Changeset 179567 in webkit

Timestamp:

Feb 3, 2015 1:37:56 PM (9 years ago)

Author:

enrica@apple.com

Message:

Additional emoji support.
​https://bugs.webkit.org/show_bug.cgi?id=141047
rdar://problem/19045135

Reviewed by Darin Adler.

Source/WebCore:

Adds support for emoji modifiers and group emoji.

Test: editing/deleting/delete-emoji.html

• platform/graphics/FontCascade.cpp:

(WebCore::FontCascade::characterRangeCodePath):

• platform/text/TextBreakIterator.cpp:

(WebCore::cursorMovementIterator):

• rendering/RenderText.cpp:

(WebCore::isEmojiGroupCandidate):
(WebCore::isEmojiModifier):
(WebCore::RenderText::previousOffsetForBackwardDeletion):

LayoutTests:

• editing/deleting/delete-emoji.html: Added.
• editing/deleting/delete-emoji-expected.txt: Added.

Location:

trunk

Files:

• LayoutTests/ChangeLog (modified) (1 diff)
• LayoutTests/editing/deleting/delete-emoji-expected.txt (added)
• LayoutTests/editing/deleting/delete-emoji.html (added)
• Source/WebCore/ChangeLog (modified) (1 diff)
• Source/WebCore/platform/graphics/FontCascade.cpp (modified) (2 diffs)
• Source/WebCore/platform/text/TextBreakIterator.cpp (modified) (8 diffs)
• Source/WebCore/rendering/RenderText.cpp (modified) (4 diffs)

trunk/LayoutTests/ChangeLog

@@ +1 @@
+2015-02-02  Enrica Casucci  <enrica@apple.com>
+
+        Additional emoji support.
+        https://bugs.webkit.org/show_bug.cgi?id=141047
+        rdar://problem/19045135
+
+        Reviewed by Darin Adler.
+
+        * editing/deleting/delete-emoji.html: Added.
+        * editing/deleting/delete-emoji-expected.txt: Added.
+
 2015-02-03  Brent Fulgham  <bfulgham@apple.com>
 

trunk/Source/WebCore/ChangeLog

@@ +1 @@
+2015-02-02  Enrica Casucci  <enrica@apple.com>
+
+        Additional emoji support.
+        https://bugs.webkit.org/show_bug.cgi?id=141047
+        rdar://problem/19045135
+
+        Reviewed by Darin Adler.
+
+        Adds support for emoji modifiers and group emoji.
+
+        Test: editing/deleting/delete-emoji.html
+
+        * platform/graphics/FontCascade.cpp:
+        (WebCore::FontCascade::characterRangeCodePath):
+        * platform/text/TextBreakIterator.cpp:
+        (WebCore::cursorMovementIterator):
+        * rendering/RenderText.cpp:
+        (WebCore::isEmojiGroupCandidate):
+        (WebCore::isEmojiModifier):
+        (WebCore::RenderText::previousOffsetForBackwardDeletion):
+
 2015-02-03  Jer Noble  <jer.noble@apple.com>
 

trunk/Source/WebCore/platform/graphics/FontCascade.cpp

@@ -641 +641 @@
     // list of ranges.
     CodePath result = Simple;
+    bool previousCharacterIsEmojiGroupCandidate = false;
     for (unsigned i = 0; i < len; i++) {
         const UChar c = characters[i];
+        if (c == zeroWidthJoiner && previousCharacterIsEmojiGroupCandidate)
+            return Complex;
+        
+        previousCharacterIsEmojiGroupCandidate = false;
         if (c < 0x2E5) // U+02E5 through U+02E9 (Modifier Letters : Tone letters)  
             continue;
@@ -763 +768 @@
                 return Complex;
 
