nyancrimew/LatinIME
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

am cea54014: am ae4e2036: [SD13] Reorder an enum alphabetically

Browse source 
* commit 'cea540145d65b605bb9bbe483c6210dd2f0d7d66':
  [SD13] Reorder an enum alphabetically
This commit is contained in:
Jean Chalard 2014-07-30 17:11:40 +00:00 committed by Android Git Automerger
commit 82f2a0b8b8
2 changed files with 79 additions and 79 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

34
java/res/values/attrs.xml
View file

@ -481,24 +481,24 @@
<declare-styleable name="KeyboardLayoutSet_Feature">
<!-- This should be aligned with ScriptUtils.SCRIPT_* -->
<attr name="supportedScript" format="enum">
<enum name="latin" value="0" />
<enum name="cyrillic" value="1" />
<enum name="greek" value="2" />
<enum name="arabic" value="3" />
<enum name="hebrew" value="4" />
<enum name="armenian" value="5" />
<enum name="georgian" value="6" />
<enum name="khmer" value="7" />
<enum name="lao" value="8" />
<enum name="myanmar" value="9" />
<enum name="sinhala" value="10" />
<enum name="thai" value="11" />
<enum name="telugu" value="12" />
<enum name="devanagari" value="13" />
<enum name="bengali" value="14" />
<enum name="arabic" value="0" />
<enum name="armenian" value="1" />
<enum name="bengali" value="2" />
<enum name="cyrillic" value="3" />
<enum name="devanagari" value="4" />
<enum name="georgian" value="5" />
<enum name="greek" value="6" />
<enum name="hebrew" value="7" />
<enum name="kannada" value="8" />
<enum name="khmer" value="9" />
<enum name="lao" value="10" />
<enum name="latin" value="11" />
<enum name="malayalam" value="12" />
<enum name="myanmar" value="13" />
<enum name="sinhala" value="14" />
<enum name="tamil" value="15" />
<enum name="kannada" value="16" />
<enum name="malayalam" value="17" />
<enum name="telugu" value="16" />
<enum name="thai" value="17" />
</attr>
</declare-styleable>

