diff options
Diffstat (limited to 'java/src')
18 files changed, 2161 insertions, 917 deletions
diff --git a/java/src/com/android/inputmethod/keyboard/internal/KeyboardTextsSet.java b/java/src/com/android/inputmethod/keyboard/internal/KeyboardTextsSet.java index e769e3cdd..98515c893 100644 --- a/java/src/com/android/inputmethod/keyboard/internal/KeyboardTextsSet.java +++ b/java/src/com/android/inputmethod/keyboard/internal/KeyboardTextsSet.java @@ -147,111 +147,117 @@ public final class KeyboardTextsSet { /* 42 */ "keylabel_for_south_slavic_row3_8", /* 43 */ "more_keys_for_cyrillic_ie", /* 44 */ "more_keys_for_cyrillic_i", - /* 45 */ "label_to_alpha_key", - /* 46 */ "single_quotes", - /* 47 */ "double_quotes", - /* 48 */ "single_angle_quotes", - /* 49 */ "double_angle_quotes", - /* 50 */ "more_keys_for_currency_dollar", - /* 51 */ "keylabel_for_currency", - /* 52 */ "more_keys_for_currency", - /* 53 */ "more_keys_for_punctuation", - /* 54 */ "more_keys_for_star", - /* 55 */ "more_keys_for_bullet", - /* 56 */ "more_keys_for_plus", - /* 57 */ "more_keys_for_left_parenthesis", - /* 58 */ "more_keys_for_right_parenthesis", - /* 59 */ "more_keys_for_less_than", - /* 60 */ "more_keys_for_greater_than", - /* 61 */ "more_keys_for_arabic_diacritics", - /* 62 */ "keyhintlabel_for_arabic_diacritics", - /* 63 */ "keylabel_for_symbols_1", - /* 64 */ "keylabel_for_symbols_2", - /* 65 */ "keylabel_for_symbols_3", - /* 66 */ "keylabel_for_symbols_4", - /* 67 */ "keylabel_for_symbols_5", - /* 68 */ "keylabel_for_symbols_6", - /* 69 */ "keylabel_for_symbols_7", - /* 70 */ "keylabel_for_symbols_8", - /* 71 */ "keylabel_for_symbols_9", - /* 72 */ "keylabel_for_symbols_0", - /* 73 */ "label_to_symbol_key", - /* 74 */ "label_to_symbol_with_microphone_key", - /* 75 */ "additional_more_keys_for_symbols_1", - /* 76 */ "additional_more_keys_for_symbols_2", - /* 77 */ "additional_more_keys_for_symbols_3", - /* 78 */ "additional_more_keys_for_symbols_4", - /* 79 */ "additional_more_keys_for_symbols_5", - /* 80 */ "additional_more_keys_for_symbols_6", - /* 81 */ "additional_more_keys_for_symbols_7", - /* 82 */ "additional_more_keys_for_symbols_8", - /* 83 */ "additional_more_keys_for_symbols_9", - /* 84 */ "additional_more_keys_for_symbols_0", - /* 85 */ "more_keys_for_symbols_1", - /* 86 */ "more_keys_for_symbols_2", - /* 87 */ "more_keys_for_symbols_3", - /* 88 */ "more_keys_for_symbols_4", - /* 89 */ "more_keys_for_symbols_5", - /* 90 */ "more_keys_for_symbols_6", - /* 91 */ "more_keys_for_symbols_7", - /* 92 */ "more_keys_for_symbols_8", - /* 93 */ "more_keys_for_symbols_9", - /* 94 */ "more_keys_for_symbols_0", - /* 95 */ "keylabel_for_comma", - /* 96 */ "more_keys_for_comma", - /* 97 */ "keylabel_for_symbols_question", - /* 98 */ "keylabel_for_symbols_semicolon", - /* 99 */ "keylabel_for_symbols_percent", - /* 100 */ "more_keys_for_symbols_exclamation", - /* 101 */ "more_keys_for_symbols_question", - /* 102 */ "more_keys_for_symbols_semicolon", - /* 103 */ "more_keys_for_symbols_percent", - /* 104 */ "keylabel_for_tablet_comma", - /* 105 */ "keyhintlabel_for_tablet_comma", - /* 106 */ "more_keys_for_tablet_comma", - /* 107 */ "keyhintlabel_for_period", - /* 108 */ "more_keys_for_period", - /* 109 */ "keylabel_for_apostrophe", - /* 110 */ "keyhintlabel_for_apostrophe", - /* 111 */ "more_keys_for_apostrophe", - /* 112 */ "more_keys_for_q", - /* 113 */ "more_keys_for_x", - /* 114 */ "keylabel_for_q", - /* 115 */ "keylabel_for_w", - /* 116 */ "keylabel_for_y", - /* 117 */ "keylabel_for_x", - /* 118 */ "keylabel_for_spanish_row2_10", - /* 119 */ "more_keys_for_am_pm", - /* 120 */ "settings_as_more_key", - /* 121 */ "shortcut_as_more_key", - /* 122 */ "action_next_as_more_key", - /* 123 */ "action_previous_as_more_key", - /* 124 */ "label_to_more_symbol_key", - /* 125 */ "label_to_more_symbol_for_tablet_key", - /* 126 */ "label_tab_key", - /* 127 */ "label_to_phone_numeric_key", - /* 128 */ "label_to_phone_symbols_key", - /* 129 */ "label_time_am", - /* 130 */ "label_time_pm", - /* 131 */ "keylabel_for_popular_domain", - /* 132 */ "more_keys_for_popular_domain", - /* 133 */ "more_keys_for_smiley", - /* 134 */ "single_laqm_raqm", - /* 135 */ "single_laqm_raqm_rtl", - /* 136 */ "single_raqm_laqm", - /* 137 */ "double_laqm_raqm", - /* 138 */ "double_laqm_raqm_rtl", - /* 139 */ "double_raqm_laqm", - /* 140 */ "single_lqm_rqm", - /* 141 */ "single_9qm_lqm", - /* 142 */ "single_9qm_rqm", - /* 143 */ "double_lqm_rqm", - /* 144 */ "double_9qm_lqm", - /* 145 */ "double_9qm_rqm", - /* 146 */ "more_keys_for_single_quote", - /* 147 */ "more_keys_for_double_quote", - /* 148 */ "more_keys_for_tablet_double_quote", - /* 149 */ "emoji_key_as_more_key", + /* 45 */ "keylabel_for_swiss_row1_11", + /* 46 */ "keylabel_for_swiss_row2_10", + /* 47 */ "keylabel_for_swiss_row2_11", + /* 48 */ "more_keys_for_swiss_row1_11", + /* 49 */ "more_keys_for_swiss_row2_10", + /* 50 */ "more_keys_for_swiss_row2_11", + /* 51 */ "label_to_alpha_key", + /* 52 */ "single_quotes", + /* 53 */ "double_quotes", + /* 54 */ "single_angle_quotes", + /* 55 */ "double_angle_quotes", + /* 56 */ "more_keys_for_currency_dollar", + /* 57 */ "keylabel_for_currency", + /* 58 */ "more_keys_for_currency", + /* 59 */ "more_keys_for_punctuation", + /* 60 */ "more_keys_for_star", + /* 61 */ "more_keys_for_bullet", + /* 62 */ "more_keys_for_plus", + /* 63 */ "more_keys_for_left_parenthesis", + /* 64 */ "more_keys_for_right_parenthesis", + /* 65 */ "more_keys_for_less_than", + /* 66 */ "more_keys_for_greater_than", + /* 67 */ "more_keys_for_arabic_diacritics", + /* 68 */ "keyhintlabel_for_arabic_diacritics", + /* 69 */ "keylabel_for_symbols_1", + /* 70 */ "keylabel_for_symbols_2", + /* 71 */ "keylabel_for_symbols_3", + /* 72 */ "keylabel_for_symbols_4", + /* 73 */ "keylabel_for_symbols_5", + /* 74 */ "keylabel_for_symbols_6", + /* 75 */ "keylabel_for_symbols_7", + /* 76 */ "keylabel_for_symbols_8", + /* 77 */ "keylabel_for_symbols_9", + /* 78 */ "keylabel_for_symbols_0", + /* 79 */ "label_to_symbol_key", + /* 80 */ "label_to_symbol_with_microphone_key", + /* 81 */ "additional_more_keys_for_symbols_1", + /* 82 */ "additional_more_keys_for_symbols_2", + /* 83 */ "additional_more_keys_for_symbols_3", + /* 84 */ "additional_more_keys_for_symbols_4", + /* 85 */ "additional_more_keys_for_symbols_5", + /* 86 */ "additional_more_keys_for_symbols_6", + /* 87 */ "additional_more_keys_for_symbols_7", + /* 88 */ "additional_more_keys_for_symbols_8", + /* 89 */ "additional_more_keys_for_symbols_9", + /* 90 */ "additional_more_keys_for_symbols_0", + /* 91 */ "more_keys_for_symbols_1", + /* 92 */ "more_keys_for_symbols_2", + /* 93 */ "more_keys_for_symbols_3", + /* 94 */ "more_keys_for_symbols_4", + /* 95 */ "more_keys_for_symbols_5", + /* 96 */ "more_keys_for_symbols_6", + /* 97 */ "more_keys_for_symbols_7", + /* 98 */ "more_keys_for_symbols_8", + /* 99 */ "more_keys_for_symbols_9", + /* 100 */ "more_keys_for_symbols_0", + /* 101 */ "keylabel_for_comma", + /* 102 */ "more_keys_for_comma", + /* 103 */ "keylabel_for_symbols_question", + /* 104 */ "keylabel_for_symbols_semicolon", + /* 105 */ "keylabel_for_symbols_percent", + /* 106 */ "more_keys_for_symbols_exclamation", + /* 107 */ "more_keys_for_symbols_question", + /* 108 */ "more_keys_for_symbols_semicolon", + /* 109 */ "more_keys_for_symbols_percent", + /* 110 */ "keylabel_for_tablet_comma", + /* 111 */ "keyhintlabel_for_tablet_comma", + /* 112 */ "more_keys_for_tablet_comma", + /* 113 */ "keyhintlabel_for_period", + /* 114 */ "more_keys_for_period", + /* 115 */ "keylabel_for_apostrophe", + /* 116 */ "keyhintlabel_for_apostrophe", + /* 117 */ "more_keys_for_apostrophe", + /* 118 */ "more_keys_for_q", + /* 119 */ "more_keys_for_x", + /* 120 */ "keylabel_for_q", + /* 121 */ "keylabel_for_w", + /* 122 */ "keylabel_for_y", + /* 123 */ "keylabel_for_x", + /* 124 */ "keylabel_for_spanish_row2_10", + /* 125 */ "more_keys_for_am_pm", + /* 126 */ "settings_as_more_key", + /* 127 */ "shortcut_as_more_key", + /* 128 */ "action_next_as_more_key", + /* 129 */ "action_previous_as_more_key", + /* 130 */ "label_to_more_symbol_key", + /* 131 */ "label_to_more_symbol_for_tablet_key", + /* 132 */ "label_tab_key", + /* 133 */ "label_to_phone_numeric_key", + /* 134 */ "label_to_phone_symbols_key", + /* 135 */ "label_time_am", + /* 136 */ "label_time_pm", + /* 137 */ "keylabel_for_popular_domain", + /* 138 */ "more_keys_for_popular_domain", + /* 139 */ "more_keys_for_smiley", + /* 140 */ "single_laqm_raqm", + /* 141 */ "single_laqm_raqm_rtl", + /* 142 */ "single_raqm_laqm", + /* 143 */ "double_laqm_raqm", + /* 144 */ "double_laqm_raqm_rtl", + /* 145 */ "double_raqm_laqm", + /* 146 */ "single_lqm_rqm", + /* 147 */ "single_9qm_lqm", + /* 148 */ "single_9qm_rqm", + /* 149 */ "double_lqm_rqm", + /* 150 */ "double_9qm_lqm", + /* 151 */ "double_9qm_rqm", + /* 152 */ "more_keys_for_single_quote", + /* 153 */ "more_keys_for_double_quote", + /* 154 */ "more_keys_for_tablet_double_quote", + /* 155 */ "emoji_key_as_more_key", }; private static final String EMPTY = ""; @@ -262,145 +268,145 @@ public final class KeyboardTextsSet { EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, - EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, - /* ~44 */ + EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, + /* ~50 */ // Label for "switch to alphabetic" key. - /* 45 */ "ABC", - /* 46 */ "!text/single_lqm_rqm", - /* 47 */ "!text/double_lqm_rqm", - /* 48 */ "!text/single_laqm_raqm", - /* 49 */ "!text/double_laqm_raqm", + /* 51 */ "ABC", + /* 52 */ "!text/single_lqm_rqm", + /* 53 */ "!text/double_lqm_rqm", + /* 54 */ "!text/single_laqm_raqm", + /* 55 */ "!text/double_laqm_raqm", // U+00A2: "¢" CENT SIGN // U+00A3: "£" POUND SIGN // U+20AC: "€" EURO SIGN // U+00A5: "¥" YEN SIGN // U+20B1: "₱" PESO SIGN - /* 50 */ "\u00A2,\u00A3,\u20AC,\u00A5,\u20B1", - /* 51 */ "$", - /* 52 */ "$,\u00A2,\u20AC,\u00A3,\u00A5,\u20B1", - /* 53 */ "!fixedColumnOrder!4,#,!,\\,,?,-,:,',@", + /* 56 */ "\u00A2,\u00A3,\u20AC,\u00A5,\u20B1", + /* 57 */ "$", + /* 58 */ "$,\u00A2,\u20AC,\u00A3,\u00A5,\u20B1", + /* 59 */ "!fixedColumnOrder!4,#,!,\\,,?,-,:,',@", // U+2020: "†" DAGGER // U+2021: "‡" DOUBLE DAGGER // U+2605: "★" BLACK STAR - /* 54 */ "\u2020,\u2021,\u2605", + /* 60 */ "\u2020,\u2021,\u2605", // U+266A: "♪" EIGHTH NOTE // U+2665: "♥" BLACK HEART SUIT // U+2660: "♠" BLACK SPADE SUIT // U+2666: "♦" BLACK DIAMOND SUIT // U+2663: "♣" BLACK CLUB SUIT - /* 55 */ "\u266A,\u2665,\u2660,\u2666,\u2663", + /* 61 */ "\u266A,\u2665,\u2660,\u2666,\u2663", // U+00B1: "±" PLUS-MINUS SIGN - /* 56 */ "\u00B1", + /* 62 */ "\u00B1", // The all letters need to be mirrored are found at // http://www.unicode.org/Public/6.1.0/ucd/BidiMirroring.txt - /* 57 */ "!fixedColumnOrder!3,<,{,[", - /* 58 */ "!fixedColumnOrder!3,>,},]", + /* 63 */ "!fixedColumnOrder!3,<,{,[", + /* 64 */ "!fixedColumnOrder!3,>,},]", // U+2039: "‹" SINGLE LEFT-POINTING ANGLE QUOTATION MARK // U+203A: "›" SINGLE RIGHT-POINTING ANGLE QUOTATION MARK // U+2264: "≤" LESS-THAN OR EQUAL TO // U+2265: "≥" GREATER-THAN EQUAL TO // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK // U+00BB: "»" RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK - /* 59 */ "!fixedColumnOrder!3,\u2039,\u2264,\u00AB", - /* 60 */ "!fixedColumnOrder!3,\u203A,\u2265,\u00BB", - /* 61 */ EMPTY, - /* 62 */ EMPTY, - /* 63 */ "1", - /* 64 */ "2", - /* 65 */ "3", - /* 66 */ "4", - /* 67 */ "5", - /* 68 */ "6", - /* 69 */ "7", - /* 70 */ "8", - /* 71 */ "9", - /* 72 */ "0", + /* 65 */ "!fixedColumnOrder!3,\u2039,\u2264,\u00AB", + /* 66 */ "!fixedColumnOrder!3,\u203A,\u2265,\u00BB", + /* 67 */ EMPTY, + /* 68 */ EMPTY, + /* 69 */ "1", + /* 70 */ "2", + /* 71 */ "3", + /* 72 */ "4", + /* 73 */ "5", + /* 74 */ "6", + /* 75 */ "7", + /* 76 */ "8", + /* 77 */ "9", + /* 78 */ "0", // Label for "switch to symbols" key. - /* 73 */ "?123", + /* 79 */ "?123", // Label for "switch to symbols with microphone" key. This string shouldn't include the "mic" // part because it'll be appended by the code. - /* 74 */ "123", - /* 75~ */ + /* 80 */ "123", + /* 81~ */ EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, EMPTY, - /* ~84 */ + /* ~90 */ // U+00B9: "¹" SUPERSCRIPT ONE // U+00BD: "½" VULGAR FRACTION ONE HALF // U+2153: "⅓" VULGAR FRACTION ONE THIRD // U+00BC: "¼" VULGAR FRACTION ONE QUARTER // U+215B: "⅛" VULGAR FRACTION ONE EIGHTH - /* 85 */ "\u00B9,\u00BD,\u2153,\u00BC,\u215B", + /* 91 */ "\u00B9,\u00BD,\u2153,\u00BC,\u215B", // U+00B2: "²" SUPERSCRIPT TWO // U+2154: "⅔" VULGAR FRACTION TWO THIRDS - /* 86 */ "\u00B2,\u2154", + /* 92 */ "\u00B2,\u2154", // U+00B3: "³" SUPERSCRIPT THREE // U+00BE: "¾" VULGAR FRACTION THREE QUARTERS // U+215C: "⅜" VULGAR FRACTION THREE EIGHTHS - /* 87 */ "\u00B3,\u00BE,\u215C", + /* 93 */ "\u00B3,\u00BE,\u215C", // U+2074: "⁴" SUPERSCRIPT FOUR - /* 88 */ "\u2074", + /* 94 */ "\u2074", // U+215D: "⅝" VULGAR FRACTION FIVE EIGHTHS - /* 89 */ "\u215D", - /* 90 */ EMPTY, + /* 95 */ "\u215D", + /* 96 */ EMPTY, // U+215E: "⅞" VULGAR FRACTION SEVEN EIGHTHS - /* 91 */ "\u215E", - /* 92 */ EMPTY, - /* 93 */ EMPTY, + /* 97 */ "\u215E", + /* 98 */ EMPTY, + /* 99 */ EMPTY, // U+207F: "ⁿ" SUPERSCRIPT LATIN SMALL LETTER N // U+2205: "∅" EMPTY SET - /* 94 */ "\u207F,\u2205", - /* 95 */ ",", - /* 96 */ EMPTY, - /* 97 */ "?", - /* 98 */ ";", - /* 99 */ "%", + /* 100 */ "\u207F,\u2205", + /* 101 */ ",", + /* 102 */ EMPTY, + /* 103 */ "?", + /* 104 */ ";", + /* 105 */ "%", // U+00A1: "¡" INVERTED EXCLAMATION MARK - /* 100 */ "\u00A1", + /* 106 */ "\u00A1", // U+00BF: "¿" INVERTED QUESTION MARK - /* 101 */ "\u00BF", - /* 102 */ EMPTY, + /* 107 */ "\u00BF", + /* 108 */ EMPTY, // U+2030: "‰" PER MILLE SIGN - /* 103 */ "\u2030", - /* 104 */ ",", - /* 105~ */ + /* 109 */ "\u2030", + /* 110 */ ",", + /* 111~ */ EMPTY, EMPTY, EMPTY, - /* ~107 */ + /* ~113 */ // U+2026: "…" HORIZONTAL ELLIPSIS - /* 108 */ "\u2026", - /* 109 */ "\'", - /* 110 */ "\"", - /* 111 */ "\"", - /* 112 */ EMPTY, - /* 113 */ EMPTY, - /* 114 */ "q", - /* 115 */ "w", - /* 116 */ "y", - /* 117 */ "x", + /* 114 */ "\u2026", + /* 115 */ "\'", + /* 116 */ "\"", + /* 117 */ "\"", /* 118 */ EMPTY, - /* 119 */ "!fixedColumnOrder!2,!hasLabels!,!text/label_time_am,!text/label_time_pm", - /* 120 */ "!icon/settings_key|!code/key_settings", - /* 121 */ "!icon/shortcut_key|!code/key_shortcut", - /* 122 */ "!hasLabels!,!text/label_next_key|!code/key_action_next", - /* 123 */ "!hasLabels!,!text/label_previous_key|!code/key_action_previous", + /* 119 */ EMPTY, + /* 120 */ "q", + /* 121 */ "w", + /* 122 */ "y", + /* 123 */ "x", + /* 124 */ EMPTY, + /* 125 */ "!fixedColumnOrder!2,!hasLabels!,!text/label_time_am,!text/label_time_pm", + /* 126 */ "!icon/settings_key|!code/key_settings", + /* 127 */ "!icon/shortcut_key|!code/key_shortcut", + /* 128 */ "!hasLabels!,!text/label_next_key|!code/key_action_next", + /* 129 */ "!hasLabels!,!text/label_previous_key|!code/key_action_previous", // Label for "switch to more symbol" modifier key. Must be short to fit on key! - /* 124 */ "= \\ <", + /* 130 */ "= \\ <", // Label for "switch to more symbol" modifier key on tablets. Must be short to fit on key! - /* 125 */ "~ [ <", + /* 131 */ "~ [ <", // Label for "Tab" key. Must be short to fit on key! - /* 126 */ "Tab", + /* 132 */ "Tab", // Label for "switch to phone numeric" key. Must be short to fit on key! - /* 127 */ "123", + /* 133 */ "123", // Label for "switch to phone symbols" key. Must be short to fit on key! // U+FF0A: "＊" FULLWIDTH ASTERISK // U+FF03: "＃" FULLWIDTH NUMBER SIGN - /* 128 */ "\uFF0A\uFF03", + /* 134 */ "\uFF0A\uFF03", // Key label for "ante meridiem" - /* 129 */ "AM", + /* 135 */ "AM", // Key label for "post meridiem" - /* 130 */ "PM", - /* 131 */ ".com", + /* 136 */ "PM", + /* 137 */ ".com", // popular web domains for the locale - most popular, displayed on the keyboard - /* 132 */ "!hasLabels!,.net,.org,.gov,.edu", - /* 133 */ "!fixedColumnOrder!5,!hasLabels!,=-O|=-O ,:-P|:-P ,;-)|;-) ,:-(|:-( ,:-)|:-) ,:-!|:-! ,:-$|:-$ ,B-)|B-) ,:O|:O ,:-*|:-* ,:-D|:-D ,:\'(|:\'( ,:-\\\\|:-\\\\ ,O:-)|O:-) ,:-[|:-[ ", + /* 138 */ "!hasLabels!,.net,.org,.gov,.edu", + /* 139 */ "!fixedColumnOrder!5,!hasLabels!,=-O|=-O ,:-P|:-P ,;-)|;-) ,:-(|:-( ,:-)|:-) ,:-!|:-! ,:-$|:-$ ,B-)|B-) ,:O|:O ,:-*|:-* ,:-D|:-D ,:\'(|:\'( ,:-\\\\|:-\\\\ ,O:-)|O:-) ,:-[|:-[ ", // U+2039: "‹" SINGLE LEFT-POINTING ANGLE QUOTATION MARK // U+203A: "›" SINGLE RIGHT-POINTING ANGLE QUOTATION MARK // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK @@ -422,25 +428,25 @@ public final class KeyboardTextsSet { // The following each quotation mark pair consist of // <opening quotation mark>, <closing quotation mark> // and is named after (single|double)_<opening quotation mark>_<closing quotation mark>. - /* 134 */ "\u2039,\u203A", - /* 135 */ "\u2039|\u203A,\u203A|\u2039", - /* 136 */ "\u203A,\u2039", - /* 137 */ "\u00AB,\u00BB", - /* 138 */ "\u00AB|\u00BB,\u00BB|\u00AB", - /* 139 */ "\u00BB,\u00AB", + /* 140 */ "\u2039,\u203A", + /* 141 */ "\u2039|\u203A,\u203A|\u2039", + /* 142 */ "\u203A,\u2039", + /* 143 */ "\u00AB,\u00BB", + /* 144 */ "\u00AB|\u00BB,\u00BB|\u00AB", + /* 145 */ "\u00BB,\u00AB", // The following each quotation mark triplet consists of // <another quotation mark>, <opening quotation mark>, <closing quotation mark> // and is named after (single|double)_<opening quotation mark>_<closing quotation mark>. - /* 140 */ "\u201A,\u2018,\u2019", - /* 141 */ "\u2019,\u201A,\u2018", - /* 142 */ "\u2018,\u201A,\u2019", - /* 143 */ "\u201E,\u201C,\u201D", - /* 144 */ "\u201D,\u201E,\u201C", - /* 145 */ "\u201C,\u201E,\u201D", - /* 146 */ "!fixedColumnOrder!5,!text/single_quotes,!text/single_angle_quotes", - /* 147 */ "!fixedColumnOrder!5,!text/double_quotes,!text/double_angle_quotes", - /* 148 */ "!fixedColumnOrder!6,!text/double_quotes,!text/single_quotes,!text/double_angle_quotes,!text/single_angle_quotes", - /* 149 */ "!icon/emoji_key|!code/key_emoji", + /* 146 */ "\u201A,\u2018,\u2019", + /* 147 */ "\u2019,\u201A,\u2018", + /* 148 */ "\u2018,\u201A,\u2019", + /* 149 */ "\u201E,\u201C,\u201D", + /* 150 */ "\u201D,\u201E,\u201C", + /* 151 */ "\u201C,\u201E,\u201D", + /* 152 */ "!fixedColumnOrder!5,!text/single_quotes,!text/single_angle_quotes", + /* 153 */ "!fixedColumnOrder!5,!text/double_quotes,!text/double_angle_quotes", + /* 154 */ "!fixedColumnOrder!6,!text/double_quotes,!text/single_quotes,!text/double_angle_quotes,!text/single_angle_quotes", + /* 155 */ "!icon/emoji_key|!code/key_emoji", }; /* Language af: Afrikaans */ @@ -502,44 +508,45 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0623: "ا" ARABIC LETTER ALEF // U+200C: ZERO WIDTH NON-JOINER // U+0628: "ب" ARABIC LETTER BEH // U+062C: "پ" ARABIC LETTER PEH - /* 45 */ "\u0623\u200C\u0628\u200C\u062C", - /* 46 */ null, - /* 47 */ null, - /* 48 */ "!text/single_laqm_raqm_rtl", - /* 49 */ "!text/double_laqm_raqm_rtl", - /* 50~ */ + /* 51 */ "\u0623\u200C\u0628\u200C\u062C", + /* 52 */ null, + /* 53 */ null, + /* 54 */ "!text/single_laqm_raqm_rtl", + /* 55 */ "!text/double_laqm_raqm_rtl", + /* 56~ */ null, null, null, - /* ~52 */ + /* ~58 */ // U+061F: "؟" ARABIC QUESTION MARK // U+060C: "،" ARABIC COMMA // U+061B: "؛" ARABIC