+            if (supplementaryCharacter >= 0x1F466 && supplementaryCharacter <= 0x1F469) {
+                previousCharacterIsEmojiGroupCandidate = true;
+                continue;
+            }
             if (supplementaryCharacter < 0xE0100) // U+E0100 through U+E01EF Unicode variation selectors.
                 continue;

trunk/Source/WebCore/platform/text/TextBreakIterator.cpp

@@ -208 +208 @@
         "$Mal1    = [\\u0D15-\\u0D39];"    // Malayalam Letter A,...,Ha
         "$RI      = [\\U0001F1E6-\\U0001F1FF];" // Emoji regional indicators
+        "$ZWJ     = \\u200D;"               // Zero width joiner
+        "$EmojiForModsAndSeqs = [\\U0001F466-\\U0001F469];" // Emoji that take Fitzpatrick modifiers AND participate in ZWJ sequences
+        "$EmojiForModsOnly = [\\u261D \\u270A-\\u270C \\U0001F385 \\U0001F3C3-\\U0001F3C4 \\U0001F3C7 \\U0001F3CA \\U0001F442-\\U0001F443 \\U0001F446-\\U0001F450 \\U0001F46E-\\U0001F478 \\U0001F47C \\U0001F481-\\U0001F483 \\U0001F485-\\U0001F487 \\U0001F4AA \\U0001F645-\\U0001F647 \\U0001F64B-\\U0001F64F \\U0001F6B4-\\U0001F6B6 \\U0001F6C0];" // Emoji that take Fitzpatrick modifiers
+        "$EmojiMods = [\\U0001F3FB-\\U0001F3FF];" // Fitzpatrick modifiers
         "!!chain;"
         "!!forward;"
@@ -226 +230 @@
         "$Kan0 $KanV $Kan1;"               // Kannada Virama (forward)
         "$Mal0 $MalV $Mal1;"               // Malayalam Virama (forward)
+        "$ZWJ $EmojiForModsAndSeqs;"       // Don't break in emoji ZWJ sequences
+        "[$EmojiForModsAndSeqs $EmojiForModsOnly] $EmojiMods;" // Don't break between relevant emoji and Fitzpatrick modifier
         "!!reverse;"
         "$LF $CR;"
@@ -243 +249 @@
         "$Kan1 $KanV $Kan0;"               // Kannada Virama (backward)
         "$Mal1 $MalV $Mal0;"               // Malayalam Virama (backward)
+        "$EmojiForModsAndSeqs $ZWJ;"       // Don't break in emoji ZWJ sequences
+        "$EmojiMods [$EmojiForModsAndSeqs $EmojiForModsOnly];" // Don't break between relevant emoji and Fitzpatrick modifier
         "!!safe_reverse;"
         "!!safe_forward;";
@@ -398 +406 @@
     "$XX = [:LineBreak = Unknown:];"
     "$ZW = [:LineBreak = ZWSpace:];"
+    "$ZWJ = \\u200D;"
+    "$EmojiForModsAndSeqs = [\\U0001F466-\\U0001F469];"
+    "$EmojiForModsOnly = [\\u261D \\u270A-\\u270C \\U0001F385 \\U0001F3C3-\\U0001F3C4 \\U0001F3C7 \\U0001F3CA \\U0001F442-\\U0001F443 \\U0001F446-\\U0001F450 \\U0001F46E-\\U0001F478 \\U0001F47C \\U0001F481-\\U0001F483 \\U0001F485-\\U0001F487 \\U0001F4AA \\U0001F645-\\U0001F647 \\U0001F64B-\\U0001F64F \\U0001F6B4-\\U0001F6B6 \\U0001F6C0];"
+    "$EmojiMods = [\\U0001F3FB-\\U0001F3FF];"
     "$dictionary = [:LineBreak = Complex_Context:];"
     "$ALPlus = [$AL $AI $SA $SG $XX];"
@@ -474 +486 @@
     "$CAN_CM $CM* [$SP $ZW];"
     "$CM+ [$SP $ZW];"
+    "[$EmojiForModsAndSeqs $EmojiMods] $ZWJ $EmojiForModsAndSeqs;"
     "$CAN_CM $CM+;"
     "$CM+;"
@@ -540 +553 @@
     "($ALcm | $HLcm | $NUcm) $OPcm;"
     "$CM+ $OPcm;"
-    "$CPcm ($ALcm | $HLcm | $NUcm);";
+    "$CPcm ($ALcm | $HLcm | $NUcm);"
+    "[$EmojiForModsAndSeqs $EmojiForModsOnly] $EmojiMods;";
 
 static const char* uax14Reverse =
@@ -578 +592 @@
     "[$SP $ZW] [$LB4NonBreaks-$CM];"
     "[$SP $ZW] $CM+ $CAN_CM;"
+    "$EmojiForModsAndSeqs $ZWJ [$EmojiForModsAndSeqs $EmojiMods];"
     "$CM+ $CAN_CM;"
     "$CM* $WJ $CM* $CAN_CM;"
@@ -634 +649 @@
     "$CM* ($ALPlus | $HL) $CM* $IS;"
     "$CM* $OP $CM* ($ALPlus | $HL | $NU);"
-    "$CM* ($ALPlus | $HL | $NU) $CM* $CP;";
+    "$CM* ($ALPlus | $HL | $NU) $CM* $CP;"
+    "$EmojiMods [$EmojiForModsAndSeqs $EmojiForModsOnly];";
 
 static const char* uax14SafeForward =

trunk/Source/WebCore/rendering/RenderText.cpp

@@ -1422 +1422 @@
 };
 
-inline bool isHangulLVT(UChar32 character)
+static inline bool isHangulLVT(UChar32 character)
 {
     return (character - HANGUL_SYLLABLE_START) % HANGUL_JONGSEONG_COUNT;
 }
 
-inline bool isMark(UChar32 c)
-{
-    int8_t charType = u_charType(c);
+static inline bool isMark(UChar32 character)
+{
+    int8_t charType = u_charType(character);
     return charType == U_NON_SPACING_MARK || charType == U_ENCLOSING_MARK || charType == U_COMBINING_SPACING_MARK;
 }
 
-inline bool isRegionalIndicator(UChar32 c)
+static inline bool isRegionalIndicator(UChar32 character)
 {
     // National flag emoji each consists of a pair of regional indicator symbols.
-    return 0x1F1E6 <= c && c <= 0x1F1FF;
+    return 0x1F1E6 <= character && character <= 0x1F1FF;
+}
+
+static inline bool isEmojiGroupCandidate(UChar32 character)
+{
+    return character >= 0x1F466 && character <= 0x1F469;
+}
+
+static inline bool isEmojiModifier(UChar32 character)
+{
+    return character >= 0x1F3FB && character <= 0x1F3FF;
 }
 
@@ -1448 +1458 @@
     UChar32 character;
     bool sawRegionalIndicator = false;
+    bool sawEmojiGroupCandidate = false;
+    bool sawEmojiModifier = false;
+    
     while (current > 0) {
         if (U16_IS_TRAIL(text[--current]))
@@ -1455 +1468 @@
 
         UChar32 character = text.characterStartingAt(current);
+
+        if (sawEmojiGroupCandidate) {
+            sawEmojiGroupCandidate = false;
+            if (character == zeroWidthJoiner)
+                continue;
+            // We could have two emoji group candidates without a joiner in between.
+            // Those should not be treated as a group.
+            U16_FWD_1_UNSAFE(text, current);
+            break;
+        }
+
+        if (sawEmojiModifier) {
+            if (isEmojiModifier(character))
+                U16_FWD_1_UNSAFE(text, current);
+            break;
+        }
 
         if (sawRegionalIndicator) {
@@ -1472 +1501 @@
         if (isRegionalIndicator(character)) {
             sawRegionalIndicator = true;
+            continue;
+        }
+        
+        if (isEmojiModifier(character)) {
+            sawEmojiModifier = true;
+            continue;
+        }
+
+        if (isEmojiGroupCandidate(character)) {
+            sawEmojiGroupCandidate = true;
             continue;
         }