_WIDGETS_COUNT (19 + RADIO_88 + RADIO_256 + RADIO_TRUE)
+#define TERMOPT_WIDGETS_COUNT (20 + RADIO_88 + RADIO_256 + RADIO_TRUE)
 
 #define TERM_OPTION_VALUE_SIZE (sizeof(union option_value) * TERM_OPTIONS)
 
@@ -230,6 +232,7 @@ terminal_options(struct terminal *term, void *xxx, struct session *ses)
 	add_dlg_checkbox(dlg, _("Transparency", term), &values[TERM_OPT_TRANSPARENCY].number);
 	add_dlg_checkbox(dlg, _("Underline", term), &values[TERM_OPT_UNDERLINE].number);
 	add_dlg_checkbox(dlg, _("UTF-8 I/O", term), &values[TERM_OPT_UTF_8_IO].number);
+	add_dlg_checkbox(dlg, _("Combining characters", term), &values[TERM_OPT_COMBINE].number);
 
 	add_dlg_button(dlg, _("~OK", term), B_ENTER, push_ok_button, NULL);
 	if (!anonymous)

src/intl/charsets.c

@@ -25,6 +25,7 @@
 #include "util/conv.h"
 #include "util/error.h"
 #include "util/fastfind.h"
+#include "util/hash.h"
 #include "util/memory.h"
 #include "util/string.h"
 
@@ -769,9 +770,68 @@ cp_to_unicode(int codepage, unsigned char **string, unsigned char *end)
 	++*string;
 	return ret;
 }
-#endif	/* CONFIG_UTF8 */
 
 
+unicode_val_T last_combined = UCS_BEGIN_COMBINED - 1;
+unicode_val_T **combined;
+struct hash *combined_hash;
+
+unicode_val_T
+get_combined(unicode_val_T *data, int length)
+{
+	struct hash_item *item;
+	unicode_val_T *key;
+	int i, indeks;
+
+	assert(length >= 1 && length <= UCS_MAX_LENGTH_COMBINED);
+	if_assert_failed return UCS_NO_CHAR;
+
+	if (!combined_hash) combined_hash = init_hash8();
+	if (!combined_hash) return UCS_NO_CHAR;
+	item = get_hash_item(combined_hash, (unsigned char *)data, length * sizeof(*data));
+
+	if (item) return (unicode_val_T)(long)item->value;
+	if (last_combined >= UCS_END_COMBINED) return UCS_NO_CHAR;
+
+	key = mem_alloc((length + 1) * sizeof(*key));
+	if (!key) return UCS_NO_CHAR;
+	for (i = 0; i < length; i++)
+		key[i] = data[i];
+	key[i] = UCS_END_COMBINED;
+
+	last_combined++;
+	indeks = last_combined - UCS_BEGIN_COMBINED;
+
+	combined = mem_realloc(combined, sizeof(*combined) * (indeks + 1));
+	if (!combined) {
+		mem_free(key);
+		last_combined--;
+		return UCS_NO_CHAR;
+	}
+	combined[indeks] = key;
+	item = add_hash_item(combined_hash, (unsigned char *)key,
+			     length * sizeof(*data), (void *)(long)(last_combined));
+	if (!item) {
+		last_combined--;
+		mem_free(key);
+		return UCS_NO_CHAR;
+	}
+	return last_combined;
+}
+
+void
+free_combined()
+{
+	int i, end = last_combined - UCS_BEGIN_COMBINED + 1; 
+
+	if (combined_hash)
+		free_hash(&combined_hash);
+	for (i = 0; i < end; i++)
+		mem_free(combined[i]);
+	mem_free_if(combined);
+}
+#endif	/* CONFIG_UTF8 */
+
 static void
 add_utf8(struct conv_table *ct, unicode_val_T u, const unsigned char *str)
 {
@@ -1480,3 +1540,4 @@ is_cp_utf8(int cp_index)
 	cp_index &= ~SYSTEM_CHARSET_FLAG;
 	return is_cp_ptr_utf8(&codepages[cp_index]);
 }
+

src/intl/charsets.h

@@ -1,6 +1,7 @@
 #ifndef EL__INTL_CHARSETS_H
 #define EL__INTL_CHARSETS_H
 
+struct hash;
 typedef uint32_t unicode_val_T;
 
 /* U+0020 SPACE.  Normally the same as ' ' or L' ' but perhaps ELinks
@@ -26,6 +27,13 @@ typedef uint32_t unicode_val_T;
  * for the second cell of a double-cell character.  */
 #define UCS_NO_CHAR ((unicode_val_T) 0xFFFFFFFD)
 