SEMICOLON - /* 53 */ "!fixedColumnOrder!8,\",\',#,-,:,!,\u060C,\u061F,@,&,\\%,+,\u061B,/,(|),)|(", + /* 59 */ "!fixedColumnOrder!8,\",\',#,-,:,!,\u060C,\u061F,@,&,\\%,+,\u061B,/,(|),)|(", // U+2605: "★" BLACK STAR // U+066D: "٭" ARABIC FIVE POINTED STAR - /* 54 */ "\u2605,\u066D", + /* 60 */ "\u2605,\u066D", // U+266A: "♪" EIGHTH NOTE - /* 55 */ "\u266A", - /* 56 */ null, + /* 61 */ "\u266A", + /* 62 */ null, // The all letters need to be mirrored are found at // http://www.unicode.org/Public/6.1.0/ucd/BidiMirroring.txt // U+FD3E: "﴾" ORNATE LEFT PARENTHESIS // U+FD3F: "﴿" ORNATE RIGHT PARENTHESIS - /* 57 */ "!fixedColumnOrder!4,\uFD3E|\uFD3F,<|>,{|},[|]", - /* 58 */ "!fixedColumnOrder!4,\uFD3F|\uFD3E,>|<,}|{,]|[", + /* 63 */ "!fixedColumnOrder!4,\uFD3E|\uFD3F,<|>,{|},[|]", + /* 64 */ "!fixedColumnOrder!4,\uFD3F|\uFD3E,>|<,}|{,]|[", // U+2264: "≤" LESS-THAN OR EQUAL TO // U+2265: "≥" GREATER-THAN EQUAL TO // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK // U+00BB: "»" RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK // U+2039: "‹" SINGLE LEFT-POINTING ANGLE QUOTATION MARK // U+203A: "›" SINGLE RIGHT-POINTING ANGLE QUOTATION MARK - /* 59 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,\u00AB|\u00BB", - /* 60 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,\u00BB|\u00AB", + /* 65 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,\u00AB|\u00BB", + /* 66 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,\u00BB|\u00AB", // U+0655: "ٕ" ARABIC HAMZA BELOW // U+0654: "ٔ" ARABIC HAMZA ABOVE // U+0652: "ْ" ARABIC SUKUN @@ -556,70 +563,70 @@ public final class KeyboardTextsSet { // U+0640: "ـ" ARABIC TATWEEL // In order to make Tatweel easily distinguishable from other punctuations, we use consecutive Tatweels only for its displayed label. // Note: The space character is needed as a preceding letter to draw Arabic diacritics characters correctly. - /* 61 */ "!fixedColumnOrder!7, \u0655|\u0655, \u0654|\u0654, \u0652|\u0652, \u064D|\u064D, \u064C|\u064C, \u064B|\u064B, \u0651|\u0651, \u0656|\u0656, \u0670|\u0670, \u0653|\u0653, \u0650|\u0650, \u064F|\u064F, \u064E|\u064E,\u0640\u0640\u0640|\u0640", - /* 62 */ "\u0651", + /* 67 */ "!fixedColumnOrder!7, \u0655|\u0655, \u0654|\u0654, \u0652|\u0652, \u064D|\u064D, \u064C|\u064C, \u064B|\u064B, \u0651|\u0651, \u0656|\u0656, \u0670|\u0670, \u0653|\u0653, \u0650|\u0650, \u064F|\u064F, \u064E|\u064E,\u0640\u0640\u0640|\u0640", + /* 68 */ "\u0651", // U+0661: "١" ARABIC-INDIC DIGIT ONE - /* 63 */ "\u0661", + /* 69 */ "\u0661", // U+0662: "٢" ARABIC-INDIC DIGIT TWO - /* 64 */ "\u0662", + /* 70 */ "\u0662", // U+0663: "٣" ARABIC-INDIC DIGIT THREE - /* 65 */ "\u0663", + /* 71 */ "\u0663", // U+0664: "٤" ARABIC-INDIC DIGIT FOUR - /* 66 */ "\u0664", + /* 72 */ "\u0664", // U+0665: "٥" ARABIC-INDIC DIGIT FIVE - /* 67 */ "\u0665", + /* 73 */ "\u0665", // U+0666: "٦" ARABIC-INDIC DIGIT SIX - /* 68 */ "\u0666", + /* 74 */ "\u0666", // U+0667: "٧" ARABIC-INDIC DIGIT SEVEN - /* 69 */ "\u0667", + /* 75 */ "\u0667", // U+0668: "٨" ARABIC-INDIC DIGIT EIGHT - /* 70 */ "\u0668", + /* 76 */ "\u0668", // U+0669: "٩" ARABIC-INDIC DIGIT NINE - /* 71 */ "\u0669", + /* 77 */ "\u0669", // U+0660: "٠" ARABIC-INDIC DIGIT ZERO - /* 72 */ "\u0660", + /* 78 */ "\u0660", // Label for "switch to symbols" key. // U+061F: "؟" ARABIC QUESTION MARK - /* 73 */ "\u0663\u0662\u0661\u061F", + /* 79 */ "\u0663\u0662\u0661\u061F", // Label for "switch to symbols with microphone" key. This string shouldn't include the "mic" // part because it'll be appended by the code. - /* 74 */ "\u0663\u0662\u0661", - /* 75 */ "1", - /* 76 */ "2", - /* 77 */ "3", - /* 78 */ "4", - /* 79 */ "5", - /* 80 */ "6", - /* 81 */ "7", - /* 82 */ "8", - /* 83 */ "9", + /* 80 */ "\u0663\u0662\u0661", + /* 81 */ "1", + /* 82 */ "2", + /* 83 */ "3", + /* 84 */ "4", + /* 85 */ "5", + /* 86 */ "6", + /* 87 */ "7", + /* 88 */ "8", + /* 89 */ "9", // U+066B: "٫" ARABIC DECIMAL SEPARATOR // U+066C: "٬" ARABIC THOUSANDS SEPARATOR - /* 84 */ "0,\u066B,\u066C", - /* 85~ */ + /* 90 */ "0,\u066B,\u066C", + /* 91~ */ null, null, null, null, null, null, null, null, null, null, - /* ~94 */ + /* ~100 */ // U+060C: "،" ARABIC COMMA - /* 95 */ "\u060C", - /* 96 */ "\\,", - /* 97 */ "\u061F", - /* 98 */ "\u061B", + /* 101 */ "\u060C", + /* 102 */ "\\,", + /* 103 */ "\u061F", + /* 104 */ "\u061B", // U+066A: "٪" ARABIC PERCENT SIGN - /* 99 */ "\u066A", - /* 100 */ null, - /* 101 */ "?", - /* 102 */ ";", + /* 105 */ "\u066A", + /* 106 */ null, + /* 107 */ "?", + /* 108 */ ";", // U+2030: "‰" PER MILLE SIGN - /* 103 */ "\\%,\u2030", - /* 104~ */ + /* 109 */ "\\%,\u2030", + /* 110~ */ null, null, null, null, null, - /* ~108 */ + /* ~114 */ // U+060C: "،" ARABIC COMMA // U+061B: "؛" ARABIC SEMICOLON // U+061F: "؟" ARABIC QUESTION MARK - /* 109 */ "\u060C", - /* 110 */ "\u061F", - /* 111 */ "\u061F,\u061B,!,:,-,/,\',\"", + /* 115 */ "\u060C", + /* 116 */ "\u061F", + /* 117 */ "\u061F,\u061B,!,:,-,/,\',\"", }; /* Language az: Azerbaijani */ @@ -694,14 +701,16 @@ public final class KeyboardTextsSet { /* ~42 */ // U+0451: "ё" CYRILLIC SMALL LETTER IO /* 43 */ "\u0451", - /* 44 */ null, + /* 44~ */ + null, null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language bg: Bulgarian */ @@ -710,15 +719,16 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ null, + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ null, // single_quotes of Bulgarian is default single_quotes_right_left. - /* 47 */ "!text/double_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language ca: Catalan */ @@ -782,22 +792,22 @@ public final class KeyboardTextsSet { /* 15~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, - /* ~52 */ + null, null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~58 */ // U+00B7: "·" MIDDLE DOT - /* 53 */ "!fixedColumnOrder!4,\u00B7,!,\\,,?,:,;,@", - /* 54~ */ + /* 59 */ "!fixedColumnOrder!4,\u00B7,!,\\,,?,:,;,@", + /* 60~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~107 */ - /* 108 */ "?,\u00B7", - /* 109~ */ + /* ~113 */ + /* 114 */ "?,\u00B7", + /* 115~ */ null, null, null, null, null, null, null, null, null, - /* ~117 */ + /* ~123 */ // U+00E7: "ç" LATIN SMALL LETTER C WITH CEDILLA - /* 118 */ "\u00E7", + /* 124 */ "\u00E7", }; /* Language cs: Czech */ @@ -871,12 +881,12 @@ public final class KeyboardTextsSet { /* 13~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language da: Danish */ @@ -940,12 +950,12 @@ public final class KeyboardTextsSet { /* 24 */ "\u00F6", /* 25~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language de: German */ @@ -991,12 +1001,25 @@ public final class KeyboardTextsSet { /* 7~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, + /* ~44 */ + // U+00FC: "ü" LATIN SMALL LETTER U WITH DIAERESIS + /* 45 */ "\u00FC", + // U+00F6: "ö" LATIN SMALL LETTER O WITH DIAERESIS + /* 46 */ "\u00F6", + // U+00E4: "ä" LATIN SMALL LETTER A WITH DIAERESIS + /* 47 */ "\u00E4", + // U+00E8: "è" LATIN SMALL LETTER E WITH GRAVE + /* 48 */ "\u00E8", + // U+00E9: "é" LATIN SMALL LETTER E WITH ACUTE + /* 49 */ "\u00E9", + // U+00E0: "à" LATIN SMALL LETTER A WITH GRAVE + /* 50 */ "\u00E0", + /* 51 */ null, + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language el: Greek */ @@ -1005,12 +1028,13 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0391: "Α" GREEK CAPITAL LETTER ALPHA // U+0392: "Β" GREEK CAPITAL LETTER BETA // U+0393: "Γ" GREEK CAPITAL LETTER GAMMA - /* 45 */ "\u0391\u0392\u0393", + /* 51 */ "\u0391\u0392\u0393", }; /* Language en: English */ @@ -1182,20 +1206,20 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, - /* ~111 */ - /* 112 */ "q", - /* 113 */ "x", + null, null, null, null, null, null, null, null, + /* ~117 */ + /* 118 */ "q", + /* 119 */ "x", // U+015D: "ŝ" LATIN SMALL LETTER S WITH CIRCUMFLEX - /* 114 */ "\u015D", + /* 120 */ "\u015D", // U+011D: "ĝ" LATIN SMALL LETTER G WITH CIRCUMFLEX - /* 115 */ "\u011D", + /* 121 */ "\u011D", // U+016D: "ŭ" LATIN SMALL LETTER U WITH BREVE - /* 116 */ "\u016D", + /* 122 */ "\u016D", // U+0109: "ĉ" LATIN SMALL LETTER C WITH CIRCUMFLEX - /* 117 */ "\u0109", + /* 123 */ "\u0109", // U+0135: "ĵ" LATIN SMALL LETTER J WITH CIRCUMFLEX - /* 118 */ "\u0135", + /* 124 */ "\u0135", }; /* Language es: Spanish */ @@ -1254,29 +1278,30 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~52 */ + null, null, null, null, null, null, + /* ~58 */ // U+00A1: "¡" INVERTED EXCLAMATION MARK // U+00BF: "¿" INVERTED QUESTION MARK - /* 53 */ "!fixedColumnOrder!4,;,!,\\,,?,:,\u00A1,@,\u00BF", - /* 54~ */ + /* 59 */ "!fixedColumnOrder!4,;,!,\\,,?,:,\u00A1,@,\u00BF", + /* 60~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~105 */ + /* ~111 */ // U+00A1: "¡" INVERTED EXCLAMATION MARK - /* 106 */ "!,\u00A1", - /* 107 */ null, + /* 112 */ "!,\u00A1", + /* 113 */ null, // U+00BF: "¿" INVERTED QUESTION MARK - /* 108 */ "?,\u00BF", - /* 109 */ "\"", - /* 110 */ "\'", - /* 111 */ "\'", - /* 112~ */ + /* 114 */ "?,\u00BF", + /* 115 */ "\"", + /* 116 */ "\'", + /* 117 */ "\'", + /* 118~ */ null, null, null, null, null, null, - /* ~117 */ + /* ~123 */ // U+00F1: "ñ" LATIN SMALL LETTER N WITH TILDE - /* 118 */ "\u00F1", + /* 124 */ "\u00F1", }; /* Language et: Estonian */ @@ -1379,10 +1404,10 @@ public final class KeyboardTextsSet { /* 23 */ "\u00F5", /* 24~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language fa: Persian */ @@ -1391,45 +1416,46 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0627: "ا" ARABIC LETTER ALEF // U+200C: ZERO WIDTH NON-JOINER // U+0628: "ب" ARABIC LETTER BEH // U+067E: "پ" ARABIC LETTER PEH - /* 45 */ "\u0627\u200C\u0628\u200C\u067E", - /* 46 */ null, - /* 47 */ null, - /* 48 */ "!text/single_laqm_raqm_rtl", - /* 49 */ "!text/double_laqm_raqm_rtl", - /* 50 */ null, - // U+FDFC: "﷼" RIAL SIGN - /* 51 */ "\uFDFC", + /* 51 */ "\u0627\u200C\u0628\u200C\u067E", /* 52 */ null, + /* 53 */ null, + /* 54 */ "!text/single_laqm_raqm_rtl", + /* 55 */ "!text/double_laqm_raqm_rtl", + /* 56 */ null, + // U+FDFC: "﷼" RIAL SIGN + /* 57 */ "\uFDFC", + /* 58 */ null, // U+061F: "؟" ARABIC QUESTION MARK // U+060C: "،" ARABIC COMMA // U+061B: "؛" ARABIC SEMICOLON - /* 53 */ "!fixedColumnOrder!8,\",\',#,-,:,!,\u060C,\u061F,@,&,\\%,+,\u061B,/,(|),)|(", + /* 59 */ "!fixedColumnOrder!8,\",\',#,-,:,!,\u060C,\u061F,@,&,\\%,+,\u061B,/,(|),)|(", // U+2605: "★" BLACK STAR // U+066D: "٭" ARABIC FIVE POINTED STAR - /* 54 */ "\u2605,\u066D", + /* 60 */ "\u2605,\u066D", // U+266A: "♪" EIGHTH NOTE - /* 55 */ "\u266A", - /* 56 */ null, + /* 61 */ "\u266A", + /* 62 */ null, // The all letters need to be mirrored are found at // http://www.unicode.org/Public/6.1.0/ucd/BidiMirroring.txt // U+FD3E: "﴾" ORNATE LEFT PARENTHESIS // U+FD3F: "﴿" ORNATE RIGHT PARENTHESIS - /* 57 */ "!fixedColumnOrder!4,\uFD3E|\uFD3F,<|>,{|},[|]", - /* 58 */ "!fixedColumnOrder!4,\uFD3F|\uFD3E,>|<,}|{,]|[", + /* 63 */ "!fixedColumnOrder!4,\uFD3E|\uFD3F,<|>,{|},[|]", + /* 64 */ "!fixedColumnOrder!4,\uFD3F|\uFD3E,>|<,}|{,]|[", // U+2264: "≤" LESS-THAN OR EQUAL TO // U+2265: "≥" GREATER-THAN EQUAL TO // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK // U+00BB: "»" RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK // U+2039: "‹" SINGLE LEFT-POINTING ANGLE QUOTATION MARK // U+203A: "›" SINGLE RIGHT-POINTING ANGLE QUOTATION MARK - /* 59 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,<|>", - /* 60 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,>|<", + /* 65 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,<|>", + /* 66 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,>|<", // U+0655: "ٕ" ARABIC HAMZA BELOW // U+0652: "ْ" ARABIC SUKUN // U+0651: "ّ" ARABIC SHADDA @@ -1446,74 +1472,74 @@ public final class KeyboardTextsSet { // U+0640: "ـ" ARABIC TATWEEL // In order to make Tatweel easily distinguishable from other punctuations, we use consecutive Tatweels only for its displayed label. // Note: The space character is needed as a preceding letter to draw Arabic diacritics characters correctly. - /* 61 */ "!fixedColumnOrder!7, \u0655|\u0655, \u0652|\u0652, \u0651|\u0651, \u064C|\u064C, \u064D|\u064D, \u064B|\u064B, \u0654|\u0654, \u0656|\u0656, \u0670|\u0670, \u0653|\u0653, \u064F|\u064F, \u0650|\u0650, \u064E|\u064E,\u0640\u0640\u0640|\u0640", - /* 62 */ "\u064B", + /* 67 */ "!fixedColumnOrder!7, \u0655|\u0655, \u0652|\u0652, \u0651|\u0651, \u064C|\u064C, \u064D|\u064D, \u064B|\u064B, \u0654|\u0654, \u0656|\u0656, \u0670|\u0670, \u0653|\u0653, \u064F|\u064F, \u0650|\u0650, \u064E|\u064E,\u0640\u0640\u0640|\u0640", + /* 68 */ "\u064B", // U+06F1: "۱" EXTENDED ARABIC-INDIC DIGIT ONE - /* 63 */ "\u06F1", + /* 69 */ "\u06F1", // U+06F2: "۲" EXTENDED ARABIC-INDIC DIGIT TWO - /* 64 */ "\u06F2", + /* 70 */ "\u06F2", // U+06F3: "۳" EXTENDED ARABIC-INDIC DIGIT THREE - /* 65 */ "\u06F3", + /* 71 */ "\u06F3", // U+06F4: "۴" EXTENDED ARABIC-INDIC DIGIT FOUR - /* 66 */ "\u06F4", + /* 72 */ "\u06F4", // U+06F5: "۵" EXTENDED ARABIC-INDIC DIGIT FIVE - /* 67 */ "\u06F5", + /* 73 */ "\u06F5", // U+06F6: "۶" EXTENDED ARABIC-INDIC DIGIT SIX - /* 68 */ "\u06F6", + /* 74 */ "\u06F6", // U+06F7: "۷" EXTENDED ARABIC-INDIC DIGIT SEVEN - /* 69 */ "\u06F7", + /* 75 */ "\u06F7", // U+06F8: "۸" EXTENDED ARABIC-INDIC DIGIT EIGHT - /* 70 */ "\u06F8", + /* 76 */ "\u06F8", // U+06F9: "۹" EXTENDED ARABIC-INDIC DIGIT NINE - /* 71 */ "\u06F9", + /* 77 */ "\u06F9", // U+06F0: "۰" EXTENDED ARABIC-INDIC DIGIT ZERO - /* 72 */ "\u06F0", + /* 78 */ "\u06F0", // Label for "switch to symbols" key. // U+061F: "؟" ARABIC QUESTION MARK - /* 73 */ "\u06F3\u06F2\u06F1\u061F", + /* 79 */ "\u06F3\u06F2\u06F1\u061F", // Label for "switch to symbols with microphone" key. This string shouldn't include the "mic" // part because it'll be appended by the code. - /* 74 */ "\u06F3\u06F2\u06F1", - /* 75 */ "1", - /* 76 */ "2", - /* 77 */ "3", - /* 78 */ "4", - /* 79 */ "5", - /* 80 */ "6", - /* 81 */ "7", - /* 82 */ "8", - /* 83 */ "9", + /* 80 */ "\u06F3\u06F2\u06F1", + /* 81 */ "1", + /* 82 */ "2", + /* 83 */ "3", + /* 84 */ "4", + /* 85 */ "5", + /* 86 */ "6", + /* 87 */ "7", + /* 88 */ "8", + /* 89 */ "9", // U+066B: "٫" ARABIC DECIMAL SEPARATOR // U+066C: "٬" ARABIC THOUSANDS SEPARATOR - /* 84 */ "0,\u066B,\u066C", - /* 85~ */ + /* 90 */ "0,\u066B,\u066C", + /* 91~ */ null, null, null, null, null, null, null, null, null, null, - /* ~94 */ + /* ~100 */ // U+060C: "،" ARABIC COMMA - /* 95 */ "\u060C", - /* 96 */ "\\,", - /* 97 */ "\u061F", - /* 98 */ "\u061B", + /* 101 */ "\u060C", + /* 102 */ "\\,", + /* 103 */ "\u061F", + /* 104 */ "\u061B", // U+066A: "٪" ARABIC PERCENT SIGN - /* 99 */ "\u066A", - /* 100 */ null, - /* 101 */ "?", - /* 102 */ ";", + /* 105 */ "\u066A", + /* 106 */ null, + /* 107 */ "?", + /* 108 */ ";", // U+2030: "‰" PER MILLE SIGN - /* 103 */ "\\%,\u2030", + /* 109 */ "\\%,\u2030", // U+060C: "،" ARABIC COMMA // U+061B: "؛" ARABIC SEMICOLON // U+061F: "؟" ARABIC QUESTION MARK // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK // U+00BB: "»" RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK - /* 104 */ "\u060C", - /* 105 */ "!", - /* 106 */ "!,\\,", - /* 107 */ "\u061F", - /* 108 */ "\u061F,?", - /* 109 */ "\u060C", - /* 110 */ "\u061F", - /* 111 */ "!fixedColumnOrder!4,:,!,\u061F,\u061B,-,/,\u00AB|\u00BB,\u00BB|\u00AB", + /* 110 */ "\u060C", + /* 111 */ "!", + /* 112 */ "!,\\,", + /* 113 */ "\u061F", + /* 114 */ "\u061F,?", + /* 115 */ "\u060C", + /* 116 */ "\u061F", + /* 117 */ "!fixedColumnOrder!4,:,!,\u061F,\u061B,-,/,\u00AB|\u00BB,\u00BB|\u00AB", }; /* Language fi: Finnish */ @@ -1614,6 +1640,23 @@ public final class KeyboardTextsSet { /* 7 */ "\u00E7,\u0107,\u010D", // U+00FF: "ÿ" LATIN SMALL LETTER Y WITH DIAERESIS /* 8 */ "%,\u00FF", + /* 9~ */ + null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, + null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, + null, null, null, null, null, null, + /* ~44 */ + // U+00E8: "è" LATIN SMALL LETTER E WITH GRAVE + /* 45 */ "\u00E8", + // U+00E9: "é" LATIN SMALL LETTER E WITH ACUTE + /* 46 */ "\u00E9", + // U+00E0: "à" LATIN SMALL LETTER A WITH GRAVE + /* 47 */ "\u00E0", + // U+00FC: "ü" LATIN SMALL LETTER U WITH DIAERESIS + /* 48 */ "\u00FC", + // U+00F6: "ö" LATIN SMALL LETTER O WITH DIAERESIS + /* 49 */ "\u00F6", + // U+00E4: "ä" LATIN SMALL LETTER A WITH DIAERESIS + /* 50 */ "\u00E4", }; /* Language hi: Hindi */ @@ -1622,55 +1665,56 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0915: "क" DEVANAGARI LETTER KA // U+0916: "ख" DEVANAGARI LETTER KHA // U+0917: "ग" DEVANAGARI LETTER GA - /* 45 */ "\u0915\u0916\u0917", - /* 46~ */ + /* 51 */ "\u0915\u0916\u0917", + /* 52~ */ null, null, null, null, null, - /* ~50 */ + /* ~56 */ // U+20B9: "₹" INDIAN RUPEE SIGN - /* 51 */ "\u20B9", - /* 52~ */ + /* 57 */ "\u20B9", + /* 58~ */ null, null, null, null, null, null, null, null, null, null, null, - /* ~62 */ + /* ~68 */ // U+0967: "१" DEVANAGARI DIGIT ONE - /* 63 */ "\u0967", + /* 69 */ "\u0967", // U+0968: "२" DEVANAGARI DIGIT TWO - /* 64 */ "\u0968", + /* 70 */ "\u0968", // U+0969: "३" DEVANAGARI DIGIT THREE - /* 65 */ "\u0969", + /* 71 */ "\u0969", // U+096A: "४" DEVANAGARI DIGIT FOUR - /* 66 */ "\u096A", + /* 72 */ "\u096A", // U+096B: "५" DEVANAGARI DIGIT FIVE - /* 67 */ "\u096B", + /* 73 */ "\u096B", // U+096C: "६" DEVANAGARI DIGIT SIX - /* 68 */ "\u096C", + /* 74 */ "\u096C", // U+096D: "७" DEVANAGARI DIGIT SEVEN - /* 69 */ "\u096D", + /* 75 */ "\u096D", // U+096E: "८" DEVANAGARI DIGIT EIGHT - /* 70 */ "\u096E", + /* 76 */ "\u096E", // U+096F: "९" DEVANAGARI DIGIT NINE - /* 71 */ "\u096F", + /* 77 */ "\u096F", // U+0966: "०" DEVANAGARI DIGIT ZERO - /* 72 */ "\u0966", + /* 78 */ "\u0966", // Label for "switch to symbols" key. - /* 73 */ "?\u0967\u0968\u0969", + /* 79 */ "?\u0967\u0968\u0969", // Label for "switch to symbols with microphone" key. This string shouldn't include the "mic" // part because it'll be appended by the code. - /* 74 */ "\u0967\u0968\u0969", - /* 75 */ "1", - /* 76 */ "2", - /* 77 */ "3", - /* 78 */ "4", - /* 79 */ "5", - /* 80 */ "6", - /* 81 */ "7", - /* 82 */ "8", - /* 83 */ "9", - /* 84 */ "0", + /* 80 */ "\u0967\u0968\u0969", + /* 81 */ "1", + /* 82 */ "2", + /* 83 */ "3", + /* 84 */ "4", + /* 85 */ "5", + /* 86 */ "6", + /* 87 */ "7", + /* 88 */ "8", + /* 89 */ "9", + /* 90 */ "0", }; /* Language hr: Croatian */ @@ -1701,12 +1745,12 @@ public final class KeyboardTextsSet { /* 13~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language hu: Hungarian */ @@ -1755,12 +1799,13 @@ public final class KeyboardTextsSet { /* 5~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, + null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language hy: Armenian */ @@ -1769,8 +1814,8 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, - /* ~52 */ + null, null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~58 */ // U+058A: "֊" ARMENIAN HYPHEN // U+055C: "՜" ARMENIAN EXCLAMATION MARK // U+055D: "՝" ARMENIAN COMMA @@ -1779,19 +1824,19 @@ public final class KeyboardTextsSet { // U+055A: "՚" ARMENIAN APOSTROPHE // U+055B: "՛" ARMENIAN EMPHASIS MARK // U+055F: "՟" ARMENIAN ABBREVIATION MARK - /* 53 */ "!fixedColumnOrder!8,!,?,\\,,.,\u058A,\u055C,\u055D,\u055E,:,;,@,\u0559,\u055A,\u055B,\u055F", - /* 54~ */ + /* 59 */ "!fixedColumnOrder!8,!,?,\\,,.,\u058A,\u055C,\u055D,\u055E,:,;,@,\u0559,\u055A,\u055B,\u055F", + /* 60~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~99 */ + /* ~105 */ // U+055C: "՜" ARMENIAN EXCLAMATION MARK // U+00A1: "¡" INVERTED EXCLAMATION MARK - /* 100 */ "\u055C,\u00A1", + /* 106 */ "\u055C,\u00A1", // U+055E: "՞" ARMENIAN QUESTION MARK // U+00BF: "¿" INVERTED QUESTION MARK - /* 101 */ "\u055E,\u00BF", + /* 107 */ "\u055E,\u00BF", }; /* Language is: Icelandic */ @@ -1857,10 +1902,10 @@ public final class KeyboardTextsSet { /* 22 */ "\u00FE", /* 23~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + null, null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language it: Italian */ @@ -1914,12 +1959,13 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+05D0: "א" HEBREW LETTER ALEF // U+05D1: "ב" HEBREW LETTER BET // U+05D2: "ג" HEBREW LETTER GIMEL - /* 45 */ "\u05D0\u05D1\u05D2", + /* 51 */ "\u05D0\u05D1\u05D2", // The following characters don't need BIDI mirroring. // U+2018: "‘" LEFT SINGLE QUOTATION MARK // U+2019: "’" RIGHT SINGLE QUOTATION MARK @@ -1927,42 +1973,42 @@ public final class KeyboardTextsSet { // U+201C: "“" LEFT DOUBLE QUOTATION MARK // U+201D: "”" RIGHT DOUBLE QUOTATION MARK // U+201E: "„" DOUBLE LOW-9 QUOTATION MARK - /* 46 */ "\u2018,\u2019,\u201A", - /* 47 */ "\u201C,\u201D,\u201E", - /* 48 */ "!text/single_laqm_raqm_rtl", - /* 49 */ "!text/double_laqm_raqm_rtl", - /* 50 */ null, + /* 52 */ "\u2018,\u2019,\u201A", + /* 53 */ "\u201C,\u201D,\u201E", + /* 54 */ "!text/single_laqm_raqm_rtl", + /* 55 */ "!text/double_laqm_raqm_rtl", + /* 56 */ null, // U+20AA: "₪" NEW SHEQEL SIGN - /* 51 */ "\u20AA", - /* 52 */ null, - /* 53 */ null, + /* 57 */ "\u20AA", + /* 58 */ null, + /* 59 */ null, // U+2605: "★" BLACK STAR - /* 54 */ "\u2605", - /* 55 */ null, + /* 60 */ "\u2605", + /* 61 */ null, // U+00B1: "±" PLUS-MINUS SIGN // U+FB29: "﬩" HEBREW LETTER ALTERNATIVE PLUS SIGN - /* 56 */ "\u00B1,\uFB29", + /* 62 */ "\u00B1,\uFB29", // The all letters need to be mirrored are found at // http://www.unicode.org/Public/6.1.0/ucd/BidiMirroring.txt - /* 57 */ "!fixedColumnOrder!3,<|>,{|},[|]", - /* 58 */ "!fixedColumnOrder!3,>|<,}|{,]|[", + /* 63 */ "!fixedColumnOrder!3,<|>,{|},[|]", + /* 64 */ "!fixedColumnOrder!3,>|<,}|{,]|[", // U+2264: "≤" LESS-THAN OR EQUAL TO // U+2265: "≥" GREATER-THAN EQUAL TO // U+00AB: "«" LEFT-POINTING DOUBLE ANGLE QUOTATION MARK // U+00BB: "»" RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK // U+2039: "‹" SINGLE LEFT-POINTING ANGLE QUOTATION MARK // U+203A: "›" SINGLE RIGHT-POINTING ANGLE QUOTATION MARK - /* 59 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,\u00AB|\u00BB", - /* 60 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,\u00BB|\u00AB", - /* 61~ */ + /* 65 */ "!fixedColumnOrder!3,\u2039|\u203A,\u2264|\u2265,\u00AB|\u00BB", + /* 66 */ "!fixedColumnOrder!3,\u203A|\u2039,\u2265|\u2264,\u00BB|\u00AB", + /* 67~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~104 */ - /* 105 */ "!", - /* 106 */ "!", - /* 107 */ "?", - /* 108 */ "?", + /* ~110 */ + /* 111 */ "!", + /* 112 */ "!", + /* 113 */ "?", + /* 114 */ "?", }; /* Language ka: Georgian */ @@ -1971,14 +2017,15 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+10D0: "ა" GEORGIAN LETTER AN // U+10D1: "ბ" GEORGIAN LETTER BAN // U+10D2: "გ" GEORGIAN LETTER GAN - /* 45 */ "\u10D0\u10D1\u10D2", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + /* 51 */ "\u10D0\u10D1\u10D2", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language kk: Kazakh */ @@ -2021,12 +2068,14 @@ public final class KeyboardTextsSet { /* ~42 */ // U+0451: "ё" CYRILLIC SMALL LETTER IO /* 43 */ "\u0451", - /* 44 */ null, + /* 44~ */ + null, null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", + /* 51 */ "\u0410\u0411\u0412", }; /* Language km: Khmer */ @@ -2035,17 +2084,18 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+1780: "ក" KHMER LETTER KA // U+1781: "ខ" KHMER LETTER KHA // U+1782: "គ" KHMER LETTER KO - /* 45 */ "\u1780\u1781\u1782", - /* 46~ */ + /* 51 */ "\u1780\u1781\u1782", + /* 52~ */ null, null, null, null, - /* ~49 */ + /* ~55 */ // U+17DB: "៛" KHMER CURRENCY SYMBOL RIEL - /* 50 */ "\u17DB,\u00A2,\u00A3,\u20AC,\u00A5,\u20B1", + /* 56 */ "\u17DB,\u00A2,\u00A3,\u20AC,\u00A5,\u20B1", }; /* Language ky: Kirghiz */ @@ -2081,12 +2131,14 @@ public final class KeyboardTextsSet { /* ~42 */ // U+0451: "ё" CYRILLIC SMALL LETTER IO /* 43 */ "\u0451", - /* 44 */ null, + /* 44~ */ + null, null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", + /* 51 */ "\u0410\u0411\u0412", }; /* Language lo: Lao */ @@ -2095,17 +2147,18 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0E81: "ກ" LAO LETTER KO // U+0E82: "ຂ" LAO LETTER KHO SUNG // U+0E84: "ຄ" LAO LETTER KHO TAM - /* 45 */ "\u0E81\u0E82\u0E84", - /* 46~ */ + /* 51 */ "\u0E81\u0E82\u0E84", + /* 52~ */ null, null, null, null, null, - /* ~50 */ + /* ~56 */ // U+20AD: "₭" KIP SIGN - /* 51 */ "\u20AD", + /* 57 */ "\u20AD", }; /* Language lt: Lithuanian */ @@ -2199,9 +2252,10 @@ public final class KeyboardTextsSet { /* 16~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language lv: Latvian */ @@ -2294,9 +2348,10 @@ public final class KeyboardTextsSet { /* 16~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language mk: Macedonian */ @@ -2318,13 +2373,16 @@ public final class KeyboardTextsSet { /* 43 */ "\u0450", // U+045D: "ѝ" CYRILLIC SMALL LETTER I WITH GRAVE /* 44 */ "\u045D", + /* 45~ */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language mn: Mongolian */ @@ -2333,17 +2391,18 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46~ */ + /* 51 */ "\u0410\u0411\u0412", + /* 52~ */ null, null, null, null, null, - /* ~50 */ + /* ~56 */ // U+20AE: "₮" TUGRIK SIGN - /* 51 */ "\u20AE", + /* 57 */ "\u20AE", }; /* Language nb: Norwegian Bokmål */ @@ -2393,10 +2452,10 @@ public final class KeyboardTextsSet { /* 24 */ "\u00E4", /* 25~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", + null, null, null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", }; /* Language ne: Nepali */ @@ -2405,55 +2464,56 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0915: "क" DEVANAGARI LETTER KA // U+0916: "ख" DEVANAGARI LETTER KHA // U+0917: "ग" DEVANAGARI LETTER GA - /* 45 */ "\u0915\u0916\u0917", - /* 46~ */ + /* 51 */ "\u0915\u0916\u0917", + /* 52~ */ null, null, null, null, null, - /* ~50 */ + /* ~56 */ // U+0930/U+0941/U+002E "रु." NEPALESE RUPEE SIGN - /* 51 */ "\u0930\u0941.", - /* 52~ */ + /* 57 */ "\u0930\u0941.", + /* 58~ */ null, null, null, null, null, null, null, null, null, null, null, - /* ~62 */ + /* ~68 */ // U+0967: "१" DEVANAGARI DIGIT ONE - /* 63 */ "\u0967", + /* 69 */ "\u0967", // U+0968: "२" DEVANAGARI DIGIT TWO - /* 64 */ "\u0968", + /* 70 */ "\u0968", // U+0969: "३" DEVANAGARI DIGIT THREE - /* 65 */ "\u0969", + /* 71 */ "\u0969", // U+096A: "४" DEVANAGARI DIGIT FOUR - /* 66 */ "\u096A", + /* 72 */ "\u096A", // U+096B: "५" DEVANAGARI DIGIT FIVE - /* 67 */ "\u096B", + /* 73 */ "\u096B", // U+096C: "६" DEVANAGARI DIGIT SIX - /* 68 */ "\u096C", + /* 74 */ "\u096C", // U+096D: "७" DEVANAGARI DIGIT SEVEN - /* 69 */ "\u096D", + /* 75 */ "\u096D", // U+096E: "८" DEVANAGARI DIGIT EIGHT - /* 70 */ "\u096E", + /* 76 */ "\u096E", // U+096F: "९" DEVANAGARI DIGIT NINE - /* 71 */ "\u096F", + /* 77 */ "\u096F", // U+0966: "०" DEVANAGARI DIGIT ZERO - /* 72 */ "\u0966", + /* 78 */ "\u0966", // Label for "switch to symbols" key. - /* 73 */ "?\u0967\u0968\u0969", + /* 79 */ "?\u0967\u0968\u0969", // Label for "switch to symbols with microphone" key. This string shouldn't include the "mic" // part because it'll be appended by the code. - /* 74 */ "\u0967\u0968\u0969", - /* 75 */ "1", - /* 76 */ "2", - /* 77 */ "3", - /* 78 */ "4", - /* 79 */ "5", - /* 80 */ "6", - /* 81 */ "7", - /* 82 */ "8", - /* 83 */ "9", - /* 84 */ "0", + /* 80 */ "\u0967\u0968\u0969", + /* 81 */ "1", + /* 82 */ "2", + /* 83 */ "3", + /* 84 */ "4", + /* 85 */ "5", + /* 86 */ "6", + /* 87 */ "7", + /* 88 */ "8", + /* 89 */ "9", + /* 90 */ "0", }; /* Language nl: Dutch */ @@ -2508,10 +2568,10 @@ public final class KeyboardTextsSet { /* 9~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", + null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", }; /* Language pl: Polish */ @@ -2569,10 +2629,10 @@ public final class KeyboardTextsSet { /* 15~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", + null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", }; /* Language pt: Portuguese */ @@ -2675,10 +2735,10 @@ public final class KeyboardTextsSet { /* 12~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_rqm", - /* 47 */ "!text/double_9qm_rqm", + null, null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_rqm", + /* 53 */ "!text/double_9qm_rqm", }; /* Language ru: Russian */ @@ -2707,14 +2767,16 @@ public final class KeyboardTextsSet { /* ~42 */ // U+0451: "ё" CYRILLIC SMALL LETTER IO /* 43 */ "\u0451", - /* 44 */ null, + /* 44~ */ + null, null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", }; /* Language sk: Slovak */ @@ -2808,11 +2870,12 @@ public final class KeyboardTextsSet { /* 16~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language sl: Slovenian */ @@ -2836,12 +2899,12 @@ public final class KeyboardTextsSet { /* 13~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, - /* ~45 */ - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, + /* ~51 */ + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language sr: Serbian */ @@ -2881,16 +2944,19 @@ public final class KeyboardTextsSet { /* 43 */ "\u0450", // U+045D: "ѝ" CYRILLIC SMALL LETTER I WITH GRAVE /* 44 */ "\u045D", + /* 45~ */ + null, null, null, null, null, null, + /* ~50 */ // END: More keys definitions for Serbian (Cyrillic) // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language sv: Swedish */ @@ -2972,10 +3038,10 @@ public final class KeyboardTextsSet { /* 24 */ "\u00E6", /* 25~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, - /* ~47 */ - /* 48 */ "!text/single_raqm_laqm", - /* 49 */ "!text/double_raqm_laqm", + null, null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~53 */ + /* 54 */ "!text/single_raqm_laqm", + /* 55 */ "!text/double_raqm_laqm", }; /* Language sw: Swahili */ @@ -3035,17 +3101,18 @@ public final class KeyboardTextsSet { null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0E01: "ก" THAI CHARACTER KO KAI // U+0E02: "ข" THAI CHARACTER KHO KHAI // U+0E04: "ค" THAI CHARACTER KHO KHWAI - /* 45 */ "\u0E01\u0E02\u0E04", - /* 46~ */ + /* 51 */ "\u0E01\u0E02\u0E04", + /* 52~ */ null, null, null, null, null, - /* ~50 */ + /* ~56 */ // U+0E3F: "฿" THAI CURRENCY SYMBOL BAHT - /* 51 */ "\u0E3F", + /* 57 */ "\u0E3F", }; /* Language tl: Tagalog */ @@ -3175,20 +3242,20 @@ public final class KeyboardTextsSet { // U+044A: "ъ" CYRILLIC SMALL LETTER HARD SIGN /* 37 */ "\u044A", /* 38~ */ - null, null, null, null, null, null, null, - /* ~44 */ + null, null, null, null, null, null, null, null, null, null, null, null, null, + /* ~50 */ // Label for "switch to alphabetic" key. // U+0410: "А" CYRILLIC CAPITAL LETTER A // U+0411: "Б" CYRILLIC CAPITAL LETTER BE // U+0412: "В" CYRILLIC CAPITAL LETTER VE - /* 45 */ "\u0410\u0411\u0412", - /* 46 */ "!text/single_9qm_lqm", - /* 47 */ "!text/double_9qm_lqm", - /* 48~ */ + /* 51 */ "\u0410\u0411\u0412", + /* 52 */ "!text/single_9qm_lqm", + /* 53 */ "!text/double_9qm_lqm", + /* 54~ */ null, null, null, - /* ~50 */ + /* ~56 */ // U+20B4: "₴" HRYVNIA SIGN - /* 51 */ "\u20B4", + /* 57 */ "\u20B4", }; /* Language vi: Vietnamese */ @@ -3273,10 +3340,11 @@ public final class KeyboardTextsSet { /* 10~ */ null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, - null, null, null, null, null, null, null, null, null, null, null, - /* ~50 */ + null, null, null, null, null, null, null, null, null, null, null, null, null, null, null, + null, null, + /* ~56 */ // U+20AB: "₫" DONG SIGN - /* 51 */ "\u20AB", + /* 57 */ "\u20AB", }; /* Language zu: Zulu */ diff --git a/java/src/com/android/inputmethod/latin/Constants.java b/java/src/com/android/inputmethod/latin/Constants.java index c4f96016c..9a9653094 100644 --- a/java/src/com/android/inputmethod/latin/Constants.java +++ b/java/src/com/android/inputmethod/latin/Constants.java @@ -174,6 +174,7 @@ public final class Constants { public static final int CODE_SLASH = '/'; public static final int CODE_COMMERCIAL_AT = '@'; public static final int CODE_PLUS = '+'; + public static final int CODE_PERCENT = '%'; public static final int CODE_CLOSING_PARENTHESIS = ')'; public static final int CODE_CLOSING_SQUARE_BRACKET = ']'; public static final int CODE_CLOSING_CURLY_BRACKET = '}'; diff --git a/java/src/com/android/inputmethod/latin/LatinIME.java b/java/src/com/android/inputmethod/latin/LatinIME.java index 10f643c49..65ff95458 100644 --- a/java/src/com/android/inputmethod/latin/LatinIME.java +++ b/java/src/com/android/inputmethod/latin/LatinIME.java @@ -196,9 +196,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen private int mLastSelectionStart = NOT_A_CURSOR_POSITION; private int mLastSelectionEnd = NOT_A_CURSOR_POSITION; - // Whether we are expecting an onUpdateSelection event to fire. If it does when we don't - // "expect" it, it means the user actually moved the cursor. - private boolean mExpectingUpdateSelection; private int mDeleteCount; private long mLastKeyTime; private final TreeSet<Long> mCurrentlyPressedHardwareKeys = CollectionUtils.newTreeSet(); @@ -1081,16 +1078,9 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen + ", ce=" + composingSpanEnd); } if (ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS) { - final boolean expectingUpdateSelectionFromLogger = - ResearchLogger.getAndClearLatinIMEExpectingUpdateSelection(); ResearchLogger.latinIME_onUpdateSelection(mLastSelectionStart, mLastSelectionEnd, oldSelStart, oldSelEnd, newSelStart, newSelEnd, composingSpanStart, - composingSpanEnd, mExpectingUpdateSelection, - expectingUpdateSelectionFromLogger, mConnection); - if (expectingUpdateSelectionFromLogger) { - // TODO: Investigate. Quitting now sounds wrong - we won't do the resetting work - return; - } + composingSpanEnd, mConnection); } final boolean selectionChanged = mLastSelectionStart != newSelStart @@ -1109,14 +1099,8 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen // TODO: revisit this when LatinIME supports hardware keyboards. // NOTE: the test harness subclasses LatinIME and overrides isInputViewShown(). // TODO: find a better way to simulate actual execution. - if (isInputViewShown() && !mExpectingUpdateSelection + if (isInputViewShown() && !mConnection.isBelatedExpectedUpdate(oldSelStart, newSelStart)) { - // TAKE CARE: there is a race condition when we enter this test even when the user - // did not explicitly move the cursor. This happens when typing fast, where two keys - // turn this flag on in succession and both onUpdateSelection() calls arrive after - // the second one - the first call successfully avoids this test, but the second one - // enters. For the moment we rely on noComposingSpan to further reduce the impact. - // TODO: the following is probably better done in resetEntireInputState(). // it should only happen when the cursor moved, and the very purpose of the // test below is to narrow down whether this happened or not. Likewise with @@ -1161,7 +1145,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen mRecapitalizeStatus.deactivate(); mKeyboardSwitcher.updateShiftState(); } - mExpectingUpdateSelection = false; // Make a note of the cursor position mLastSelectionStart = newSelStart; @@ -1502,6 +1485,7 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen || codePoint == Constants.CODE_CLOSING_CURLY_BRACKET || codePoint == Constants.CODE_CLOSING_ANGLE_BRACKET || codePoint == Constants.CODE_PLUS + || codePoint == Constants.CODE_PERCENT || Character.getType(codePoint) == Character.OTHER_SYMBOL; } @@ -1739,7 +1723,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen } handleCharacter(primaryCode, keyX, keyY, spaceState); } - mExpectingUpdateSelection = true; return didAutoCorrect; } @@ -1805,7 +1788,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen } else { commitTyped(LastComposedWord.NOT_A_SEPARATOR); } - mExpectingUpdateSelection = true; } final int codePointBeforeCursor = mConnection.getCodePointBeforeCursor(); if (Character.isLetterOrDigit(codePointBeforeCursor) @@ -1995,7 +1977,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen mWordComposer.setBatchInputWord(batchInputText); mConnection.setComposingText(batchInputText, 1); } - mExpectingUpdateSelection = true; mConnection.endBatchEdit(); if (ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS) { ResearchLogger.latinIME_onEndBatchInput(batchInputText, 0, suggestedWords); @@ -2049,9 +2030,7 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen } private void handleBackspace(final int spaceState) { - // We revert these in this method if the deletion doesn't happen. mDeleteCount++; - mExpectingUpdateSelection = true; // In many cases, we may have to put the keyboard in auto-shift state again. However // we want to wait a few milliseconds before doing it to avoid the keyboard flashing @@ -2144,10 +2123,7 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen } final int codePointBeforeCursor = mConnection.getCodePointBeforeCursor(); if (codePointBeforeCursor == Constants.NOT_A_CODE) { - // Nothing to delete before the cursor. We have to revert the deletion states - // that were updated at the beginning of this method. - mDeleteCount--; - mExpectingUpdateSelection = false; + // Nothing to delete before the cursor. return; } final int lengthToDelete = @@ -2318,9 +2294,9 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen if (!mRecapitalizeStatus.isSetAt(mLastSelectionStart, mLastSelectionEnd)) { mLastSelectionStart = mRecapitalizeStatus.getNewCursorStart(); mLastSelectionEnd = mRecapitalizeStatus.getNewCursorEnd(); - mConnection.setSelection(mLastSelectionStart, mLastSelectionEnd); } } + mConnection.finishComposingText(); mRecapitalizeStatus.rotate(); final int numCharsDeleted = mLastSelectionEnd - mLastSelectionStart; mConnection.setSelection(mLastSelectionEnd, mLastSelectionEnd); @@ -2656,7 +2632,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen ResearchLogger.latinIme_commitCurrentAutoCorrection(typedWord, autoCorrection, separator, mWordComposer.isBatchMode(), suggestedWords); } - mExpectingUpdateSelection = true; commitChosenWord(autoCorrection, LastComposedWord.COMMIT_TYPE_DECIDED_WORD, separator); if (!typedWord.equals(autoCorrection)) { @@ -2727,7 +2702,6 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen // typed word. final String replacedWord = mWordComposer.getTypedWord(); LatinImeLogger.logOnManualSuggestion(replacedWord, suggestion, index, suggestedWords); - mExpectingUpdateSelection = true; commitChosenWord(suggestion, LastComposedWord.COMMIT_TYPE_MANUAL_PICK, LastComposedWord.NOT_A_SEPARATOR); if (ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS) { diff --git a/java/src/com/android/inputmethod/latin/makedict/AbstractDictDecoder.java b/java/src/com/android/inputmethod/latin/makedict/AbstractDictDecoder.java index 