Changeset 41588 in webkit

Timestamp:

Mar 11, 2009, 9:05:04 AM (16 years ago)

Author:

ap@webkit.org

Message:

2009-03-11 Hironori Bono <​hbono@chromium.org>

Reviewed by Alexey Proskuryakov.

​https://bugs.webkit.org/show_bug.cgi?id=15790
Ligatures aren't accounted for when manipulating VisiblePositions

Changed cursorMovementIterator rules to prevent a cursor from moving in the middle of an
Indic ligature which uses a virama sign.

Test: editing/deleting/skip-virama-001.html

• platform/text/TextBreakIteratorICU.cpp: (WebCore::cursorMovementIterator):

Location:

trunk

Files:

• LayoutTests/ChangeLog (modified) (1 diff)
• LayoutTests/editing/deleting/skip-virama-001-expected.txt (added)
• LayoutTests/editing/deleting/skip-virama-001.html (added)
• LayoutTests/platform/mac-tiger/Skipped (modified) (1 diff)
• WebCore/ChangeLog (modified) (1 diff)
• WebCore/platform/text/TextBreakIteratorICU.cpp (modified) (2 diffs)

trunk/LayoutTests/ChangeLog

@@ +1 @@
+2009-03-11  Hironori Bono  <hbono@chromium.org>
+
+        Reviewed by Alexey Proskuryakov.
+
+        https://bugs.webkit.org/show_bug.cgi?id=15790
+        Ligatures aren't accounted for when manipulating VisiblePositions
+
+        Verify that we can prevent a cursor from moving in the middle of an Indic ligature.
+
+        * editing/deleting/skip-virama-001-expected.txt: Added.
+        * editing/deleting/skip-virama-001.html: Added.
+
+        * platform/mac-tiger/Skipped: This fix requires a customized cursorMovementIterator,
+        which is not enabled on Tiger (see bug 24513).
+
 2009-03-11  Jan Michael Alonzo  <jmalonzo@webkit.org>
 
         Reviewed by Holger Freyther.
 
-        Gtk] Implement LayoutTestControllerGtk::setPrivateBrowsingEnabled
+        [Gtk] Implement LayoutTestControllerGtk::setPrivateBrowsingEnabled
         https://bugs.webkit.org/show_bug.cgi?id=24487
 

trunk/LayoutTests/platform/mac-tiger/Skipped

@@ -23 +23 @@
 http/tests/xmlhttprequest/upload-onprogress-event.html
 http/tests/xmlhttprequest/upload-progress-events.html
+
+# Customized cursorMovementIterator doesn't work with ICU 3.2
+editing/deleting/skip-virama-001.html

trunk/WebCore/ChangeLog

@@ +1 @@
+2009-03-11  Hironori Bono  <hbono@chromium.org>
+
+        Reviewed by Alexey Proskuryakov.
+
+        https://bugs.webkit.org/show_bug.cgi?id=15790
+        Ligatures aren't accounted for when manipulating VisiblePositions
+
+        Changed cursorMovementIterator rules to prevent a cursor from moving in the middle of an
+        Indic ligature which uses a virama sign.
+        
+        Test: editing/deleting/skip-virama-001.html
+
+        * platform/text/TextBreakIteratorICU.cpp:
+        (WebCore::cursorMovementIterator):
+
 2009-03-11  Alexey Proskuryakov  <ap@webkit.org>
 

trunk/WebCore/platform/text/TextBreakIteratorICU.cpp

@@ -149 +149 @@
     return characterBreakIterator(string, length);
 #else
-    // This rule set is based on character-break iterator rules of ICU 3.8
-    // <http://source.icu-project.org/repos/icu/icu/tags/release-3-8/source/data/brkitr/char.txt>.
+    // This rule set is based on character-break iterator rules of ICU 4.0
+    // <http://source.icu-project.org/repos/icu/icu/tags/release-4-0/source/data/brkitr/char.txt>.
+    // The major differences from the original ones are listed below:
+    // * Replaced '[\p{Grapheme_Cluster_Break = SpacingMark}]' with '[\p{General_Category = Spacing Mark} - $Extend]' for ICU 3.8 or earlier;
+    // * Removed rules that prevent a cursor from moving after prepend characters (Bug 24342);
+    // * Added rules that prevent a cursor from moving after virama signs of Indic languages except Tamil (Bug 15790), and;
+    // * Added rules that prevent a cursor from moving before Japanese half-width katakara voiced marks.
     static const char* kRules =
         "$CR      = [\\p{Grapheme_Cluster_Break = CR}];"
         "$LF      = [\\p{Grapheme_Cluster_Break = LF}];"
         "$Control = [\\p{Grapheme_Cluster_Break = Control}];"
-        "$VoiceMarks = [\\uff9e\\uff9f];"
+        "$VoiceMarks = [\\uFF9E\\uFF9F];"  // Japanese half-width katakana voiced marks
         "$Extend  = [\\p{Grapheme_Cluster_Break = Extend} $VoiceMarks];"
+        "$SpacingMark = [[\\p{General_Category = Spacing Mark}] - $Extend];"
         "$L       = [\\p{Grapheme_Cluster_Break = L}];"
         "$V       = [\\p{Grapheme_Cluster_Break = V}];"
@@ -162 +168 @@
         "$LV      = [\\p{Grapheme_Cluster_Break = LV}];"
         "$LVT     = [\\p{Grapheme_Cluster_Break = LVT}];"
-        "$HangulSyllable = $L+ | ($L* ($LV? $V+ | $LV | $LVT) $T*) | $T+;"
+        "$Hin0    = [\\u0905-\\u0939];"    // Devanagari Letter A,...,Ha
+        "$HinV    = \\u094D;"              // Devanagari Sign Virama
+        "$Hin1    = [\\u0915-\\u0939];"    // Devanagari Letter Ka,...,Ha
+        "$Ben0    = [\\u0985-\\u09B9];"    // Bengali Letter A,...,Ha
+        "$BenV    = \\u09CD;"              // Bengali Sign Virama
+        "$Ben1    = [\\u0995-\\u09B9];"    // Bengali Letter Ka,...,Ha
+        "$Pan0    = [\\u0A05-\\u0A39];"    // Gurmukhi Letter A,...,Ha
+        "$PanV    = \\u0A4D;"              // Gurmukhi Sign Virama
+        "$Pan1    = [\\u0A15-\\u0A39];"    // Gurmukhi Letter Ka,...,Ha
+        "$Guj0    = [\\u0A85-\\u0AB9];"    // Gujarati Letter A,...,Ha
+        "$GujV    = \\u0ACD;"              // Gujarati Sign Virama
+        "$Guj1    = [\\u0A95-\\u0AB9];"    // Gujarati Letter Ka,...,Ha
+        "$Ori0    = [\\u0B05-\\u0B39];"    // Oriya Letter A,...,Ha
+        "$OriV    = \\u0B4D;"              // Oriya Sign Virama
+        "$Ori1    = [\\u0B15-\\u0B39];"    // Oriya Letter Ka,...,Ha
+        "$Tel0    = [\\u0C05-\\u0C39];"    // Telugu Letter A,...,Ha
+        "$TelV    = \\u0C4D;"              // Telugu Sign Virama
+        "$Tel1    = [\\u0C14-\\u0C39];"    // Telugu Letter Ka,...,Ha
+        "$Kan0    = [\\u0C85-\\u0CB9];"    // Kannada Letter A,...,Ha
+        "$KanV    = \\u0CCD;"              // Kannada Sign Virama
+        "$Kan1    = [\\u0C95-\\u0CB9];"    // Kannada Letter A,...,Ha
+        "$Mal0    = [\\u0D05-\\u0D39];"    // Malayalam Letter A,...,Ha
+        "$MalV    = \\u0D4D;"              // Malayalam Sign Virama
+        "$Mal1    = [\\u0D15-\\u0D39];"    // Malayalam Letter A,...,Ha
+        "!!chain;"
         "!!forward;"
         "$CR $LF;"
-        "([^$Control $CR $LF] | $HangulSyllable) $Extend*;"
+        "$L ($L | $V | $LV | $LVT);"
+        "($LV | $V) ($V | $T);"
+        "($LVT | $T) $T;"
+        "[^$Control $CR $LF] $Extend;"
+        "[^$Control $CR $LF] $SpacingMark;"
+        "$Hin0 $HinV $Hin1;"               // Devanagari Virama (forward)
+        "$Ben0 $BenV $Ben1;"               // Bengali Virama (forward)
+        "$Pan0 $PanV $Pan1;"               // Gurmukhi Virama (forward)
+        "$Guj0 $GujV $Guj1;"               // Gujarati Virama (forward)
+        "$Ori0 $OriV $Ori1;"               // Oriya Virama (forward)
+        "$Tel0 $TelV $Tel1;"               // Telugu Virama (forward)
+        "$Kan0 $KanV $Kan1;"               // Kannada Virama (forward)
+        "$Mal0 $MalV $Mal1;"               // Malayalam Virama (forward)
         "!!reverse;"
-        "$BackHangulSyllable = $L+ | ($T* ($V+$LV? | $LV | $LVT) $L*) | $T+;"
-        "$BackOneCluster = ($LF $CR) | ($Extend* ([^$Control $CR $LF] | $BackHangulSyllable));"
-        "$BackOneCluster;"
+        "$LF $CR;"
+        "($L | $V | $LV | $LVT) $L;"
+        "($V | $T) ($LV | $V);"
+        "$T ($LVT | $T);"
+        "$Extend      [^$Control $CR $LF];"
+        "$SpacingMark [^$Control $CR $LF];"
+        "$Hin1 $HinV $Hin0;"               // Devanagari Virama (backward)
+        "$Ben1 $BenV $Ben0;"               // Bengali Virama (backward)
+        "$Pan1 $PanV $Pan0;"               // Gurmukhi Virama (backward)
+        "$Guj1 $GujV $Guj0;"               // Gujarati Virama (backward)
+        "$Ori1 $OriV $Ori0;"               // Gujarati Virama (backward)
+        "$Tel1 $TelV $Tel0;"               // Telugu Virama (backward)
+        "$Kan1 $KanV $Kan0;"               // Kannada Virama (backward)
+        "$Mal1 $MalV $Mal0;"               // Malayalam Virama (backward)
         "!!safe_reverse;"
-        "$V+ $L;"
-        "!!safe_forward;"
-        "$V+ $T;";
-    static bool createdInputCursorIterator = false;
-    static TextBreakIterator* staticInputCursorIterator;
-    return setUpIteratorWithRules(createdInputCursorIterator, staticInputCursorIterator, kRules, string, length);
+        "!!safe_forward;";
+    static bool createdCursorMovementIterator = false;
+    static TextBreakIterator* staticCursorMovementIterator;
+    return setUpIteratorWithRules(createdCursorMovementIterator, staticCursorMovementIterator, kRules, string, length);
 #endif // BUILDING_ON_TIGER
 }