+#define UCS_END_COMBINED ((unicode_val_T) 0xFFFFFFFC)
+
+#define UCS_BEGIN_COMBINED ((unicode_val_T) (UCS_END_COMBINED - (unicode_val_T) 10000))
+
+/* Base character and up to 5 combining characters. */
+#define UCS_MAX_LENGTH_COMBINED 6
+
 /* If ELinks should display a double-cell character but there is only
  * one cell available, it displays this character instead.  This must
  * be a single-cell character but need not be unique.  Possible values
@@ -146,6 +154,14 @@ unicode_val_T unicode_fold_label_case(unicode_val_T);
 inline int strlen_utf8(unsigned char **);
 inline unicode_val_T utf8_to_unicode(unsigned char **, const unsigned char *);
 unicode_val_T cp_to_unicode(int, unsigned char **, unsigned char *);
+
+
+extern unicode_val_T last_combined;
+extern unicode_val_T **combined;
+extern struct hash *combined_hash;
+unicode_val_T get_combined(unicode_val_T *, int);
+void free_combined();
+
 #endif /* CONFIG_UTF8 */
 
 unicode_val_T cp2u(int, unsigned char);

src/main/main.c

@@ -49,6 +49,7 @@
 #include "util/color.h"
 #include "util/error.h"
 #include "util/file.h"
+#include "util/hash.h"
 #include "util/memdebug.h"
 #include "util/memory.h"
 #include "viewer/dump/dump.h"
@@ -302,6 +303,9 @@ terminate_all_subsystems(void)
 	done_options();
 	done_event();
 	terminate_osdep();
+#ifdef CONFIG_UTF8
+	free_combined();
+#endif
 }
 
 void

src/terminal/screen.c

@@ -223,6 +223,9 @@ struct screen_driver {
 		 * is the same as is_cp_utf8(charsets[0]), except the
 		 * latter might crash if UTF-8 I/O is disabled.  */
 		unsigned int utf8_cp:1;
+
+		/* Whether the terminal supports combining characters. */
+		unsigned int combine:1;
 #endif /* CONFIG_UTF8 */
 	} opt;
 
@@ -240,6 +243,7 @@ static const struct screen_driver_opt dumb_screen_driver_opt = {
 	/* transparent: */	1,
 #ifdef CONFIG_UTF8
 	/* utf8_cp: */		0,
+	/* combine */		0,
 #endif /* CONFIG_UTF8 */
 };
 
@@ -253,6 +257,7 @@ static const struct screen_driver_opt vt100_screen_driver_opt = {
 	/* transparent: */	1,
 #ifdef CONFIG_UTF8
 	/* utf8_cp: */		0,
+	/* combine */		0,
 #endif /* CONFIG_UTF8 */
 };
 
@@ -266,6 +271,7 @@ static const struct screen_driver_opt linux_screen_driver_opt = {
 	/* transparent: */	1,
 #ifdef CONFIG_UTF8
 	/* utf8_cp: */		0,
+	/* combine */		0,
 #endif /* CONFIG_UTF8 */
 };
 
@@ -279,6 +285,7 @@ static const struct screen_driver_opt koi8_screen_driver_opt = {
 	/* transparent: */	1,
 #ifdef CONFIG_UTF8
 	/* utf8_cp: */		0,
+	/* combine */		0,
 #endif /* CONFIG_UTF8 */
 };
 
@@ -292,6 +299,7 @@ static const struct screen_driver_opt freebsd_screen_driver_opt = {
 	/* transparent: */	1,
 #ifdef CONFIG_UTF8
 	/* utf8_cp: */		0,
+	/* combine */		0,
 #endif /* CONFIG_UTF8 */
 };
 
@@ -325,6 +333,7 @@ set_screen_driver_opt(struct screen_driver *driver, struct option *term_spec)
 	copy_struct(&driver->opt, screen_driver_opts[driver->type]);
 
 #ifdef CONFIG_UTF8
+	driver->opt.combine = get_opt_bool_tree(term_spec, "combine", NULL);
 	/* Force UTF-8 I/O if the UTF-8 charset is selected.  Various
 	 * places assume that the terminal's charset is unibyte if
 	 * UTF-8 I/O is disabled.  (bug 827) */
@@ -639,6 +648,21 @@ add_char_data(struct string *screen, struct screen_driver *driver,
 		}
 		if (data == UCS_NO_CHAR)
 			return;
+		if (data >= UCS_BEGIN_COMBINED && data <= last_combined) {
+			unicode_val_T *text = combined[data - UCS_BEGIN_COMBINED];
+
+			if (driver->opt.combine) {
+				/* XTerm */
+				while (*text != UCS_END_COMBINED) {
+					add_to_string(screen, encode_utf8(*text));
+					text++;
+				}
+				return;
+			} else {
+				/* Others */
+				data = *text;
+			}
+		}
 		if (!isscreensafe_ucs(data))
 			data = UCS_SPACE;
 		add_to_string(screen, encode_utf8(data));