9f7f502ea..fda97dafc 100644 --- a/java/src/com/android/inputmethod/latin/makedict/AbstractDictDecoder.java +++ b/java/src/com/android/inputmethod/latin/makedict/AbstractDictDecoder.java @@ -60,7 +60,8 @@ public abstract class AbstractDictDecoder implements DictDecoder { 0 != (optionsFlags & FormatSpec.GERMAN_UMLAUT_PROCESSING_FLAG), 0 != (optionsFlags & FormatSpec.FRENCH_LIGATURE_PROCESSING_FLAG)), new FormatOptions(version, - 0 != (optionsFlags & FormatSpec.SUPPORTS_DYNAMIC_UPDATE))); + 0 != (optionsFlags & FormatSpec.SUPPORTS_DYNAMIC_UPDATE), + 0 != (optionsFlags & FormatSpec.CONTAINS_TIMESTAMP_FLAG))); return header; } diff --git a/java/src/com/android/inputmethod/latin/makedict/BinaryDictDecoderUtils.java b/java/src/com/android/inputmethod/latin/makedict/BinaryDictDecoderUtils.java index 216492b4d..8a8ceaa8c 100644 --- a/java/src/com/android/inputmethod/latin/makedict/BinaryDictDecoderUtils.java +++ b/java/src/com/android/inputmethod/latin/makedict/BinaryDictDecoderUtils.java @@ -169,6 +169,14 @@ public final class BinaryDictDecoderUtils { return size; } + static int getCharArraySize(final int[] chars, final int start, final int end) { + int size = 0; + for (int i = start; i < end; ++i) { + size += getCharSize(chars[i]); + } + return size; + } + /** * Writes a char array to a byte buffer. * @@ -200,8 +208,7 @@ public final class BinaryDictDecoderUtils { * @param word the string to write. * @return the size written, in bytes. */ - static int writeString(final byte[] buffer, final int origin, - final String word) { + static int writeString(final byte[] buffer, final int origin, final String word) { final int length = word.length(); int index = origin; for (int i = 0; i < length; i = word.offsetByCodePoints(i, 1)) { @@ -223,22 +230,62 @@ public final class BinaryDictDecoderUtils { * * This will also write the terminator byte. * - * @param buffer the OutputStream to write to. + * @param stream the OutputStream to write to. * @param word the string to write. + * @return the size written, in bytes. */ - static void writeString(final OutputStream buffer, final String word) throws IOException { + static int writeString(final OutputStream stream, final String word) throws IOException { final int length = word.length(); + int written = 0; for (int i = 0; i < length; i = word.offsetByCodePoints(i, 1)) { final int codePoint = word.codePointAt(i); - if (1 == getCharSize(codePoint)) { - buffer.write((byte) codePoint); + final int charSize = getCharSize(codePoint); + if (1 == charSize) { + stream.write((byte) codePoint); } else { - buffer.write((byte) (0xFF & (codePoint >> 16))); - buffer.write((byte) (0xFF & (codePoint >> 8))); - buffer.write((byte) (0xFF & codePoint)); + stream.write((byte) (0xFF & (codePoint >> 16))); + stream.write((byte) (0xFF & (codePoint >> 8))); + stream.write((byte) (0xFF & codePoint)); } + written += charSize; + } + stream.write(FormatSpec.PTNODE_CHARACTERS_TERMINATOR); + written += FormatSpec.PTNODE_TERMINATOR_SIZE; + return written; + } + + /** + * Writes an array of code points with our character format to an OutputStream. + * + * This will also write the terminator byte. + * + * @param stream the OutputStream to write to. + * @param codePoints the array of code points + * @return the size written, in bytes. + */ + // TODO: Merge this method with writeCharArray and rename the various write* methods to + // make the difference clear. + static int writeCodePoints(final OutputStream stream, final int[] codePoints, + final int startIndex, final int endIndex) + throws IOException { + int written = 0; + for (int i = startIndex; i < endIndex; ++i) { + final int codePoint = codePoints[i]; + final int charSize = getCharSize(codePoint); + if (1 == charSize) { + stream.write((byte) codePoint); + } else { + stream.write((byte) (0xFF & (codePoint >> 16))); + stream.write((byte) (0xFF & (codePoint >> 8))); + stream.write((byte) (0xFF & codePoint)); + } + written += charSize; + } + if (endIndex - startIndex > 1) { + stream.write(FormatSpec.PTNODE_CHARACTERS_TERMINATOR); + written += FormatSpec.PTNODE_TERMINATOR_SIZE; } - buffer.write(FormatSpec.PTNODE_CHARACTERS_TERMINATOR); + return written; } /** diff --git a/java/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java b/java/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java index f761829de..bc1a2579e 100644 --- a/java/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java +++ b/java/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java @@ -17,6 +17,7 @@ package com.android.inputmethod.latin.makedict; import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.CharEncoding; +import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.DictBuffer; import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; import com.android.inputmethod.latin.makedict.FusionDictionary.DictionaryOptions; @@ -245,6 +246,26 @@ public class BinaryDictEncoderUtils { } } + static void writeUIntToDictBuffer(final DictBuffer dictBuffer, final int value, + final int size) { + switch(size) { + case 4: + dictBuffer.put((byte) ((value >> 24) & 0xFF)); + /* fall through */ + case 3: + dictBuffer.put((byte) ((value >> 16) & 0xFF)); + /* fall through */ + case 2: + dictBuffer.put((byte) ((value >> 8) & 0xFF)); + /* fall through */ + case 1: + dictBuffer.put((byte) (value & 0xFF)); + break; + default: + /* nop */ + } + } + // End utility methods // This method is responsible for finding a nice ordering of the nodes that favors run-time diff --git a/java/src/com/android/inputmethod/latin/makedict/BinaryDictIOUtils.java b/java/src/com/android/inputmethod/latin/makedict/BinaryDictIOUtils.java index 0f7d2f6c9..8d14e4d60 100644 --- a/java/src/com/android/inputmethod/latin/makedict/BinaryDictIOUtils.java +++ b/java/src/com/android/inputmethod/latin/makedict/BinaryDictIOUtils.java @@ -245,8 +245,7 @@ public final class BinaryDictIOUtils { /** * @return the size written, in bytes. Always 3 bytes. */ - static int writeSInt24ToBuffer(final DictBuffer dictBuffer, - final int value) { + static int writeSInt24ToBuffer(final DictBuffer dictBuffer, final int value) { final int absValue = Math.abs(value); dictBuffer.put((byte)(((value < 0 ? 0x80 : 0) | (absValue >> 16)) & 0xFF)); dictBuffer.put((byte)((absValue >> 8) & 0xFF)); @@ -301,35 +300,6 @@ public final class BinaryDictIOUtils { } /** - * Write a string to a stream. - * - * @param destination the stream to write. - * @param word the string to be written. - * @return the size written, in bytes. - * @throws IOException - */ - private static int writeString(final OutputStream destination, final String word) - throws IOException { - int size = 0; - final int length = word.length(); - for (int i = 0; i < length; i = word.offsetByCodePoints(i, 1)) { - final int codePoint = word.codePointAt(i); - if (CharEncoding.getCharSize(codePoint) == 1) { - destination.write((byte)codePoint); - size++; - } else { - destination.write((byte)(0xFF & (codePoint >> 16))); - destination.write((byte)(0xFF & (codePoint >> 8))); - destination.write((byte)(0xFF & codePoint)); - size += 3; - } - } - destination.write((byte)FormatSpec.PTNODE_CHARACTERS_TERMINATOR); - size += FormatSpec.PTNODE_TERMINATOR_SIZE; - return size; - } - - /** * Write a PtNode to an output stream from a PtNodeInfo. * A PtNode is an in-memory representation of a node in the patricia trie. * A PtNode info is a container for low-level information about how the @@ -387,7 +357,7 @@ public final class BinaryDictIOUtils { destination.write((byte)BinaryDictEncoderUtils.makeShortcutFlags( shortcutIterator.hasNext(), target.mFrequency)); size++; - size += writeString(destination, target.mWord); + size += CharEncoding.writeString(destination, target.mWord); } } @@ -445,6 +415,25 @@ public final class BinaryDictIOUtils { } /** + * Writes a PtNodeCount to the stream. + * + * @param destination the stream to write. + * @param ptNodeCount the count. + * @return the size written in bytes. + */ + static int writePtNodeCount(final OutputStream destination, final int ptNodeCount) + throws IOException { + final int countSize = BinaryDictIOUtils.getPtNodeCountSize(ptNodeCount); + // the count must fit on one byte or two bytes. + // Please see comments in FormatSpec. + if (countSize != 1 && countSize != 2) { + throw new RuntimeException("Strange size from getPtNodeCountSize : " + countSize); + } + BinaryDictEncoderUtils.writeUIntToStream(destination, ptNodeCount, countSize); + return countSize; + } + + /** * Write a node array to the stream. * * @param destination the stream to write. @@ -454,18 +443,7 @@ public final class BinaryDictIOUtils { */ static int writeNodes(final OutputStream destination, final PtNodeInfo[] infos) throws IOException { - int size = getPtNodeCountSize(infos.length); - switch (getPtNodeCountSize(infos.length)) { - case 1: - destination.write((byte)infos.length); - break; - case 2: - destination.write((byte)(infos.length >> 8)); - destination.write((byte)(infos.length & 0xFF)); - break; - default: - throw new RuntimeException("Invalid node count size."); - } + int size = writePtNodeCount(destination, infos.length); for (final PtNodeInfo info : infos) size += writePtNode(destination, info); writeSInt24ToStream(destination, FormatSpec.NO_FORWARD_LINK_ADDRESS); return size + FormatSpec.FORWARD_LINK_ADDRESS_SIZE; diff --git a/java/src/com/android/inputmethod/latin/makedict/DictDecoder.java b/java/src/com/android/inputmethod/latin/makedict/DictDecoder.java index 3dbeee099..91543986d 100644 --- a/java/src/com/android/inputmethod/latin/makedict/DictDecoder.java +++ b/java/src/com/android/inputmethod/latin/makedict/DictDecoder.java @@ -35,6 +35,7 @@ import java.util.TreeMap; /** * An interface of binary dictionary decoders. */ +// TODO: Straighten out responsibility for the buffer's file pointer. public interface DictDecoder { /** @@ -43,7 +44,7 @@ public interface DictDecoder { public FileHeader readHeader() throws IOException, UnsupportedFormatException; /** - * Reads PtNode from nodeAddress. + * Reads PtNode from ptNodePos. * @param ptNodePos the position of PtNode. * @param formatOptions the format options. * @return PtNodeInfo. diff --git a/java/src/com/android/inputmethod/latin/makedict/DynamicBinaryDictIOUtils.java b/java/src/com/android/inputmethod/latin/makedict/DynamicBinaryDictIOUtils.java index 336277196..971b4ff9f 100644 --- a/java/src/com/android/inputmethod/latin/makedict/DynamicBinaryDictIOUtils.java +++ b/java/src/com/android/inputmethod/latin/makedict/DynamicBinaryDictIOUtils.java @@ -22,6 +22,7 @@ import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.DictBuffer; import com.android.inputmethod.latin.makedict.FormatSpec.FileHeader; import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions; import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString; +import com.android.inputmethod.latin.utils.CollectionUtils; import java.io.IOException; import java.io.OutputStream; @@ -36,7 +37,7 @@ import java.util.Arrays; @UsedForTesting public final class DynamicBinaryDictIOUtils { private static final boolean DBG = false; - private static final int MAX_JUMPS = 10000; + static final int MAX_JUMPS = 10000; private DynamicBinaryDictIOUtils() { // This utility class is not publicly instantiable. @@ -217,6 +218,25 @@ public final class DynamicBinaryDictIOUtils { } /** + * Converts a list of WeightedString to a list of PendingAttribute. + */ + public static ArrayList<PendingAttribute> resolveBigramPositions(final DictUpdater dictUpdater, + final ArrayList<WeightedString> bigramStrings) + throws IOException, UnsupportedFormatException { + if (bigramStrings == null) return CollectionUtils.newArrayList(); + final ArrayList<PendingAttribute> bigrams = CollectionUtils.newArrayList(); + for (final WeightedString bigram : bigramStrings) { + final int pos = dictUpdater.getTerminalPosition(bigram.mWord); + if (pos == FormatSpec.NOT_VALID_WORD) { + // TODO: figure out what is the correct thing to do here. + } else { + bigrams.add(new PendingAttribute(bigram.mFrequency, pos)); + } + } + return bigrams; + } + + /** * Insert a word into a binary dictionary. * * @param dictUpdater the dict updater. @@ -238,18 +258,9 @@ public final class DynamicBinaryDictIOUtils { final ArrayList<WeightedString> shortcuts, final boolean isNotAWord, final boolean isBlackListEntry) throws IOException, UnsupportedFormatException { - final ArrayList<PendingAttribute> bigrams = new ArrayList<PendingAttribute>(); + final ArrayList<PendingAttribute> bigrams = resolveBigramPositions(dictUpdater, + bigramStrings); final DictBuffer dictBuffer = dictUpdater.getDictBuffer(); - if (bigramStrings != null) { - for (final WeightedString bigram : bigramStrings) { - int position = dictUpdater.getTerminalPosition(bigram.mWord); - if (position == FormatSpec.NOT_VALID_WORD) { - // TODO: figure out what is the correct thing to do here. - } else { - bigrams.add(new PendingAttribute(bigram.mFrequency, position)); - } - } - } final boolean isTerminal = true; final boolean hasBigrams = !bigrams.isEmpty(); diff --git a/java/src/com/android/inputmethod/latin/makedict/FormatSpec.java b/java/src/com/android/inputmethod/latin/makedict/FormatSpec.java index 5a5d7af6b..b99aca2c8 100644 --- a/java/src/com/android/inputmethod/latin/makedict/FormatSpec.java +++ b/java/src/com/android/inputmethod/latin/makedict/FormatSpec.java @@ -37,13 +37,15 @@ public final class FormatSpec { * sion * * o | - * p | not used 4 bits - * t | has bigrams ? 1 bit, 1 = yes, 0 = no : CONTAINS_BIGRAMS_FLAG - * i | FRENCH_LIGATURE_PROCESSING_FLAG - * o | supports dynamic updates ? 1 bit, 1 = yes, 0 = no : SUPPORTS_DYNAMIC_UPDATE - * n | GERMAN_UMLAUT_PROCESSING_FLAG - * f | - * lags + * p | not used 3 bits + * t | each unigram and bigram entry has a time stamp? + * i | 1 bit, 1 = yes, 0 = no : CONTAINS_TIMESTAMP_FLAG + * o | has bigrams ? 1 bit, 1 = yes, 0 = no : CONTAINS_BIGRAMS_FLAG + * n | FRENCH_LIGATURE_PROCESSING_FLAG + * f | supports dynamic updates ? 1 bit, 1 = yes, 0 = no : SUPPORTS_DYNAMIC_UPDATE + * l | GERMAN_UMLAUT_PROCESSING_FLAG + * a | + * gs * * h | * e | size of the file header, 4bytes @@ -211,6 +213,7 @@ public final class FormatSpec { static final int SUPPORTS_DYNAMIC_UPDATE = 0x2; static final int FRENCH_LIGATURE_PROCESSING_FLAG = 0x4; static final int CONTAINS_BIGRAMS_FLAG = 0x8; + static final int CONTAINS_TIMESTAMP_FLAG = 0x10; // TODO: Make this value adaptative to content data, store it in the header, and // use it in the reading code. @@ -263,6 +266,7 @@ public final class FormatSpec { // These values are used only by version 4 or later. static final String TRIE_FILE_EXTENSION = ".trie"; static final String FREQ_FILE_EXTENSION = ".freq"; + static final String UNIGRAM_TIMESTAMP_FILE_EXTENSION = ".timestamp"; // tat = Terminal Address Table static final String TERMINAL_ADDRESS_TABLE_FILE_EXTENSION = ".tat"; static final String BIGRAM_FILE_EXTENSION = ".bigram"; @@ -271,19 +275,25 @@ public final class FormatSpec { static final String CONTENT_TABLE_FILE_SUFFIX = "_index"; static final int FREQUENCY_AND_FLAGS_SIZE = 2; static final int TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE = 3; + static final int UNIGRAM_TIMESTAMP_SIZE = 4; // With the English main dictionary as of October 2013, the size of bigram address table is - // is 584KB with the block size being 4. - // This is 91% of that of full address table. - static final int BIGRAM_ADDRESS_TABLE_BLOCK_SIZE = 4; - static final int BIGRAM_CONTENT_COUNT = 1; + // is 345KB with the block size being 16. + // This is 54% of that of full address table. + static final int BIGRAM_ADDRESS_TABLE_BLOCK_SIZE = 16; + static final int BIGRAM_CONTENT_COUNT = 2; static final int BIGRAM_FREQ_CONTENT_INDEX = 0; + static final int BIGRAM_TIMESTAMP_CONTENT_INDEX = 1; static final String BIGRAM_FREQ_CONTENT_ID = "_freq"; + static final String BIGRAM_TIMESTAMP_CONTENT_ID = "_timestamp"; + static final int BIGRAM_TIMESTAMP_SIZE = 4; + static final int BIGRAM_COUNTER_SIZE = 1; + static final int BIGRAM_LEVEL_SIZE = 1; static final int SHORTCUT_CONTENT_COUNT = 1; static final int SHORTCUT_CONTENT_INDEX = 0; // With the English main dictionary as of October 2013, the size of shortcut address table is - // 29KB with the block size being 64. + // 26KB with the block size being 64. // This is only 4.4% of that of full address table. static final int SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE = 64; static final String SHORTCUT_CONTENT_ID = "_shortcut"; @@ -321,6 +331,7 @@ public final class FormatSpec { public final int mVersion; public final boolean mSupportsDynamicUpdate; public final boolean mHasTerminalId; + public final boolean mHasTimestamp; @UsedForTesting public FormatOptions(final int version) { this(version, false); @@ -328,6 +339,11 @@ public final class FormatSpec { @UsedForTesting public FormatOptions(final int version, final boolean supportsDynamicUpdate) { + this(version, supportsDynamicUpdate, false /* hasTimestamp */); + } + + public FormatOptions(final int version, final boolean supportsDynamicUpdate, + final boolean hasTimestamp) { mVersion = version; if (version < FIRST_VERSION_WITH_DYNAMIC_UPDATE && supportsDynamicUpdate) { throw new RuntimeException("Dynamic updates are only supported with versions " @@ -335,6 +351,7 @@ public final class FormatSpec { } mSupportsDynamicUpdate = supportsDynamicUpdate; mHasTerminalId = (version >= FIRST_VERSION_WITH_TERMINAL_ID); + mHasTimestamp = hasTimestamp; } } diff --git a/java/src/com/android/inputmethod/latin/makedict/SparseTableContentReader.java b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentReader.java new file mode 100644 index 000000000..06088b651 --- /dev/null +++ b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentReader.java @@ -0,0 +1,120 @@ +/* + * Copyright (C) 2013 The Android Open Source Project + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package com.android.inputmethod.latin.makedict; + +import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.DictBuffer; +import com.android.inputmethod.latin.makedict.DictDecoder.DictionaryBufferFactory; + +import java.io.File; +import java.io.FileNotFoundException; +import java.io.IOException; + +/** + * An auxiliary class for reading SparseTable and data written by SparseTableContentWriter. + */ +public class SparseTableContentReader { + + /** + * An interface of a function which is passed to SparseTableContentReader.read. + */ + public interface SparseTableContentReaderInterface { + /** + * Reads data. + * + * @param buffer the DictBuffer. The position of the buffer is set to the head of data. + */ + public void read(final DictBuffer buffer); + } + + protected final int mContentCount; + protected final int mBlockSize; + protected final File mBaseDir; + protected final File mLookupTableFile; + protected final File[] mAddressTableFiles; + protected final File[] mContentFiles; + protected DictBuffer mLookupTableBuffer; + protected final DictBuffer[] mAddressTableBuffers; + private final DictBuffer[] mContentBuffers; + protected final DictionaryBufferFactory mFactory; + + /** + * Sole constructor of SparseTableContentReader. + * + * @param name the name of SparseTable. + * @param blockSize the block size of the content table. + * @param baseDir the directory which contains the files of the content table. + * @param contentFilenames the file names of content files. + * @param contentIds the ids of contents. These ids are used for a suffix of a name of + * address files and content files. + * @param factory the DictionaryBufferFactory which is used for opening the files. + */ + public SparseTableContentReader(final String name, final int blockSize, final File baseDir, + final String[] contentFilenames, final String[] contentIds, + final DictionaryBufferFactory factory) { + if (contentFilenames.length != contentIds.length) { + throw new RuntimeException("The length of contentFilenames and the length of" + + " contentIds are different " + contentFilenames.length + ", " + + contentIds.length); + } + mBlockSize = blockSize; + mBaseDir = baseDir; + mFactory = factory; + mContentCount = contentFilenames.length; + mLookupTableFile = new File(baseDir, name + FormatSpec.LOOKUP_TABLE_FILE_SUFFIX); + mAddressTableFiles = new File[mContentCount]; + mContentFiles = new File[mContentCount]; + for (int i = 0; i < mContentCount; ++i) { + mAddressTableFiles[i] = new File(mBaseDir, + name + FormatSpec.CONTENT_TABLE_FILE_SUFFIX + contentIds[i]); + mContentFiles[i] = new File(mBaseDir, contentFilenames[i] + contentIds[i]); + } + mAddressTableBuffers = new DictBuffer[mContentCount]; + mContentBuffers = new DictBuffer[mContentCount]; + } + + public void openBuffers() throws FileNotFoundException, IOException { + mLookupTableBuffer = mFactory.getDictionaryBuffer(mLookupTableFile); + for (int i = 0; i < mContentCount; ++i) { + mAddressTableBuffers[i] = mFactory.getDictionaryBuffer(mAddressTableFiles[i]); + mContentBuffers[i] = mFactory.getDictionaryBuffer(mContentFiles[i]); + } + } + + protected void read(final int contentIndex, final int index, + final SparseTableContentReaderInterface reader) { + if (index < 0 || (index / mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES + >= mLookupTableBuffer.limit()) { + return; + } + + mLookupTableBuffer.position((index / mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES); + final int posInAddressTable = mLookupTableBuffer.readInt(); + if (posInAddressTable == SparseTable.NOT_EXIST) { + return; + } + + mAddressTableBuffers[contentIndex].position( + (posInAddressTable + index % mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES); + final int address = mAddressTableBuffers[contentIndex].readInt(); + if (address == SparseTable.NOT_EXIST) { + return; + } + + mContentBuffers[contentIndex].position(address); + reader.read(mContentBuffers[contentIndex]); + } +}
\ No newline at end of file diff --git a/java/src/com/android/inputmethod/latin/makedict/SparseTableContentUpdater.java b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentUpdater.java new file mode 100644 index 000000000..4518f21b9 --- /dev/null +++ b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentUpdater.java @@ -0,0 +1,123 @@ +/* + * Copyright (C) 2013 The Android Open Source Project + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package com.android.inputmethod.latin.makedict; + +import com.android.inputmethod.latin.makedict.DictDecoder.DictionaryBufferFactory; + +import java.io.File; +import java.io.FileOutputStream; +import java.io.IOException; +import java.io.OutputStream; + +/** + * An auxiliary class for updating data associated with SparseTable. + */ +public class SparseTableContentUpdater extends SparseTableContentReader { + protected OutputStream mLookupTableOutStream; + protected OutputStream[] mAddressTableOutStreams; + protected OutputStream[] mContentOutStreams; + + public SparseTableContentUpdater(final String name, final int blockSize, + final File baseDir, final String[] contentFilenames, final String[] contentIds, + final DictionaryBufferFactory factory) { + super(name, blockSize, baseDir, contentFilenames, contentIds, factory); + mAddressTableOutStreams = new OutputStream[mContentCount]; + mContentOutStreams = new OutputStream[mContentCount]; + } + + protected void openStreamsAndBuffers() throws IOException { + openBuffers(); + mLookupTableOutStream = new FileOutputStream(mLookupTableFile, true /* append */); + for (int i = 0; i < mContentCount; ++i) { + mAddressTableOutStreams[i] = new FileOutputStream(mAddressTableFiles[i], + true /* append */); + mContentOutStreams[i] = new FileOutputStream(mContentFiles[i], true /* append */); + } + } + + /** + * Set the contentIndex-th elements of contentId-th table. + * + * @param contentId the id of the content table. + * @param contentIndex the index where to set the valie. + * @param value the value to set. + */ + protected void setContentValue(final int contentId, final int contentIndex, final int value) + throws IOException { + if ((contentIndex / mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES + >= mLookupTableBuffer.limit()) { + // Need to extend the lookup table + final int currentSize = mLookupTableBuffer.limit() + / SparseTable.SIZE_OF_INT_IN_BYTES; + final int target = contentIndex / mBlockSize + 1; + for (int i = currentSize; i < target; ++i) { + BinaryDictEncoderUtils.writeUIntToStream(mLookupTableOutStream, + SparseTable.NOT_EXIST, SparseTable.SIZE_OF_INT_IN_BYTES); + } + // We need to reopen the byte buffer of the lookup table because a MappedByteBuffer in + // Java isn't expanded automatically when the underlying file is expanded. + reopenLookupTable(); + } + + mLookupTableBuffer.position((contentIndex / mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES); + int posInAddressTable = mLookupTableBuffer.readInt(); + if (posInAddressTable == SparseTable.NOT_EXIST) { + // Need to extend the address table + mLookupTableBuffer.position(mLookupTableBuffer.position() + - SparseTable.SIZE_OF_INT_IN_BYTES); + posInAddressTable = mAddressTableBuffers[0].limit() / mBlockSize; + BinaryDictEncoderUtils.writeUIntToDictBuffer(mLookupTableBuffer, + posInAddressTable, SparseTable.SIZE_OF_INT_IN_BYTES); + for (int i = 0; i < mContentCount; ++i) { + for (int j = 0; j < mBlockSize; ++j) { + BinaryDictEncoderUtils.writeUIntToStream(mAddressTableOutStreams[i], + SparseTable.NOT_EXIST, SparseTable.SIZE_OF_INT_IN_BYTES); + } + } + // We need to reopen the byte buffers of the address tables because a MappedByteBuffer + // in Java isn't expanded automatically when the underlying file is expanded. + reopenAddressTables(); + } + posInAddressTable += (contentIndex % mBlockSize) * SparseTable.SIZE_OF_INT_IN_BYTES; + + mAddressTableBuffers[contentId].position(posInAddressTable); + BinaryDictEncoderUtils.writeUIntToDictBuffer(mAddressTableBuffers[contentId], + value, SparseTable.SIZE_OF_INT_IN_BYTES); + } + + private void reopenLookupTable() throws IOException { + mLookupTableOutStream.flush(); + mLookupTableBuffer = mFactory.getDictionaryBuffer(mLookupTableFile); + } + + private void reopenAddressTables() throws IOException { + for (int i = 0; i < mContentCount; ++i) { + mAddressTableOutStreams[i].flush(); + mAddressTableBuffers[i] = mFactory.getDictionaryBuffer(mAddressTableFiles[i]); + } + } + + protected void close() throws IOException { + mLookupTableOutStream.close(); + for (final OutputStream stream : mAddressTableOutStreams) { + stream.close(); + } + for (final OutputStream stream : mContentOutStreams) { + stream.close(); + } + } +} diff --git a/java/src/com/android/inputmethod/latin/makedict/SparseTableContentWriter.java b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentWriter.java new file mode 100644 index 000000000..49f0fd624 --- /dev/null +++ b/java/src/com/android/inputmethod/latin/makedict/SparseTableContentWriter.java @@ -0,0 +1,93 @@ +/* + * Copyright (C) 2013 The Android Open Source Project + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package com.android.inputmethod.latin.makedict; + +import java.io.File; +import java.io.FileNotFoundException; +import java.io.FileOutputStream; +import java.io.IOException; +import java.io.OutputStream; + +/** + * An auxiliary class for writing data associated with SparseTable to files. + */ +public class SparseTableContentWriter { + public interface SparseTableContentWriterInterface { + public void write(final OutputStream outStream) throws IOException; + } + + private final int mContentCount; + private final SparseTable mSparseTable; + private final File mLookupTableFile; + protected final File mBaseDir; + private final File[] mAddressTableFiles; + private final File[] mContentFiles; + protected final OutputStream[] mContentOutStreams; + + /** + * Sole constructor of SparseTableContentWriter. + * + * @param name the name of SparseTable. + * @param initialCapacity the initial capacity of SparseTable. + * @param blockSize the block size of the content table. + * @param baseDir the directory which contains the files of the content table. + * @param contentFilenames the file names of content files. + * @param contentIds the ids of contents. These ids are used for a suffix of a name of address + * files and content files. + */ + public SparseTableContentWriter(final String name, final int initialCapacity, + final int blockSize, final File baseDir, final String[] contentFilenames, + final String[] contentIds) { + if (contentFilenames.length != contentIds.length) { + throw new RuntimeException("The length of contentFilenames and the length of" + + " contentIds are different " + contentFilenames.length + ", " + + contentIds.length); + } + mContentCount = contentFilenames.length; + mSparseTable = new SparseTable(initialCapacity, blockSize, mContentCount); + mLookupTableFile = new File(baseDir, name + FormatSpec.LOOKUP_TABLE_FILE_SUFFIX); + mAddressTableFiles = new File[mContentCount]; + mContentFiles = new File[mContentCount]; + mBaseDir = baseDir; + for (int i = 0; i < mContentCount; ++i) { + mAddressTableFiles[i] = new File(mBaseDir, + name + FormatSpec.CONTENT_TABLE_FILE_SUFFIX + contentIds[i]); + mContentFiles[i] = new File(mBaseDir, contentFilenames[i] + contentIds[i]); + } + mContentOutStreams = new OutputStream[mContentCount]; + } + + public void openStreams() throws FileNotFoundException { + for (int i = 0; i < mContentCount; ++i) { + mContentOutStreams[i] = new FileOutputStream(mContentFiles[i]); + } + } + + protected void write(final int contentIndex, final int index, + final SparseTableContentWriterInterface writer) throws IOException { + mSparseTable.set(contentIndex, index, (int) mContentFiles[contentIndex].length()); + writer.write(mContentOutStreams[contentIndex]); + mContentOutStreams[contentIndex].flush(); + } + + public void closeStreams() throws IOException { + mSparseTable.writeToFiles(mLookupTableFile, mAddressTableFiles); + for (int i = 0; i < mContentCount; ++i) { + mContentOutStreams[i].close(); + } + } +}
\ No newline at end of file diff --git a/java/src/com/android/inputmethod/latin/makedict/Ver4DictDecoder.java b/java/src/com/android/inputmethod/latin/makedict/Ver4DictDecoder.java index 53729075f..f0fed3fda 100644 --- a/java/src/com/android/inputmethod/latin/makedict/Ver4DictDecoder.java +++ b/java/src/com/android/inputmethod/latin/makedict/Ver4DictDecoder.java @@ -40,21 +40,45 @@ import java.util.Arrays; public class Ver4DictDecoder extends AbstractDictDecoder { private static final String TAG = Ver4DictDecoder.class.getSimpleName(); - private static final int FILETYPE_TRIE = 1; - private static final int FILETYPE_FREQUENCY = 2; - private static final int FILETYPE_TERMINAL_ADDRESS_TABLE = 3; - private static final int FILETYPE_BIGRAM_FREQ = 4; - private static final int FILETYPE_SHORTCUT = 5; - - private final File mDictDirectory; - private final DictionaryBufferFactory mBufferFactory; + protected static final int FILETYPE_TRIE = 1; + protected static final int FILETYPE_FREQUENCY = 2; + protected static final int FILETYPE_TERMINAL_ADDRESS_TABLE = 3; + protected static final int FILETYPE_BIGRAM_FREQ = 4; + protected static final int FILETYPE_SHORTCUT = 5; + + protected final File mDictDirectory; + protected final DictionaryBufferFactory mBufferFactory; protected DictBuffer mDictBuffer; - private DictBuffer mFrequencyBuffer; - private DictBuffer mTerminalAddressTableBuffer; - private DictBuffer mBigramBuffer; - private DictBuffer mShortcutBuffer; - private SparseTable mBigramAddressTable; - private SparseTable mShortcutAddressTable; + protected DictBuffer mFrequencyBuffer; + protected DictBuffer mTerminalAddressTableBuffer; + private BigramContentReader mBigramReader; + private ShortcutContentReader mShortcutReader; + + /** + * Raw PtNode info straight out of a trie file in version 4 dictionary. + */ + protected static final class Ver4PtNodeInfo { + public final int mFlags; + public final int[] mCharacters; + public final int mTerminalId; + public final int mChildrenPos; + public final int mParentPos; + public final int mNodeSize; + public int mStartIndexOfCharacters; + public int mEndIndexOfCharacters; // exclusive + + public Ver4PtNodeInfo(final int flags, final int[] characters, final int terminalId, + final int childrenPos, final int parentPos, final int nodeSize) { + mFlags = flags; + mCharacters = characters; + mTerminalId = terminalId; + mChildrenPos = childrenPos; + mParentPos = parentPos; + mNodeSize = nodeSize; + mStartIndexOfCharacters = 0; + mEndIndexOfCharacters = characters.length; + } + } @UsedForTesting /* package */ Ver4DictDecoder(final File dictDirectory, final int factoryFlag) { @@ -79,7 +103,7 @@ public class Ver4DictDecoder extends AbstractDictDecoder { mDictBuffer = mFrequencyBuffer = null; } - private File getFile(final int fileType) { + protected File getFile(final int fileType) { if (fileType == FILETYPE_TRIE) { return new File(mDictDirectory, mDictDirectory.getName() + FormatSpec.TRIE_FILE_EXTENSION); @@ -108,10 +132,12 @@ public class Ver4DictDecoder extends AbstractDictDecoder { mFrequencyBuffer = mBufferFactory.getDictionaryBuffer(getFile(FILETYPE_FREQUENCY)); mTerminalAddressTableBuffer = mBufferFactory.getDictionaryBuffer( getFile(FILETYPE_TERMINAL_ADDRESS_TABLE)); - mBigramBuffer = mBufferFactory.getDictionaryBuffer(getFile(FILETYPE_BIGRAM_FREQ)); - loadBigramAddressSparseTable(); - mShortcutBuffer = mBufferFactory.getDictionaryBuffer(getFile(FILETYPE_SHORTCUT)); - loadShortcutAddressSparseTable(); + mBigramReader = new BigramContentReader(mDictDirectory.getName(), + mDictDirectory, mBufferFactory, false); + mBigramReader.openBuffers(); + mShortcutReader = new ShortcutContentReader(mDictDirectory.getName(), mDictDirectory, + mBufferFactory); + mShortcutReader.openBuffers(); } @Override @@ -119,6 +145,7 @@ public class Ver4DictDecoder extends AbstractDictDecoder { return mDictBuffer != null; } + @UsedForTesting /* package */ DictBuffer getDictBuffer() { return mDictBuffer; } @@ -136,25 +163,113 @@ public class Ver4DictDecoder extends AbstractDictDecoder { return header; } - private void loadBigramAddressSparseTable() throws IOException { - final File lookupIndexFile = new File(mDictDirectory, mDictDirectory.getName() - + FormatSpec.BIGRAM_FILE_EXTENSION + FormatSpec.LOOKUP_TABLE_FILE_SUFFIX); - final File freqsFile = new File(mDictDirectory, mDictDirectory.getName() - + FormatSpec.BIGRAM_FILE_EXTENSION + FormatSpec.CONTENT_TABLE_FILE_SUFFIX - + FormatSpec.BIGRAM_FREQ_CONTENT_ID); - mBigramAddressTable = SparseTable.readFromFiles(lookupIndexFile, new File[] { freqsFile }, - FormatSpec.BIGRAM_ADDRESS_TABLE_BLOCK_SIZE); + /** + * An auxiliary class for reading bigrams. + */ + protected static class BigramContentReader extends SparseTableContentReader { + private final boolean mHasTimestamp; + + public BigramContentReader(final String name, final File baseDir, + final DictionaryBufferFactory factory, final boolean hasTimestamp) { + super(name + FormatSpec.BIGRAM_FILE_EXTENSION, + FormatSpec.BIGRAM_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + getContentFilenames(name, hasTimestamp), getContentIds(hasTimestamp), factory); + mHasTimestamp = hasTimestamp; + } + + // TODO: Consolidate this method and BigramContentWriter.getContentFilenames. + protected static String[] getContentFilenames(final String name, + final boolean hasTimestamp) { + final String[] contentFilenames; + if (hasTimestamp) { + contentFilenames = new String[] { name + FormatSpec.BIGRAM_FILE_EXTENSION, + name + FormatSpec.BIGRAM_FILE_EXTENSION }; + } else { + contentFilenames = new String[] { name + FormatSpec.BIGRAM_FILE_EXTENSION }; + } + return contentFilenames; + } + + // TODO: Consolidate this method and BigramContentWriter.getContentIds. + protected static String[] getContentIds(final boolean hasTimestamp) { + final String[] contentIds; + if (hasTimestamp) { + contentIds = new String[] { FormatSpec.BIGRAM_FREQ_CONTENT_ID, + FormatSpec.BIGRAM_TIMESTAMP_CONTENT_ID }; + } else { + contentIds = new String[] { FormatSpec.BIGRAM_FREQ_CONTENT_ID }; + } + return contentIds; + } + + public ArrayList<PendingAttribute> readTargetsAndFrequencies(final int terminalId, + final DictBuffer terminalAddressTableBuffer) { + final ArrayList<PendingAttribute> bigrams = CollectionUtils.newArrayList(); + read(FormatSpec.BIGRAM_FREQ_CONTENT_INDEX, terminalId, + new SparseTableContentReaderInterface() { + @Override + public void read(final DictBuffer buffer) { + while (bigrams.size() < FormatSpec.MAX_BIGRAMS_IN_A_PTNODE) { + // If bigrams.size() reaches FormatSpec.MAX_BIGRAMS_IN_A_PTNODE, + // remaining bigram entries are ignored. + final int bigramFlags = buffer.readUnsignedByte(); + final int targetTerminalId = buffer.readUnsignedInt24(); + terminalAddressTableBuffer.position(targetTerminalId + * FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE); + final int targetAddress = + terminalAddressTableBuffer.readUnsignedInt24(); + bigrams.add(new PendingAttribute(bigramFlags + & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY, + targetAddress)); + if (0 == (bigramFlags + & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) { + break; + } + } + if (bigrams.size() >= FormatSpec.MAX_BIGRAMS_IN_A_PTNODE) { + throw new RuntimeException("Too many bigrams in a PtNode (" + + bigrams.size() + " but max is " + + FormatSpec.MAX_BIGRAMS_IN_A_PTNODE + ")"); + } + } + }); + if (bigrams.isEmpty()) return null; + return bigrams; + } } - // TODO: Let's have something like SparseTableContentsReader in this class. - private void loadShortcutAddressSparseTable() throws IOException { - final File lookupIndexFile = new File(mDictDirectory, mDictDirectory.getName() - + FormatSpec.SHORTCUT_FILE_EXTENSION + FormatSpec.LOOKUP_TABLE_FILE_SUFFIX); - final File contentFile = new File(mDictDirectory, mDictDirectory.getName() - + FormatSpec.SHORTCUT_FILE_EXTENSION + FormatSpec.CONTENT_TABLE_FILE_SUFFIX - + FormatSpec.SHORTCUT_CONTENT_ID); - mShortcutAddressTable = SparseTable.readFromFiles(lookupIndexFile, - new File[] { contentFile }, FormatSpec.SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE); + /** + * An auxiliary class for reading shortcuts. + */ + protected static class ShortcutContentReader extends SparseTableContentReader { + public ShortcutContentReader(final String name, final File baseDir, + final DictionaryBufferFactory factory) { + super(name + FormatSpec.SHORTCUT_FILE_EXTENSION, + FormatSpec.SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + new String[] { name + FormatSpec.SHORTCUT_FILE_EXTENSION }, + new String[] { FormatSpec.SHORTCUT_CONTENT_ID }, factory); + } + + public ArrayList<WeightedString> readShortcuts(final int terminalId) { + final ArrayList<WeightedString> shortcuts = CollectionUtils.newArrayList(); + read(FormatSpec.SHORTCUT_CONTENT_INDEX, terminalId, + new SparseTableContentReaderInterface() { + @Override + public void read(final DictBuffer buffer) { + while (true) { + final int flags = buffer.readUnsignedByte(); + final String word = CharEncoding.readString(buffer); + shortcuts.add(new WeightedString(word, + flags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY)); + if (0 == (flags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) { + break; + } + } + } + }); + if (shortcuts.isEmpty()) return null; + return shortcuts; + } } protected static class PtNodeReader extends AbstractDictDecoder.PtNodeReader { @@ -168,102 +283,82 @@ public class Ver4DictDecoder extends AbstractDictDecoder { } } - private ArrayList<WeightedString> readShortcuts(final int terminalId) { - if (mShortcutAddressTable.get(0, terminalId) == SparseTable.NOT_EXIST) return null; - - final ArrayList<WeightedString> ret = CollectionUtils.newArrayList(); - final int posOfShortcuts = mShortcutAddressTable.get(FormatSpec.SHORTCUT_CONTENT_INDEX, - terminalId); - mShortcutBuffer.position(posOfShortcuts); - while (true) { - final int flags = mShortcutBuffer.readUnsignedByte(); - final String word = CharEncoding.readString(mShortcutBuffer); - ret.add(new WeightedString(word, - flags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY)); - if (0 == (flags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) break; - } - return ret; - } + private final int[] mCharacterBufferForReadingVer4PtNodeInfo + = new int[FormatSpec.MAX_WORD_LENGTH]; + /** + * Reads PtNode from ptNodePos in the trie file and returns Ver4PtNodeInfo. + * + * @param ptNodePos the position of PtNode. + * @param options the format options. + * @return Ver4PtNodeInfo. + */ // TODO: Make this buffer thread safe. // TODO: Support words longer than FormatSpec.MAX_WORD_LENGTH. - private final int[] mCharacterBuffer = new int[FormatSpec.MAX_WORD_LENGTH]; - @Override - public PtNodeInfo readPtNode(int ptNodePos, FormatOptions options) { - int addressPointer = ptNodePos; + protected Ver4PtNodeInfo readVer4PtNodeInfo(final int ptNodePos, final FormatOptions options) { + int readingPos = ptNodePos; final int flags = PtNodeReader.readPtNodeOptionFlags(mDictBuffer); - addressPointer += FormatSpec.PTNODE_FLAGS_SIZE; + readingPos += FormatSpec.PTNODE_FLAGS_SIZE; - final int parentAddress = PtNodeReader.readParentAddress(mDictBuffer, options); + final int parentPos = PtNodeReader.readParentAddress(mDictBuffer, options); if (BinaryDictIOUtils.supportsDynamicUpdate(options)) { - addressPointer += FormatSpec.PARENT_ADDRESS_SIZE; + readingPos += FormatSpec.PARENT_ADDRESS_SIZE; } final int characters[]; if (0 != (flags & FormatSpec.FLAG_HAS_MULTIPLE_CHARS)) { int index = 0; int character = CharEncoding.readChar(mDictBuffer); - addressPointer += CharEncoding.getCharSize(character); + readingPos += CharEncoding.getCharSize(character); while (FormatSpec.INVALID_CHARACTER != character && index < FormatSpec.MAX_WORD_LENGTH) { - mCharacterBuffer[index++] = character; + mCharacterBufferForReadingVer4PtNodeInfo[index++] = character; character = CharEncoding.readChar(mDictBuffer); - addressPointer += CharEncoding.getCharSize(character); + readingPos += CharEncoding.getCharSize(character); } - characters = Arrays.copyOfRange(mCharacterBuffer, 0, index); + characters = Arrays.copyOfRange(mCharacterBufferForReadingVer4PtNodeInfo, 0, index); } else { final int character = CharEncoding.readChar(mDictBuffer); - addressPointer += CharEncoding.getCharSize(character); + readingPos += CharEncoding.getCharSize(character); characters = new int[] { character }; } final int terminalId; if (0 != (FormatSpec.FLAG_IS_TERMINAL & flags)) { terminalId = PtNodeReader.readTerminalId(mDictBuffer); - addressPointer += FormatSpec.PTNODE_TERMINAL_ID_SIZE; + readingPos += FormatSpec.PTNODE_TERMINAL_ID_SIZE; } else { terminalId = PtNode.NOT_A_TERMINAL; } + int childrenPos = PtNodeReader.readChildrenAddress(mDictBuffer, flags, options); + if (childrenPos != FormatSpec.NO_CHILDREN_ADDRESS) { + childrenPos += readingPos; + } + readingPos += BinaryDictIOUtils.getChildrenAddressSize(flags, options); + + return new Ver4PtNodeInfo(flags, characters, terminalId, childrenPos, parentPos, + readingPos - ptNodePos); + } + + @Override + public PtNodeInfo readPtNode(int ptNodePos, FormatOptions options) { + final Ver4PtNodeInfo nodeInfo = readVer4PtNodeInfo(ptNodePos, options); + final int frequency; - if (0 != (FormatSpec.FLAG_IS_TERMINAL & flags)) { - frequency = PtNodeReader.readFrequency(mFrequencyBuffer, terminalId); + if (0 != (FormatSpec.FLAG_IS_TERMINAL & nodeInfo.mFlags)) { + frequency = PtNodeReader.readFrequency(mFrequencyBuffer, nodeInfo.mTerminalId); } else { frequency = PtNode.NOT_A_TERMINAL; } - int childrenAddress = PtNodeReader.readChildrenAddress(mDictBuffer, flags, options); - if (childrenAddress != FormatSpec.NO_CHILDREN_ADDRESS) { - childrenAddress += addressPointer; - } - addressPointer += BinaryDictIOUtils.getChildrenAddressSize(flags, options); - final ArrayList<WeightedString> shortcutTargets = readShortcuts(terminalId); - - final ArrayList<PendingAttribute> bigrams; - if (0 != (flags & FormatSpec.FLAG_HAS_BIGRAMS)) { - bigrams = new ArrayList<PendingAttribute>(); - final int posOfBigrams = mBigramAddressTable.get(0 /* contentTableIndex */, terminalId); - mBigramBuffer.position(posOfBigrams); - while (bigrams.size() < FormatSpec.MAX_BIGRAMS_IN_A_PTNODE) { - // If bigrams.size() reaches FormatSpec.MAX_BIGRAMS_IN_A_PTNODE, - // remaining bigram entries are ignored. - final int bigramFlags = mBigramBuffer.readUnsignedByte(); - final int targetTerminalId = mBigramBuffer.readUnsignedInt24(); - mTerminalAddressTableBuffer.position( - targetTerminalId * FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE); - final int targetAddress = mTerminalAddressTableBuffer.readUnsignedInt24(); - bigrams.add(new PendingAttribute( - bigramFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY, - targetAddress)); - if (0 == (bigramFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) break; - } - if (bigrams.size() >= FormatSpec.MAX_BIGRAMS_IN_A_PTNODE) { - throw new RuntimeException("Too many bigrams in a PtNode (" + bigrams.size() - + " but max is " + FormatSpec.MAX_BIGRAMS_IN_A_PTNODE + ")"); - } - } else { - bigrams = null; - } - return new PtNodeInfo(ptNodePos, addressPointer, flags, characters, frequency, - parentAddress, childrenAddress, shortcutTargets, bigrams); + + final ArrayList<WeightedString> shortcutTargets = mShortcutReader.readShortcuts( + nodeInfo.mTerminalId); + final ArrayList<PendingAttribute> bigrams = mBigramReader.readTargetsAndFrequencies( + nodeInfo.mTerminalId, mTerminalAddressTableBuffer); + + return new PtNodeInfo(ptNodePos, ptNodePos + nodeInfo.mNodeSize, nodeInfo.mFlags, + nodeInfo.mCharacters, frequency, nodeInfo.mParentPos, nodeInfo.mChildrenPos, + shortcutTargets, bigrams); } private void deleteDictFiles() { @@ -314,10 +409,14 @@ public class Ver4DictDecoder extends AbstractDictDecoder { @Override public boolean readAndFollowForwardLink() { - final int nextAddress = mDictBuffer.readUnsignedInt24(); - if (nextAddress >= 0 && nextAddress < mDictBuffer.limit()) { - mDictBuffer.position(nextAddress); - return true; + final int forwardLinkPos = mDictBuffer.position(); + int nextRelativePos = BinaryDictDecoderUtils.readSInt24(mDictBuffer); + if (nextRelativePos != FormatSpec.NO_FORWARD_LINK_ADDRESS) { + final int nextPos = forwardLinkPos + nextRelativePos; + if (nextPos >= 0 && nextPos < mDictBuffer.limit()) { + mDictBuffer.position(nextPos); + return true; + } } return false; } diff --git a/java/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java b/java/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java index f9dcacf77..4b3acdc8e 100644 --- a/java/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java +++ b/java/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java @@ -25,6 +25,7 @@ import com.android.inputmethod.latin.makedict.FusionDictionary.DictionaryOptions import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNodeArray; import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString; +import com.android.inputmethod.latin.utils.CollectionUtils; import java.io.File; import java.io.FileNotFoundException; @@ -32,6 +33,8 @@ import java.io.FileOutputStream; import java.io.IOException; import java.io.OutputStream; import java.util.ArrayList; +import java.util.Collections; +import java.util.Comparator; import java.util.Iterator; /** @@ -45,6 +48,7 @@ public class Ver4DictEncoder implements DictEncoder { private int mHeaderSize; private OutputStream mTrieOutStream; private OutputStream mFreqOutStream; + private OutputStream mUnigramTimestampOutStream; private OutputStream mTerminalAddressTableOutStream; private File mDictDir; private String mBaseFilename; @@ -56,73 +60,41 @@ public class Ver4DictEncoder implements DictEncoder { mDictPlacedDir = dictPlacedDir; } - private interface SparseTableContentWriterInterface { - public void write(final OutputStream outStream) throws IOException; - } + private static class BigramContentWriter extends SparseTableContentWriter { + private final boolean mWriteTimestamp; - private static class SparseTableContentWriter { - private final int mContentCount; - private final SparseTable mSparseTable; - private final File mLookupTableFile; - protected final File mBaseDir; - private final File[] mAddressTableFiles; - private final File[] mContentFiles; - protected final OutputStream[] mContentOutStreams; - - public SparseTableContentWriter(final String name, final int contentCount, - final int initialCapacity, final int blockSize, final File baseDir, - final String[] contentFilenames, final String[] contentIds) { - if (contentFilenames.length != contentIds.length) { - throw new RuntimeException("The length of contentFilenames and the length of" - + " contentIds are different " + contentFilenames.length + ", " - + contentIds.length); - } - mContentCount = contentCount; - mSparseTable = new SparseTable(initialCapacity, blockSize, contentCount); - mLookupTableFile = new File(baseDir, name + FormatSpec.LOOKUP_TABLE_FILE_SUFFIX); - mAddressTableFiles = new File[mContentCount]; - mContentFiles = new File[mContentCount]; - mBaseDir = baseDir; - for (int i = 0; i < mContentCount; ++i) { - mAddressTableFiles[i] = new File(mBaseDir, - name + FormatSpec.CONTENT_TABLE_FILE_SUFFIX + contentIds[i]); - mContentFiles[i] = new File(mBaseDir, contentFilenames[i] + contentIds[i]); - } - mContentOutStreams = new OutputStream[mContentCount]; + public BigramContentWriter(final String name, final int initialCapacity, + final File baseDir, final boolean writeTimestamp) { + super(name + FormatSpec.BIGRAM_FILE_EXTENSION, initialCapacity, + FormatSpec.BIGRAM_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + getContentFilenames(name, writeTimestamp), getContentIds(writeTimestamp)); + mWriteTimestamp = writeTimestamp; } - public void openStreams() throws FileNotFoundException { - for (int i = 0; i < mContentCount; ++i) { - mContentOutStreams[i] = new FileOutputStream(mContentFiles[i]); + private static String[] getContentFilenames(final String name, + final boolean writeTimestamp) { + final String[] contentFilenames; + if (writeTimestamp) { + contentFilenames = new String[] { name + FormatSpec.BIGRAM_FILE_EXTENSION, + name + FormatSpec.BIGRAM_FILE_EXTENSION }; + } else { + contentFilenames = new String[] { name + FormatSpec.BIGRAM_FILE_EXTENSION }; } + return contentFilenames; } - protected void write(final int contentIndex, final int index, - final SparseTableContentWriterInterface writer) throws IOException { - mSparseTable.set(contentIndex, index, (int) mContentFiles[contentIndex].length()); - writer.write(mContentOutStreams[contentIndex]); - mContentOutStreams[contentIndex].flush(); - } - - public void closeStreams() throws IOException { - mSparseTable.writeToFiles(mLookupTableFile, mAddressTableFiles); - for (int i = 0; i < mContentCount; ++i) { - mContentOutStreams[i].close(); + private static String[] getContentIds(final boolean writeTimestamp) { + final String[] contentIds; + if (writeTimestamp) { + contentIds = new String[] { FormatSpec.BIGRAM_FREQ_CONTENT_ID, + FormatSpec.BIGRAM_TIMESTAMP_CONTENT_ID }; + } else { + contentIds = new String[] { FormatSpec.BIGRAM_FREQ_CONTENT_ID }; } - } - } - - private static class BigramContentWriter extends SparseTableContentWriter { - - public BigramContentWriter(final String name, final int initialCapacity, - final File baseDir) { - super(name + FormatSpec.BIGRAM_FILE_EXTENSION, FormatSpec.BIGRAM_CONTENT_COUNT, - initialCapacity, FormatSpec.BIGRAM_ADDRESS_TABLE_BLOCK_SIZE, baseDir, - new String[] { name + FormatSpec.BIGRAM_FILE_EXTENSION }, - new String[] { FormatSpec.BIGRAM_FREQ_CONTENT_ID }); + return contentIds; } - public void writeBigramsForOneWord(final int terminalId, + public void writeBigramsForOneWord(final int terminalId, final int bigramCount, final Iterator<WeightedString> bigramIterator, final FusionDictionary dict) throws IOException { write(FormatSpec.BIGRAM_FREQ_CONTENT_INDEX, terminalId, @@ -130,8 +102,16 @@ public class Ver4DictEncoder implements DictEncoder { @Override public void write(final OutputStream outStream) throws IOException { writeBigramsForOneWordInternal(outStream, bigramIterator, dict); - } - }); + }}); + if (mWriteTimestamp) { + write(FormatSpec.BIGRAM_TIMESTAMP_CONTENT_INDEX, terminalId, + new SparseTableContentWriterInterface() { + @Override + public void write(final OutputStream outStream) throws IOException { + initBigramTimestampsCountersAndLevelsForOneWordInternal(outStream, + bigramCount); + }}); + } } private void writeBigramsForOneWordInternal(final OutputStream outStream, @@ -151,13 +131,26 @@ public class Ver4DictEncoder implements DictEncoder { FormatSpec.PTNODE_ATTRIBUTE_MAX_ADDRESS_SIZE); } } + + private void initBigramTimestampsCountersAndLevelsForOneWordInternal( + final OutputStream outStream, final int bigramCount) throws IOException { + for (int i = 0; i < bigramCount; ++i) { + // TODO: Figure out what initial values should be. + BinaryDictEncoderUtils.writeUIntToStream(outStream, 0 /* value */, + FormatSpec.BIGRAM_TIMESTAMP_SIZE); + BinaryDictEncoderUtils.writeUIntToStream(outStream, 0 /* value */, + FormatSpec.BIGRAM_COUNTER_SIZE); + BinaryDictEncoderUtils.writeUIntToStream(outStream, 0 /* value */, + FormatSpec.BIGRAM_LEVEL_SIZE); + } + } } private static class ShortcutContentWriter extends SparseTableContentWriter { public ShortcutContentWriter(final String name, final int initialCapacity, final File baseDir) { - super(name + FormatSpec.SHORTCUT_FILE_EXTENSION, FormatSpec.SHORTCUT_CONTENT_COUNT, - initialCapacity, FormatSpec.SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + super(name + FormatSpec.SHORTCUT_FILE_EXTENSION, initialCapacity, + FormatSpec.SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE, baseDir, new String[] { name + FormatSpec.SHORTCUT_FILE_EXTENSION }, new String[] { FormatSpec.SHORTCUT_CONTENT_ID }); } @@ -193,18 +186,20 @@ public class Ver4DictEncoder implements DictEncoder { mDictDir = new File(mDictPlacedDir, mBaseFilename); final File trieFile = new File(mDictDir, mBaseFilename + FormatSpec.TRIE_FILE_EXTENSION); final File freqFile = new File(mDictDir, mBaseFilename + FormatSpec.FREQ_FILE_EXTENSION); + final File timestampFile = new File(mDictDir, + mBaseFilename + FormatSpec.UNIGRAM_TIMESTAMP_FILE_EXTENSION); final File terminalAddressTableFile = new File(mDictDir, mBaseFilename + FormatSpec.TERMINAL_ADDRESS_TABLE_FILE_EXTENSION); if (!mDictDir.isDirectory()) { if (mDictDir.exists()) mDictDir.delete(); mDictDir.mkdirs(); } - if (!trieFile.exists()) trieFile.createNewFile(); - if (!freqFile.exists()) freqFile.createNewFile(); - if (!terminalAddressTableFile.exists()) terminalAddressTableFile.createNewFile(); mTrieOutStream = new FileOutputStream(trieFile); mFreqOutStream = new FileOutputStream(freqFile); mTerminalAddressTableOutStream = new FileOutputStream(terminalAddressTableFile); + if (formatOptions.mHasTimestamp) { + mUnigramTimestampOutStream = new FileOutputStream(timestampFile); + } } private void close() throws IOException { @@ -218,6 +213,9 @@ public class Ver4DictEncoder implements DictEncoder { if (mTerminalAddressTableOutStream != null) { mTerminalAddressTableOutStream.close(); } + if (mUnigramTimestampOutStream != null) { + mUnigramTimestampOutStream.close(); + } } finally { mTrieOutStream = null; mFreqOutStream = null; @@ -246,10 +244,29 @@ public class Ver4DictEncoder implements DictEncoder { MakedictLog.i("Flattening the tree..."); ArrayList<PtNodeArray> flatNodes = BinaryDictEncoderUtils.flattenTree(dict.mRootNodeArray); int terminalCount = 0; + final ArrayList<PtNode> nodes = CollectionUtils.newArrayList(); for (final PtNodeArray array : flatNodes) { for (final PtNode node : array.mData) { - if (node.isTerminal()) node.mTerminalId = terminalCount++; + if (node.isTerminal()) { + nodes.add(node); + node.mTerminalId = terminalCount++; + } + } + } + Collections.sort(nodes, new Comparator<PtNode>() { + @Override + public int compare(final PtNode lhs, final PtNode rhs) { + if (lhs.mFrequency != rhs.mFrequency) { + return lhs.mFrequency < rhs.mFrequency ? -1 : 1; + } + if (lhs.mTerminalId < rhs.mTerminalId) return -1; + if (lhs.mTerminalId > rhs.mTerminalId) return 1; + return 0; } + }); + int count = 0; + for (final PtNode node : nodes) { + node.mTerminalId = count++; } MakedictLog.i("Computing addresses..."); @@ -257,7 +274,11 @@ public class Ver4DictEncoder implements DictEncoder { if (MakedictLog.DBG) BinaryDictEncoderUtils.checkFlatPtNodeArrayList(flatNodes); writeTerminalData(flatNodes, terminalCount); - mBigramWriter = new BigramContentWriter(mBaseFilename, terminalCount, mDictDir); + if (formatOptions.mHasTimestamp) { + initUnigramTimestamps(terminalCount); + } + mBigramWriter = new BigramContentWriter(mBaseFilename, terminalCount, mDictDir, + formatOptions.mHasTimestamp); writeBigrams(flatNodes, dict); mShortcutWriter = new ShortcutContentWriter(mBaseFilename, terminalCount, mDictDir); writeShortcuts(flatNodes); @@ -348,7 +369,7 @@ public class Ver4DictEncoder implements DictEncoder { for (final PtNodeArray nodeArray : flatNodes) { for (final PtNode ptNode : nodeArray.mData) { if (ptNode.mBigrams != null) { - mBigramWriter.writeBigramsForOneWord(ptNode.mTerminalId, + mBigramWriter.writeBigramsForOneWord(ptNode.mTerminalId, ptNode.mBigrams.size(), ptNode.mBigrams.iterator(), dict); } } @@ -408,4 +429,11 @@ public class Ver4DictEncoder implements DictEncoder { mFreqOutStream.write(freqBuf); mTerminalAddressTableOutStream.write(terminalAddressTableBuf); } + + private void initUnigramTimestamps(final int terminalCount) throws IOException { + // Initial value of time stamps for each word is 0. + final byte[] unigramTimestampBuf = + new byte[terminalCount * FormatSpec.UNIGRAM_TIMESTAMP_SIZE]; + mUnigramTimestampOutStream.write(unigramTimestampBuf); + } } diff --git a/java/src/com/android/inputmethod/latin/makedict/Ver4DictUpdater.java b/java/src/com/android/inputmethod/latin/makedict/Ver4DictUpdater.java index 3d8f186ba..65860ee72 100644 --- a/java/src/com/android/inputmethod/latin/makedict/Ver4DictUpdater.java +++ b/java/src/com/android/inputmethod/latin/makedict/Ver4DictUpdater.java @@ -17,23 +17,124 @@ package com.android.inputmethod.latin.makedict; import com.android.inputmethod.annotations.UsedForTesting; +import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.CharEncoding; +import com.android.inputmethod.latin.makedict.FormatSpec.FileHeader; +import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions; +import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString; +import com.android.inputmethod.latin.utils.CollectionUtils; + +import android.util.Log; import java.io.File; +import java.io.FileOutputStream; import java.io.IOException; +import java.io.OutputStream; import java.util.ArrayList; +import java.util.Arrays; +import java.util.Iterator; /** * An implementation of DictUpdater for version 4 binary dictionary. */ @UsedForTesting public class Ver4DictUpdater extends Ver4DictDecoder implements DictUpdater { + private static final String TAG = Ver4DictUpdater.class.getSimpleName(); + + private OutputStream mDictStream; + private final File mFrequencyFile; @UsedForTesting public Ver4DictUpdater(final File dictDirectory, final int factoryType) { // DictUpdater must have an updatable DictBuffer. super(dictDirectory, ((factoryType & MASK_DICTBUFFER) == USE_BYTEARRAY) ? USE_BYTEARRAY : USE_WRITABLE_BYTEBUFFER); + mFrequencyFile = getFile(FILETYPE_FREQUENCY); + } + + private static class BigramContentUpdater extends SparseTableContentUpdater { + private final boolean mHasTimestamp; + + public BigramContentUpdater(final String name, final File baseDir, + final boolean hasTimestamp) { + super(name + FormatSpec.BIGRAM_FILE_EXTENSION, + FormatSpec.BIGRAM_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + BigramContentReader.getContentFilenames(name, hasTimestamp), + BigramContentReader.getContentIds(hasTimestamp), + new DictionaryBufferFromWritableByteBufferFactory()); + mHasTimestamp = hasTimestamp; + } + + public void insertBigramEntries(final int terminalId, final int frequency, + final ArrayList<PendingAttribute> entries) throws IOException { + if (terminalId < 0) { + throw new RuntimeException("Invalid terminal id : " + terminalId); + } + openStreamsAndBuffers(); + + if (entries == null || entries.isEmpty()) { + setContentValue(FormatSpec.BIGRAM_FREQ_CONTENT_INDEX, terminalId, + SparseTable.NOT_EXIST); + return; + } + final int positionOfEntries = + (int) mContentFiles[FormatSpec.BIGRAM_FREQ_CONTENT_INDEX].length(); + setContentValue(FormatSpec.BIGRAM_FREQ_CONTENT_INDEX, terminalId, positionOfEntries); + + final Iterator<PendingAttribute> bigramIterator = entries.iterator(); + while (bigramIterator.hasNext()) { + final PendingAttribute entry = bigramIterator.next(); + final int flags = BinaryDictEncoderUtils.makeBigramFlags(bigramIterator.hasNext(), + 0 /* offset */, entry.mFrequency, frequency, "" /* word */); + BinaryDictEncoderUtils.writeUIntToStream( + mContentOutStreams[FormatSpec.BIGRAM_FREQ_CONTENT_INDEX], flags, + FormatSpec.PTNODE_ATTRIBUTE_FLAGS_SIZE); + BinaryDictEncoderUtils.writeUIntToStream( + mContentOutStreams[FormatSpec.BIGRAM_FREQ_CONTENT_INDEX], entry.mAddress, + FormatSpec.PTNODE_ATTRIBUTE_MAX_ADDRESS_SIZE); + } + close(); + } + } + + private static class ShortcutContentUpdater extends SparseTableContentUpdater { + public ShortcutContentUpdater(final String name, final File baseDir) { + super(name + FormatSpec.SHORTCUT_FILE_EXTENSION, + FormatSpec.SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE, baseDir, + new String[] { name + FormatSpec.SHORTCUT_FILE_EXTENSION }, + new String[] { FormatSpec.SHORTCUT_CONTENT_ID }, + new DictionaryBufferFromWritableByteBufferFactory()); + } + + public void insertShortcuts(final int terminalId, + final ArrayList<WeightedString> shortcuts) throws IOException { + if (terminalId < 0) { + throw new RuntimeException("Invalid terminal id : " + terminalId); + } + openStreamsAndBuffers(); + if (shortcuts == null || shortcuts.isEmpty()) { + setContentValue(FormatSpec.SHORTCUT_CONTENT_INDEX, terminalId, + SparseTable.NOT_EXIST); + return; + } + + final int positionOfShortcuts = + (int) mContentFiles[FormatSpec.SHORTCUT_CONTENT_INDEX].length(); + setContentValue(FormatSpec.SHORTCUT_CONTENT_INDEX, terminalId, positionOfShortcuts); + + final Iterator<WeightedString> shortcutIterator = shortcuts.iterator(); + while (shortcutIterator.hasNext()) { + final WeightedString target = shortcutIterator.next(); + final int shortcutFlags = BinaryDictEncoderUtils.makeShortcutFlags( + shortcutIterator.hasNext(), target.mFrequency); + BinaryDictEncoderUtils.writeUIntToStream( + mContentOutStreams[FormatSpec.SHORTCUT_CONTENT_INDEX], shortcutFlags, + FormatSpec.PTNODE_ATTRIBUTE_FLAGS_SIZE); + CharEncoding.writeString(mContentOutStreams[FormatSpec.SHORTCUT_CONTENT_INDEX], + target.mWord); + } + close(); + } } @Override @@ -49,11 +150,622 @@ public class Ver4DictUpdater extends Ver4DictDecoder implements DictUpdater { } } - @Override + private int getNewTerminalId() { + // The size of frequency file is FormatSpec.FREQUENCY_AND_FLAGS_SIZE * number of terminals + // because each terminal always has a frequency. + // So we can get a fresh terminal id by this logic. + // CAVEAT: we are reading the file size from the disk each time: beware of race conditions, + // even on one thread. + return (int) (mFrequencyFile.length() / FormatSpec.FREQUENCY_AND_FLAGS_SIZE); + } + + private void updateParentPosIfNotMoved(final int nodePos, final int newParentPos, + final FormatOptions formatOptions) { + final int originalPos = getPosition(); + setPosition(nodePos); + final int flags = PtNodeReader.readPtNodeOptionFlags(mDictBuffer); + if (!BinaryDictIOUtils.isMovedPtNode(flags, formatOptions)) { + final int parentOffset = newParentPos - nodePos; + BinaryDictIOUtils.writeSInt24ToBuffer(mDictBuffer, parentOffset); + } + setPosition(originalPos); + } + + private void updateParentPositions(final int nodeArrayPos, final int newParentPos, + final FormatOptions formatOptions) { + final int originalPos = mDictBuffer.position(); + mDictBuffer.position(nodeArrayPos); + int jumpCount = 0; + do { + final int count = readPtNodeCount(); + for (int i = 0; i < count; ++i) { + updateParentPosIfNotMoved(getPosition(), newParentPos, formatOptions); + skipPtNode(formatOptions); + } + if (!readAndFollowForwardLink()) break; + } while (jumpCount++ < DynamicBinaryDictIOUtils.MAX_JUMPS); + setPosition(originalPos); + } + + private void updateChildrenPos(final int nodePos, final int newChildrenPos, + final FormatOptions options) { + final int originalPos = getPosition(); + setPosition(nodePos); + final int flags = PtNodeReader.readPtNodeOptionFlags(mDictBuffer); + PtNodeReader.readParentAddress(mDictBuffer, options); + BinaryDictIOUtils.skipString(mDictBuffer, + (flags & FormatSpec.FLAG_HAS_MULTIPLE_CHARS) != 0); + if ((flags & FormatSpec.FLAG_IS_TERMINAL) != 0) PtNodeReader.readTerminalId(mDictBuffer); + final int basePos = getPosition(); + BinaryDictIOUtils.writeSInt24ToBuffer(mDictBuffer, newChildrenPos - basePos); + setPosition(originalPos); + } + + private void updateTerminalPosition(final int terminalId, final int position) { + if (terminalId == PtNode.NOT_A_TERMINAL + || terminalId * FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE + >= mTerminalAddressTableBuffer.limit()) return; + mTerminalAddressTableBuffer.position(terminalId + * FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE); + BinaryDictEncoderUtils.writeUIntToDictBuffer(mTerminalAddressTableBuffer, position, + FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE); + } + + private void updateForwardLink(final int nodeArrayPos, final int newForwardLink, + final FormatOptions formatOptions) { + final int originalPos = getPosition(); + setPosition(nodeArrayPos); + int jumpCount = 0; + while (jumpCount++ < DynamicBinaryDictIOUtils.MAX_JUMPS) { + final int ptNodeCount = readPtNodeCount(); + for (int i = 0; i < ptNodeCount; ++i) { + skipPtNode(formatOptions); + } + final int forwardLinkPos = getPosition(); + if (!readAndFollowForwardLink()) { + setPosition(forwardLinkPos); + BinaryDictIOUtils.writeSInt24ToBuffer(mDictBuffer, newForwardLink - forwardLinkPos); + break; + } + } + setPosition(originalPos); + } + + private void markPtNodeAsMoved(final int nodePos, final int newNodePos, + final FormatOptions options) { + final int originalPos = getPosition(); + updateParentPosIfNotMoved(nodePos, newNodePos, options); + setPosition(nodePos); + final int currentFlags = PtNodeReader.readPtNodeOptionFlags(mDictBuffer); + setPosition(nodePos); + mDictBuffer.put((byte) (FormatSpec.FLAG_IS_MOVED + | (currentFlags & (~FormatSpec.MASK_MOVE_AND_DELETE_FLAG)))); + final int offset = newNodePos - nodePos; + BinaryDictIOUtils.writeSInt24ToBuffer(mDictBuffer, offset); + setPosition(originalPos); + } + + /** + * Writes a PtNode to an output stream from a Ver4PtNodeInfo. + * + * @param nodePos the position of the head of the PtNode. + * @param info the PtNode info to be written. + * @return the size written, in bytes. + */ + private int writePtNode(final int nodePos, final Ver4PtNodeInfo info) throws IOException { + int written = 0; + + // Write flags. + mDictStream.write((byte) (info.mFlags & 0xFF)); + written += FormatSpec.PTNODE_FLAGS_SIZE; + + // Write the parent position. + final int parentOffset = info.mParentPos == FormatSpec.NO_PARENT_ADDRESS ? + FormatSpec.NO_PARENT_ADDRESS : info.mParentPos - nodePos; + BinaryDictIOUtils.writeSInt24ToStream(mDictStream, parentOffset); + written += FormatSpec.PARENT_ADDRESS_SIZE; + + // Write a string. + if (((info.mFlags & FormatSpec.FLAG_HAS_MULTIPLE_CHARS) != 0) + != (info.mEndIndexOfCharacters - info.mStartIndexOfCharacters > 1)) { + throw new RuntimeException("Inconsistent flags : hasMultipleChars = " + + ((info.mFlags & FormatSpec.FLAG_HAS_MULTIPLE_CHARS) != 0) + ", length = " + + (info.mEndIndexOfCharacters - info.mStartIndexOfCharacters)); + } + written += CharEncoding.writeCodePoints(mDictStream, info.mCharacters, + info.mStartIndexOfCharacters, info.mEndIndexOfCharacters); + + // Write the terminal id. + if ((info.mFlags & FormatSpec.FLAG_IS_TERMINAL) != 0) { + BinaryDictEncoderUtils.writeUIntToStream(mDictStream, info.mTerminalId, + FormatSpec.PTNODE_TERMINAL_ID_SIZE); + written += FormatSpec.PTNODE_TERMINAL_ID_SIZE; + } + + // Write the children position. + final int childrenOffset = info.mChildrenPos == FormatSpec.NO_CHILDREN_ADDRESS + ? 0 : info.mChildrenPos - (nodePos + written); + BinaryDictIOUtils.writeSInt24ToStream(mDictStream, childrenOffset); + written += FormatSpec.SIGNED_CHILDREN_ADDRESS_SIZE; + + return written; + } + + /** + * Helper method to split and move PtNode. + * + * @param ptNodeArrayPos the position of PtNodeArray which contains the split and moved PtNode. + * @param splittedPtNodeToMovePos the position of the split and moved PtNode. + * @param newParent the parent PtNode after splitting. + * @param newChildren the children PtNodes after splitting. + * @param newParentStartPos where to write the new parent. + * @param formatOptions the format options. + */ + private void writeSplittedPtNodes(final int ptNodeArrayPos, final int splittedPtNodeToMovePos, + final Ver4PtNodeInfo newParent, final Ver4PtNodeInfo[] newChildren, + final int newParentStartPos, + final FormatOptions formatOptions) throws IOException { + updateTerminalPosition(newParent.mTerminalId, + newParentStartPos + 1 /* size of PtNodeCount */); + int written = writePtNodeArray(newParentStartPos, new Ver4PtNodeInfo[] { newParent }, + FormatSpec.NO_FORWARD_LINK_ADDRESS); + final int childrenStartPos = newParentStartPos + written; + writePtNodeArray(childrenStartPos, newChildren, FormatSpec.NO_FORWARD_LINK_ADDRESS); + int childrenNodePos = childrenStartPos + 1 /* size of PtNodeCount */; + for (final Ver4PtNodeInfo info : newChildren) { + updateTerminalPosition(info.mTerminalId, childrenNodePos); + childrenNodePos += computePtNodeSize(info.mCharacters, info.mStartIndexOfCharacters, + info.mEndIndexOfCharacters, + (info.mFlags & FormatSpec.FLAG_IS_TERMINAL) != 0); + } + + // Mark as moved. + markPtNodeAsMoved(splittedPtNodeToMovePos, newParentStartPos + 1 /* size of PtNodeCount */, + formatOptions); + updateForwardLink(ptNodeArrayPos, newParentStartPos, formatOptions); + } + + /** + * Writes a node array to the stream. + * + * @param nodeArrayPos the position of the head of the node array. + * @param infos an array of Ver4PtNodeInfo to be written. + * @return the written length in bytes. + */ + private int writePtNodeArray(final int nodeArrayPos, final Ver4PtNodeInfo[] infos, + final int forwardLink) throws IOException { + int written = BinaryDictIOUtils.writePtNodeCount(mDictStream, infos.length); + for (int i = 0; i < infos.length; ++i) { + written += writePtNode(nodeArrayPos + written, infos[i]); + } + BinaryDictIOUtils.writeSInt24ToStream(mDictStream, forwardLink); + written += FormatSpec.FORWARD_LINK_ADDRESS_SIZE; + return written; + } + + private int computePtNodeSize(final int[] codePoints, final int startIndex, final int endIndex, + final boolean isTerminal) { + return FormatSpec.PTNODE_FLAGS_SIZE + FormatSpec.PARENT_ADDRESS_SIZE + + CharEncoding.getCharArraySize(codePoints, startIndex, endIndex) + + (endIndex - startIndex > 1 ? FormatSpec.PTNODE_TERMINATOR_SIZE : 0) + + (isTerminal ? FormatSpec.PTNODE_TERMINAL_ID_SIZE : 0) + + FormatSpec.SIGNED_CHILDREN_ADDRESS_SIZE; + } + + private void writeNewSinglePtNodeWithAttributes(final int[] codePoints, + final boolean hasShortcuts, final int terminalId, final boolean hasBigrams, + final boolean isNotAWord, final boolean isBlackListEntry, final int parentPos, + final FormatOptions formatOptions) throws IOException { + final int newNodeArrayPos = mDictBuffer.limit(); + final int newNodeFlags = BinaryDictEncoderUtils.makePtNodeFlags(codePoints.length > 1, + terminalId != PtNode.NOT_A_TERMINAL, FormatSpec.FLAG_IS_NOT_MOVED, hasShortcuts, + hasBigrams, isNotAWord, isBlackListEntry, formatOptions); + final Ver4PtNodeInfo info = new Ver4PtNodeInfo(newNodeFlags, codePoints, terminalId, + FormatSpec.NO_CHILDREN_ADDRESS, parentPos, 0 /* nodeSize */); + writePtNodeArray(newNodeArrayPos, new Ver4PtNodeInfo[] { info }, + FormatSpec.NO_FORWARD_LINK_ADDRESS); + } + + private int setMultipleCharsInFlags(final int currentFlags, final boolean hasMultipleChars) { + final int flags; + if (hasMultipleChars) { + flags = currentFlags | FormatSpec.FLAG_HAS_MULTIPLE_CHARS; + } else { + flags = currentFlags & (~FormatSpec.FLAG_HAS_MULTIPLE_CHARS); + } + return flags; + } + + private int setIsNotAWordInFlags(final int currentFlags, final boolean isNotAWord) { + final int flags; + if (isNotAWord) { + flags = currentFlags | FormatSpec.FLAG_IS_NOT_A_WORD; + } else { + flags = currentFlags & (~FormatSpec.FLAG_IS_NOT_A_WORD); + } + return flags; + } + + private int setIsBlackListEntryInFlags(final int currentFlags, final boolean isBlackListEntry) { + final int flags; + if (isBlackListEntry) { + flags = currentFlags | FormatSpec.FLAG_IS_BLACKLISTED; + } else { + flags = currentFlags & (~FormatSpec.FLAG_IS_BLACKLISTED); + } + return flags; + } + + /** + * Splits a PtNode. + * + * abcd - ef + * + * -> inserting "abc" + * + * abc - d - ef + * + * @param nodeArrayToSplitPos the position of PtNodeArray which contains the PtNode to split. + * @param nodeToSplitPos the position of the PtNode to split. + * @param nodeToSplitInfo the information of the PtNode to split. + * @param indexToSplit the index where to split in the code points array. + * @param parentOfNodeToSplitPos the absolute position of a parent of the node to split. + * @param newTerminalId the terminal id of the inserted node (corresponds to "d"). + * @param hasShortcuts whether the inserted word should have shortcuts. + * @param hasBigrams whether the inserted word should have bigrams. + * @param isNotAWord whether the inserted word should be not a word. + * @param isBlackListEntry whether the inserted word should be a black list entry. + * @param formatOptions the format options. + */ + private void splitOnly(final int nodeArrayToSplitPos, final int nodeToSplitPos, + final Ver4PtNodeInfo nodeToSplitInfo, final int indexToSplit, + final int parentOfNodeToSplitPos, final int newTerminalId, final boolean hasShortcuts, + final boolean hasBigrams, final boolean isNotAWord, final boolean isBlackListEntry, + final FormatOptions formatOptions) throws IOException { + final int parentNodeArrayStartPos = mDictBuffer.limit(); + final int parentNodeStartPos = parentNodeArrayStartPos + 1 /* size of PtNodeCount */; + final int parentFlags = BinaryDictEncoderUtils.makePtNodeFlags(indexToSplit > 1, + true /* isTerminal */, FormatSpec.FLAG_IS_NOT_MOVED, hasShortcuts, hasBigrams, + isNotAWord, isBlackListEntry, formatOptions); + final Ver4PtNodeInfo parentInfo = new Ver4PtNodeInfo(parentFlags, + nodeToSplitInfo.mCharacters, newTerminalId, parentNodeStartPos + + computePtNodeSize(nodeToSplitInfo.mCharacters, 0, indexToSplit, true) + + FormatSpec.FORWARD_LINK_ADDRESS_SIZE, + parentOfNodeToSplitPos, 0 /* nodeSize */); + parentInfo.mStartIndexOfCharacters = 0; + parentInfo.mEndIndexOfCharacters = indexToSplit; + + // Write the child. + final int childrenFlags = setMultipleCharsInFlags(nodeToSplitInfo.mFlags, + nodeToSplitInfo.mCharacters.length - indexToSplit > 1); + final Ver4PtNodeInfo childrenInfo = new Ver4PtNodeInfo(childrenFlags, + nodeToSplitInfo.mCharacters, nodeToSplitInfo.mTerminalId, + nodeToSplitInfo.mChildrenPos, parentNodeStartPos, 0 /* nodeSize */); + childrenInfo.mStartIndexOfCharacters = indexToSplit; + childrenInfo.mEndIndexOfCharacters = nodeToSplitInfo.mCharacters.length; + if (nodeToSplitInfo.mChildrenPos != FormatSpec.NO_CHILDREN_ADDRESS) { + updateParentPositions(nodeToSplitInfo.mChildrenPos, + parentInfo.mChildrenPos + 1 /* size of PtNodeCount */, formatOptions); + } + + writeSplittedPtNodes(nodeArrayToSplitPos, nodeToSplitPos, parentInfo, + new Ver4PtNodeInfo[] { childrenInfo }, parentNodeArrayStartPos, formatOptions); + } + + /** + * Split and branch a PtNode. + * + * ab - cd + * + * -> inserting "ac" + * + * a - b - cd + * | + * - c + * + * @param nodeArrayToSplitPos the position of PtNodeArray which contains the PtNode to split. + * @param nodeToSplitPos the position of the PtNode to split. + * @param nodeToSplitInfo the information of the PtNode to split. + * @param indexToSplit the index where to split in the code points array. + * @param parentOfNodeToSplitPos the absolute position of parent of the node to split. + * @param newWordSuffixCodePoints the suffix of the newly inserted word (corresponds to "c"). + * @param startIndexOfNewWordSuffixCodePoints the start index in newWordSuffixCodePoints where + * the suffix starts. + * @param newTerminalId the terminal id of the inserted node (correspond to "c"). + * @param hasShortcuts whether the inserted word should have shortcuts. + * @param hasBigrams whether the inserted word should have bigrams. + * @param isNotAWord whether the inserted word should be not a word. + * @param isBlackListEntry whether the inserted word should be a black list entry. + * @param formatOptions the format options. + */ + private void splitAndBranch(final int nodeArrayToSplitPos, final int nodeToSplitPos, + final Ver4PtNodeInfo nodeToSplitInfo, final int indexToSplit, + final int parentOfNodeToSplitPos, final int[] newWordSuffixCodePoints, + final int startIndexOfNewWordSuffixCodePoints, + final int newTerminalId, + final boolean hasShortcuts, final boolean hasBigrams, final boolean isNotAWord, + final boolean isBlackListEntry, final FormatOptions formatOptions) throws IOException { + final int parentNodeArrayStartPos = mDictBuffer.limit(); + final int parentNodeStartPos = parentNodeArrayStartPos + 1 /* size of PtNodeCount */; + final int parentFlags = BinaryDictEncoderUtils.makePtNodeFlags( + indexToSplit > 1, + false /* isTerminal */, FormatSpec.FLAG_IS_NOT_MOVED, + false /* hasShortcut */, false /* hasBigrams */, + false /* isNotAWord */, false /* isBlackListEntry */, formatOptions); + final Ver4PtNodeInfo parentInfo = new Ver4PtNodeInfo(parentFlags, + nodeToSplitInfo.mCharacters, PtNode.NOT_A_TERMINAL, + parentNodeStartPos + + computePtNodeSize(nodeToSplitInfo.mCharacters, 0, indexToSplit, false) + + FormatSpec.FORWARD_LINK_ADDRESS_SIZE, + parentOfNodeToSplitPos, 0 /* nodeSize */); + parentInfo.mStartIndexOfCharacters = 0; + parentInfo.mEndIndexOfCharacters = indexToSplit; + + final int childrenNodeArrayStartPos = parentNodeStartPos + + computePtNodeSize(nodeToSplitInfo.mCharacters, 0, indexToSplit, false) + + FormatSpec.FORWARD_LINK_ADDRESS_SIZE; + final int firstChildrenFlags = BinaryDictEncoderUtils.makePtNodeFlags( + newWordSuffixCodePoints.length - startIndexOfNewWordSuffixCodePoints > 1, + true /* isTerminal */, FormatSpec.FLAG_IS_NOT_MOVED, hasShortcuts, hasBigrams, + isNotAWord, isBlackListEntry, formatOptions); + final Ver4PtNodeInfo firstChildrenInfo = new Ver4PtNodeInfo(firstChildrenFlags, + newWordSuffixCodePoints, newTerminalId, + FormatSpec.NO_CHILDREN_ADDRESS, parentNodeStartPos, + 0 /* nodeSize */); + firstChildrenInfo.mStartIndexOfCharacters = startIndexOfNewWordSuffixCodePoints; + firstChildrenInfo.mEndIndexOfCharacters = newWordSuffixCodePoints.length; + + final int secondChildrenStartPos = childrenNodeArrayStartPos + 1 /* size of ptNodeCount */ + + computePtNodeSize(newWordSuffixCodePoints, startIndexOfNewWordSuffixCodePoints, + newWordSuffixCodePoints.length, true /* isTerminal */); + final int secondChildrenFlags = setMultipleCharsInFlags(nodeToSplitInfo.mFlags, + nodeToSplitInfo.mCharacters.length - indexToSplit > 1); + final Ver4PtNodeInfo secondChildrenInfo = new Ver4PtNodeInfo(secondChildrenFlags, + nodeToSplitInfo.mCharacters, nodeToSplitInfo.mTerminalId, + nodeToSplitInfo.mChildrenPos, parentNodeStartPos, 0 /* nodeSize */); + secondChildrenInfo.mStartIndexOfCharacters = indexToSplit; + secondChildrenInfo.mEndIndexOfCharacters = nodeToSplitInfo.mCharacters.length; + if (nodeToSplitInfo.mChildrenPos != FormatSpec.NO_CHILDREN_ADDRESS) { + updateParentPositions(nodeToSplitInfo.mChildrenPos, secondChildrenStartPos, + formatOptions); + } + + writeSplittedPtNodes(nodeArrayToSplitPos, nodeToSplitPos, parentInfo, + new Ver4PtNodeInfo[] { firstChildrenInfo, secondChildrenInfo }, + parentNodeArrayStartPos, formatOptions); + } + + /** + * Inserts a word into the trie file and returns the position of inserted terminal node. + * If the insertion is failed, returns FormatSpec.NOT_VALID_WORD. + */ + @UsedForTesting + private int insertWordToTrie(final String word, final int newTerminalId, + final boolean isNotAWord, final boolean isBlackListEntry, final boolean hasBigrams, + final boolean hasShortcuts) throws IOException, UnsupportedFormatException { + setPosition(0); + final FileHeader header = readHeader(); + + final int[] codePoints = FusionDictionary.getCodePoints(word); + final int wordLen = codePoints.length; + + int wordPos = 0; + for (int depth = 0; depth < FormatSpec.MAX_WORD_LENGTH; /* nop */) { + final int nodeArrayPos = getPosition(); + final int ptNodeCount = readPtNodeCount(); + boolean goToChildren = false; + int parentPos = FormatSpec.NO_PARENT_ADDRESS; + for (int i = 0; i < ptNodeCount; ++i) { + final int nodePos = getPosition(); + final Ver4PtNodeInfo nodeInfo = readVer4PtNodeInfo(nodePos, header.mFormatOptions); + if (BinaryDictIOUtils.isMovedPtNode(nodeInfo.mFlags, header.mFormatOptions)) { + continue; + } + if (nodeInfo.mParentPos != FormatSpec.NO_PARENT_ADDRESS) { + parentPos = nodePos + nodeInfo.mParentPos; + } + + final boolean firstCharacterMatched = + codePoints[wordPos] == nodeInfo.mCharacters[0]; + boolean allCharactersMatched = true; + int firstDifferentCharacterIndex = -1; + for (int p = 0; p < nodeInfo.mCharacters.length; ++p) { + if (wordPos + p >= codePoints.length) break; + if (codePoints[wordPos + p] != nodeInfo.mCharacters[p]) { + if (firstDifferentCharacterIndex == -1) { + firstDifferentCharacterIndex = p; + } + allCharactersMatched = false; + } + } + + if (!firstCharacterMatched) { + // Go to the next sibling node. + continue; + } + + if (!allCharactersMatched) { + final int parentNodeArrayStartPos = mDictBuffer.limit(); + splitAndBranch(nodeArrayPos, nodePos, nodeInfo, firstDifferentCharacterIndex, + parentPos, codePoints, wordPos + firstDifferentCharacterIndex, + newTerminalId, hasShortcuts, hasBigrams, isNotAWord, + isBlackListEntry, header.mFormatOptions); + + return parentNodeArrayStartPos + computePtNodeSize(codePoints, wordPos, + wordPos + firstDifferentCharacterIndex, false) + + FormatSpec.FORWARD_LINK_ADDRESS_SIZE + 1 /* size of PtNodeCount */; + } + + if (wordLen - wordPos < nodeInfo.mCharacters.length) { + final int parentNodeArrayStartPos = mDictBuffer.limit(); + splitOnly(nodeArrayPos, nodePos, nodeInfo, wordLen - wordPos, parentPos, + newTerminalId, hasShortcuts, hasBigrams, isNotAWord, isBlackListEntry, + header.mFormatOptions); + + // Return the position of the inserted word. + return parentNodeArrayStartPos + 1 /* size of PtNodeCount */; + } + + wordPos += nodeInfo.mCharacters.length; + if (wordPos == wordLen) { + // This dictionary already contains the word. + Log.e(TAG, "Something went wrong. If the word is already contained, " + + " there is no need to insert new PtNode."); + return FormatSpec.NOT_VALID_WORD; + } + if (nodeInfo.mChildrenPos == FormatSpec.NO_CHILDREN_ADDRESS) { + // There are no children. + // We need to add a new node as a child of this node. + final int newNodeArrayPos = mDictBuffer.limit(); + final int[] newNodeCodePoints = Arrays.copyOfRange(codePoints, wordPos, + codePoints.length); + writeNewSinglePtNodeWithAttributes(newNodeCodePoints, hasShortcuts, + newTerminalId, hasBigrams, isNotAWord, isBlackListEntry, nodePos, + header.mFormatOptions); + updateChildrenPos(nodePos, newNodeArrayPos, header.mFormatOptions); + return newNodeArrayPos + 1 /* size of PtNodeCount */; + } else { + // Found the matched node. + // Go to the children of this node. + setPosition(nodeInfo.mChildrenPos); + goToChildren = true; + depth++; + break; + } + } + + if (goToChildren) continue; + if (!readAndFollowForwardLink()) { + // Add a new node that contains [wordPos, word.length()-1]. + // and update the forward link. + final int newNodeArrayPos = mDictBuffer.limit(); + final int[] newCodePoints = Arrays.copyOfRange(codePoints, wordPos, + codePoints.length); + writeNewSinglePtNodeWithAttributes(newCodePoints, hasShortcuts, newTerminalId, + hasBigrams, isNotAWord, isBlackListEntry, parentPos, header.mFormatOptions); + updateForwardLink(nodeArrayPos, newNodeArrayPos, header.mFormatOptions); + return newNodeArrayPos + 1 /* size of PtNodeCount */; + } + } + return FormatSpec.NOT_VALID_WORD; + } + + private void updateFrequency(final int terminalId, final int frequency) { + mFrequencyBuffer.position(terminalId * FormatSpec.FREQUENCY_AND_FLAGS_SIZE); + BinaryDictEncoderUtils.writeUIntToDictBuffer(mFrequencyBuffer, frequency, + FormatSpec.FREQUENCY_AND_FLAGS_SIZE); + } + + private void insertFrequency(final int frequency) throws IOException { + final OutputStream frequencyStream = new FileOutputStream(mFrequencyFile, + true /* append */); + BinaryDictEncoderUtils.writeUIntToStream(frequencyStream, frequency, + FormatSpec.FREQUENCY_AND_FLAGS_SIZE); + frequencyStream.close(); + } + + private void insertTerminalPosition(final int posOfTerminal) throws IOException { + final OutputStream terminalPosStream = new FileOutputStream( + getFile(FILETYPE_TERMINAL_ADDRESS_TABLE), true /* append */); + BinaryDictEncoderUtils.writeUIntToStream(terminalPosStream, posOfTerminal, + FormatSpec.TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE); + terminalPosStream.close(); + } + + private void insertBigrams(final int terminalId, final int frequency, + final ArrayList<PendingAttribute> bigramAddresses) + throws IOException, UnsupportedFormatException { + openDictBuffer(); + final BigramContentUpdater updater = new BigramContentUpdater(mDictDirectory.getName(), + mDictDirectory, false); + + // Convert addresses to terminal ids. + final ArrayList<PendingAttribute> bigrams = CollectionUtils.newArrayList(); + mDictBuffer.position(0); + final FileHeader header = readHeader(); + for (PendingAttribute attr : bigramAddresses) { + mDictBuffer.position(attr.mAddress); + final Ver4PtNodeInfo info = readVer4PtNodeInfo(attr.mAddress, header.mFormatOptions); + if (info.mTerminalId == PtNode.NOT_A_TERMINAL) { + throw new RuntimeException("We can't have a bigram target that's not a terminal."); + } + bigrams.add(new PendingAttribute(frequency, info.mTerminalId)); + } + updater.insertBigramEntries(terminalId, frequency, bigrams); + close(); + } + + private void insertShortcuts(final int terminalId, final ArrayList<WeightedString> shortcuts) + throws IOException { + final ShortcutContentUpdater updater = new ShortcutContentUpdater(mDictDirectory.getName(), + mDictDirectory); + updater.insertShortcuts(terminalId, shortcuts); + } + + private void openBuffersAndStream() throws IOException { + openDictBuffer(); + mDictStream = new FileOutputStream(getFile(FILETYPE_TRIE), true /* append */); + } + + private void close() throws IOException { + if (mDictStream != null) { + mDictStream.close(); + mDictStream = null; + } + mDictBuffer = null; + mFrequencyBuffer = null; + mTerminalAddressTableBuffer = null; + } + + private void updateAttributes(final int posOfWord, final int frequency, + final ArrayList<WeightedString> bigramStrings, + final ArrayList<WeightedString> shortcuts, final boolean isNotAWord, + final boolean isBlackListEntry) throws IOException, UnsupportedFormatException { + mDictBuffer.position(0); + final FileHeader header = readHeader(); + mDictBuffer.position(posOfWord); + final Ver4PtNodeInfo info = readVer4PtNodeInfo(posOfWord, header.mFormatOptions); + final int terminalId = info.mTerminalId; + + // Update the flags. + final int newFlags = setIsNotAWordInFlags( + setIsBlackListEntryInFlags(info.mFlags, isBlackListEntry), isNotAWord); + mDictBuffer.position(posOfWord); + mDictBuffer.put((byte) newFlags); + + updateFrequency(terminalId, frequency); + insertBigrams(terminalId, frequency, + DynamicBinaryDictIOUtils.resolveBigramPositions(this, bigramStrings)); + insertShortcuts(terminalId, shortcuts); + } + + @Override @UsedForTesting public void insertWord(final String word, final int frequency, final ArrayList<WeightedString> bigramStrings, final ArrayList<WeightedString> shortcuts, final boolean isNotAWord, final boolean isBlackListEntry) throws IOException, UnsupportedFormatException { - // TODO: Implement this method. + final int newTerminalId = getNewTerminalId(); + + openBuffersAndStream(); + final int posOfWord = getTerminalPosition(word); + if (posOfWord != FormatSpec.NOT_VALID_WORD) { + // The word is already contained in the dictionary. + updateAttributes(posOfWord, frequency, bigramStrings, shortcuts, isNotAWord, + isBlackListEntry); + close(); + return; + } + + // Insert new PtNode into trie. + final int posOfTerminal = insertWordToTrie(word, newTerminalId, isNotAWord, + isBlackListEntry, bigramStrings != null && !bigramStrings.isEmpty(), + shortcuts != null && !shortcuts.isEmpty()); + insertFrequency(frequency); + insertTerminalPosition(posOfTerminal); + close(); + + insertBigrams(newTerminalId, frequency, + DynamicBinaryDictIOUtils.resolveBigramPositions(this, bigramStrings)); + insertShortcuts(newTerminalId, shortcuts); } } diff --git a/java/src/com/android/inputmethod/latin/utils/SubtypeLocaleUtils.java b/java/src/com/android/inputmethod/latin/utils/SubtypeLocaleUtils.java index 102a41b4e..fdbe81ab6 100644 --- a/java/src/com/android/inputmethod/latin/utils/SubtypeLocaleUtils.java +++ b/java/src/com/android/inputmethod/latin/utils/SubtypeLocaleUtils.java @@ -197,7 +197,9 @@ public final class SubtypeLocaleUtils { // es_US spanish F Español (EE.UU.) exception // fr azerty F Français // fr_CA qwerty F Français (Canada) + // fr_CH swiss F Français (Suisse) // de qwertz F Deutsch + // de_CH swiss T Deutsch (Schweiz) // zz qwerty F No language (QWERTY) in system locale // fr qwertz T Français (QWERTZ) // de qwerty T Deutsch (QWERTY) @@ -298,7 +300,9 @@ public final class SubtypeLocaleUtils { // es_US spanish F Es Español Español (EE.UU.) exception // fr azerty F Fr Français Français // fr_CA qwerty F Fr Français Français (Canada) + // fr_CH swiss F Fr Français Français (Suisse) // de qwertz F De Deutsch Deutsch + // de_CH swiss T De Deutsch Deutsch (Schweiz) // zz qwerty F QWERTY QWERTY // fr qwertz T Fr Français Français // de qwerty T De Deutsch Deutsch diff --git a/java/src/com/android/inputmethod/research/ResearchLogger.java b/java/src/com/android/inputmethod/research/ResearchLogger.java index da9c61103..ad509923a 100644 --- a/java/src/com/android/inputmethod/research/ResearchLogger.java +++ b/java/src/com/android/inputmethod/research/ResearchLogger.java @@ -102,10 +102,6 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang && ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS_DEBUG; private static final boolean DEBUG_REPLAY_AFTER_FEEDBACK = false && ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS_DEBUG; - // Whether the TextView contents are logged at the end of the session. true will disclose - // private info. - private static final boolean LOG_FULL_TEXTVIEW_CONTENTS = false - && ProductionFlag.USES_DEVELOPMENT_ONLY_DIAGNOSTICS_DEBUG; // Whether the feedback dialog preserves the editable text across invocations. Should be false // for normal research builds so users do not have to delete the same feedback string they // entered earlier. Should be true for builds internal to a development team so when the text @@ -168,9 +164,6 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang // U+E001 is in the "private-use area" /* package for test */ static final String WORD_REPLACEMENT_STRING = "\uE001"; protected static final int SUSPEND_DURATION_IN_MINUTES = 1; - // set when LatinIME should ignore an onUpdateSelection() callback that - // arises from operations in this class - private static boolean sLatinIMEExpectingUpdateSelection = false; // used to check whether words are not unique private Suggest mSuggest; @@ -1126,12 +1119,6 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang new Object[] { applicationSpecifiedCompletions }); } - public static boolean getAndClearLatinIMEExpectingUpdateSelection() { - boolean returnValue = sLatinIMEExpectingUpdateSelection; - sLatinIMEExpectingUpdateSelection = false; - return returnValue; - } - /** * The IME is finishing; it is either being destroyed, or is about to be hidden. * @@ -1149,51 +1136,12 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang // if called from finishViews(), which is called from hideWindow() and onDestroy(). These // are the situations in which we want to finish up the researchLog. if (ic != null && !finishingInput) { - final boolean isTextTruncated; - final String text; - if (LOG_FULL_TEXTVIEW_CONTENTS) { - // Capture the TextView contents. This will trigger onUpdateSelection(), so we - // set sLatinIMEExpectingUpdateSelection so that when onUpdateSelection() is called, - // it can tell that it was generated by the logging code, and not by the user, and - // therefore keep user-visible state as is. - ic.beginBatchEdit(); - ic.performContextMenuAction(android.R.id.selectAll); - CharSequence charSequence = ic.getSelectedText(0); - if (savedSelectionStart != -1 && savedSelectionEnd != -1) { - ic.setSelection(savedSelectionStart, savedSelectionEnd); - } - ic.endBatchEdit(); - sLatinIMEExpectingUpdateSelection = true; - if (TextUtils.isEmpty(charSequence)) { - isTextTruncated = false; - text = ""; - } else { - if (charSequence.length() > MAX_INPUTVIEW_LENGTH_TO_CAPTURE) { - int length = MAX_INPUTVIEW_LENGTH_TO_CAPTURE; - // do not cut in the middle of a supplementary character - final char c = charSequence.charAt(length - 1); - if (Character.isHighSurrogate(c)) { - length--; - } - final CharSequence truncatedCharSequence = charSequence.subSequence(0, - length); - isTextTruncated = true; - text = truncatedCharSequence.toString(); - } else { - isTextTruncated = false; - text = charSequence.toString(); - } - } - } else { - isTextTruncated = true; - text = ""; - } final ResearchLogger researchLogger = getInstance(); // Assume that OUTPUT_ENTIRE_BUFFER is only true when we don't care about privacy (e.g. // during a live user test), so the normal isPotentiallyPrivate and // isPotentiallyRevealing flags do not apply researchLogger.enqueueEvent(LOGSTATEMENT_LATINIME_ONFINISHINPUTVIEWINTERNAL, - isTextTruncated, text); + true /* isTextTruncated */, "" /* text */); researchLogger.commitCurrentLogUnit(); getInstance().stop(); } @@ -1213,9 +1161,7 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang public static void latinIME_onUpdateSelection(final int lastSelectionStart, final int lastSelectionEnd, final int oldSelStart, final int oldSelEnd, final int newSelStart, final int newSelEnd, final int composingSpanStart, - final int composingSpanEnd, final boolean expectingUpdateSelection, - final boolean expectingUpdateSelectionFromLogger, - final RichInputConnection connection) { + final int composingSpanEnd, final RichInputConnection connection) { String word = ""; if (connection != null) { TextRange range = connection.getWordRangeAtCursor(WHITESPACE_SEPARATORS, 1); @@ -1227,8 +1173,8 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang final String scrubbedWord = researchLogger.scrubWord(word); researchLogger.enqueueEvent(LOGSTATEMENT_LATINIME_ONUPDATESELECTION, lastSelectionStart, lastSelectionEnd, oldSelStart, oldSelEnd, newSelStart, newSelEnd, - composingSpanStart, composingSpanEnd, expectingUpdateSelection, - expectingUpdateSelectionFromLogger, scrubbedWord); + composingSpanStart, composingSpanEnd, false /* expectingUpdateSelection */, + false /* expectingUpdateSelectionFromLogger */, scrubbedWord); } /** |