124
java/src/com/android/inputmethod/latin/utils/ScriptUtils.java
View file

@ -26,24 +26,24 @@ public class ScriptUtils {
// Used for hardware keyboards
public static final int SCRIPT_UNKNOWN = -1;
// TODO: should we use ISO 15924 identifiers instead?
public static final int SCRIPT_LATIN = 0;
public static final int SCRIPT_CYRILLIC = 1;
public static final int SCRIPT_GREEK = 2;
public static final int SCRIPT_ARABIC = 3;
public static final int SCRIPT_HEBREW = 4;
public static final int SCRIPT_ARMENIAN = 5;
public static final int SCRIPT_GEORGIAN = 6;
public static final int SCRIPT_KHMER = 7;
public static final int SCRIPT_LAO = 8;
public static final int SCRIPT_MYANMAR = 9;
public static final int SCRIPT_SINHALA = 10;
public static final int SCRIPT_THAI = 11;
public static final int SCRIPT_TELUGU = 12;
public static final int SCRIPT_DEVANAGARI = 13;
public static final int SCRIPT_BENGALI = 14;
public static final int SCRIPT_ARABIC = 0;
public static final int SCRIPT_ARMENIAN = 1;
public static final int SCRIPT_BENGALI = 2;
public static final int SCRIPT_CYRILLIC = 3;
public static final int SCRIPT_DEVANAGARI = 4;
public static final int SCRIPT_GEORGIAN = 5;
public static final int SCRIPT_GREEK = 6;
public static final int SCRIPT_HEBREW = 7;
public static final int SCRIPT_KANNADA = 8;
public static final int SCRIPT_KHMER = 9;
public static final int SCRIPT_LAO = 10;
public static final int SCRIPT_LATIN = 11;
public static final int SCRIPT_MALAYALAM = 12;
public static final int SCRIPT_MYANMAR = 13;
public static final int SCRIPT_SINHALA = 14;
public static final int SCRIPT_TAMIL = 15;
public static final int SCRIPT_KANNADA = 16;
public static final int SCRIPT_MALAYALAM = 17;
public static final int SCRIPT_TELUGU = 16;
public static final int SCRIPT_THAI = 17;
public static final TreeMap<String, Integer> mSpellCheckerLanguageToScript;
static {
// List of the supported languages and their associated script. We won't check
@ -83,27 +83,6 @@ public class ScriptUtils {
*/
public static boolean isLetterPartOfScript(final int codePoint, final int scriptId) {
switch (scriptId) {
case SCRIPT_LATIN:
// Our supported latin script dictionaries (EFIGS) at the moment only include
// characters in the C0, C1, Latin Extended A and B, IPA extensions unicode
// blocks. As it happens, those are back-to-back in the code range 0x40 to 0x2AF,
// so the below is a very efficient way to test for it. As for the 0-0x3F, it's
// excluded from isLetter anyway.
return codePoint <= 0x2AF && Character.isLetter(codePoint);
case SCRIPT_CYRILLIC:
// All Cyrillic characters are in the 400~52F block. There are some in the upper
// Unicode range, but they are archaic characters that are not used in modern
// Russian and are not used by our dictionary.
return codePoint >= 0x400 && codePoint <= 0x52F && Character.isLetter(codePoint);
case SCRIPT_GREEK:
// Greek letters are either in the 370~3FF range (Greek & Coptic), or in the
// 1F00~1FFF range (Greek extended). Our dictionary contains both sort of characters.
// Our dictionary also contains a few words with 0xF2; it would be best to check
// if that's correct, but a web search does return results for these words so
// they are probably okay.
return (codePoint >= 0x370 && codePoint <= 0x3FF)
|| (codePoint >= 0x1F00 && codePoint <= 0x1FFF)
|| codePoint == 0xF2;
case SCRIPT_ARABIC:
// Arabic letters can be in any of the following blocks:
// Arabic U+0600..U+06FF
@ -116,23 +95,46 @@ public class ScriptUtils {
|| (codePoint >= 0x8A0 && codePoint <= 0x8FF)
|| (codePoint >= 0xFB50 && codePoint <= 0xFDFF)
|| (codePoint >= 0xFE70 && codePoint <= 0xFEFF);
case SCRIPT_HEBREW:
// Hebrew letters are in the Hebrew unicode block, which spans from U+0590 to U+05FF,
// or in the Alphabetic Presentation Forms block, U+FB00..U+FB4F, but only in the
// Hebrew part of that block, which is U+FB1D..U+FB4F.
return (codePoint >= 0x590 && codePoint <= 0x5FF
|| codePoint >= 0xFB1D && codePoint <= 0xFB4F);
case SCRIPT_ARMENIAN:
// Armenian letters are in the Armenian unicode block, U+0530..U+058F and
// Alphabetic Presentation Forms block, U+FB00..U+FB4F, but only in the Armenian part
// of that block, which is U+FB13..U+FB17.
return (codePoint >= 0x530 && codePoint <= 0x58F
|| codePoint >= 0xFB13 && codePoint <= 0xFB17);
case SCRIPT_BENGALI:
// Bengali unicode block is U+0980..U+09FF
return (codePoint >= 0x980 && codePoint <= 0x9FF);
case SCRIPT_CYRILLIC:
// All Cyrillic characters are in the 400~52F block. There are some in the upper
// Unicode range, but they are archaic characters that are not used in modern
// Russian and are not used by our dictionary.
return codePoint >= 0x400 && codePoint <= 0x52F && Character.isLetter(codePoint);
case SCRIPT_DEVANAGARI:
// Devanagari unicode block is +0900..U+097F
return (codePoint >= 0x900 && codePoint <= 0x97F);
case SCRIPT_GEORGIAN:
// Georgian letters are in the Georgian unicode block, U+10A0..U+10FF,
// or Georgian supplement block, U+2D00..U+2D2F
return (codePoint >= 0x10A0 && codePoint <= 0x10FF
|| codePoint >= 0x2D00 && codePoint <= 0x2D2F);
case SCRIPT_GREEK:
// Greek letters are either in the 370~3FF range (Greek & Coptic), or in the
// 1F00~1FFF range (Greek extended). Our dictionary contains both sort of characters.
// Our dictionary also contains a few words with 0xF2; it would be best to check
// if that's correct, but a web search does return results for these words so
// they are probably okay.
return (codePoint >= 0x370 && codePoint <= 0x3FF)
|| (codePoint >= 0x1F00 && codePoint <= 0x1FFF)
|| codePoint == 0xF2;
case SCRIPT_HEBREW:
// Hebrew letters are in the Hebrew unicode block, which spans from U+0590 to U+05FF,
// or in the Alphabetic Presentation Forms block, U+FB00..U+FB4F, but only in the
// Hebrew part of that block, which is U+FB1D..U+FB4F.
return (codePoint >= 0x590 && codePoint <= 0x5FF
|| codePoint >= 0xFB1D && codePoint <= 0xFB4F);
case SCRIPT_KANNADA:
// Kannada unicode block is U+0C80..U+0CFF
return (codePoint >= 0xC80 && codePoint <= 0xCFF);
case SCRIPT_KHMER:
// Khmer letters are in unicode block U+1780..U+17FF, and the Khmer symbols block
// is U+19E0..U+19FF
@ -141,6 +143,16 @@ public class ScriptUtils {
case SCRIPT_LAO:
// The Lao block is U+0E80..U+0EFF
return (codePoint >= 0xE80 && codePoint <= 0xEFF);
case SCRIPT_LATIN:
// Our supported latin script dictionaries (EFIGS) at the moment only include
// characters in the C0, C1, Latin Extended A and B, IPA extensions unicode
// blocks. As it happens, those are back-to-back in the code range 0x40 to 0x2AF,
// so the below is a very efficient way to test for it. As for the 0-0x3F, it's
// excluded from isLetter anyway.
return codePoint <= 0x2AF && Character.isLetter(codePoint);
case SCRIPT_MALAYALAM:
// Malayalam unicode block is U+0D00..U+0D7F
return (codePoint >= 0xD00 && codePoint <= 0xD7F);
case SCRIPT_MYANMAR:
// Myanmar has three unicode blocks :
// Myanmar U+1000..U+109F
@ -152,27 +164,15 @@ public class ScriptUtils {
case SCRIPT_SINHALA:
// Sinhala unicode block is U+0D80..U+0DFF
return (codePoint >= 0xD80 && codePoint <= 0xDFF);
case SCRIPT_THAI:
// Thai unicode block is U+0E00..U+0E7F
return (codePoint >= 0xE00 && codePoint <= 0xE7F);
case SCRIPT_TELUGU:
// Telugu unicode block is U+0C00..U+0C7F
return (codePoint >= 0xC00 && codePoint <= 0xC7F);
case SCRIPT_DEVANAGARI:
// Devanagari unicode block is +0900..U+097F
return (codePoint >= 0x900 && codePoint <= 0x97F);
case SCRIPT_BENGALI:
// Bengali unicode block is U+0980..U+09FF
return (codePoint >= 0x980 && codePoint <= 0x9FF);
case SCRIPT_TAMIL:
// Tamil unicode block is U+0B80..U+0BFF
return (codePoint >= 0xB80 && codePoint <= 0xBFF);
case SCRIPT_KANNADA:
// Kannada unicode block is U+0C80..U+0CFF
return (codePoint >= 0xC80 && codePoint <= 0xCFF);
case SCRIPT_MALAYALAM:
// Malayalam unicode block is U+0D00..U+0D7F
return (codePoint >= 0xD00 && codePoint <= 0xD7F);
case SCRIPT_TELUGU:
// Telugu unicode block is U+0C00..U+0C7F
return (codePoint >= 0xC00 && codePoint <= 0xC7F);
case SCRIPT_THAI:
// Thai unicode block is U+0E00..U+0E7F
return (codePoint >= 0xE00 && codePoint <= 0xE7F);
case SCRIPT_UNKNOWN:
return true;
default: