organicmaps/icu
13
Fork 0
mirror of https://github.com/unicode-org/icu.git synced 2025-04-15 09:45:26 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 1

ICU-6365 Only write the language code after a linefeed if there are more characters after the linefeed. Update data driven test to check for this.

Browse source 
X-SVN-Rev: 24343
This commit is contained in:
Michael Ow 2008-07-03 15:27:10 +00:00
parent b422a026dd
commit d7ccec764b
3 changed files with 30 additions and 26 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

48
icu4j/src/com/ibm/icu/charset/CharsetISCII.java
View file

@ -838,7 +838,7 @@ class CharsetISCII extends CharsetICU {
//data.contextCharToUnicode; /* contains previous ISCII codepoint visited */
//this.toUnicodeStatus; /* contains the mapping to Unicode of the above codepoint */
while (!cr.isError() && source.hasRemaining()) {
while (source.hasRemaining()) {
targetUniChar = UConverterConstants.missingCharMarker;
if (target.hasRemaining()) {
@ -1006,16 +1006,16 @@ class CharsetISCII extends CharsetICU {
* convert the code point given based on the delta provided.
*/
cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_RRA, (short)0);
if (!cr.isError()) {
if (!cr.isOverflow()) {
cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_SIGN_VIRAMA, (short)0);
if (!cr.isError()) {
if (!cr.isOverflow()) {
cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_HA, (short)0);
} else {
this.charErrorBufferArray[this.charErrorBufferLength++] = 0x0939;
this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_HA;
}
} else {
this.charErrorBufferArray[this.charErrorBufferLength++] = 0x094D;
this.charErrorBufferArray[this.charErrorBufferLength++] = 0x0939;
this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_SIGN_VIRAMA;
this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_HA;
}
this.toUnicodeStatus = UConverterConstants.missingCharMarker;
data.contextCharToUnicode = NO_CHAR_MARKER;
@ -1125,7 +1125,7 @@ class CharsetISCII extends CharsetICU {
} //end of while
if (!cr.isError() && !cr.isOverflow() && flush && !source.hasRemaining()) {
if (cr.isUnderflow() && flush && !source.hasRemaining()) {
/*end of the input stream */
if (data.contextCharToUnicode == ATR || data.contextCharToUnicode == EXT || data.contextCharToUnicode == ISCII_INV) {
/* set toUBytes[] */
@ -1224,31 +1224,35 @@ class CharsetISCII extends CharsetICU {
}
/* writing the char to the output stream */
while (!cr.isError() && source.hasRemaining()) {
while (source.hasRemaining()) {
if (!target.hasRemaining()) {
return CoderResult.OVERFLOW;
}
targetByteUnit = UConverterConstants.missingCharMarker;
/* Write the language code following LF only if LF is not the last character. */
if (fromUnicodeStatus == LF) {
targetByteUnit = ATR << 8;
targetByteUnit += (byte)lookupInitialData[range].isciiLang;
fromUnicodeStatus = 0x0000;
/* now append ATR and language code */
cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
if (cr.isOverflow()) {
break;
}
}
sourceChar = source.get();
tempContextFromUnicode = converterData.contextCharFromUnicode;
targetByteUnit = UConverterConstants.missingCharMarker;
/* check if input is in ASCII and C0 control codes range */
if (sourceChar <= ASCII_END) {
fromUnicodeStatus = sourceChar;
cr = WriteToTargetFromU(offsets, source, target, sourceChar);
if (cr.isError()) {
if (cr.isOverflow()) {
break;
}
if (sourceChar == LF) {
targetByteUnit = ATR << 8;
targetByteUnit += (byte)lookupInitialData[range].isciiLang;
fromUnicodeStatus = sourceChar;
/* now append ATR and language code */
cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
if (cr.isOverflow()) {
break;
}
}
continue;
}
@ -1346,7 +1350,7 @@ class CharsetISCII extends CharsetICU {
targetByteUnit = targetByteUnit << 16 | ISCII_HALANT << 8 | targetByteUnit;
/*write targetByteUnit to target */
cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
if (cr.isError()) {
if (cr.isOverflow()) {
break;
}
} else if (targetByteUnit != UConverterConstants.missingCharMarker) {
@ -1355,7 +1359,7 @@ class CharsetISCII extends CharsetICU {
}
/*write targetByteUnit to target */
cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
if (cr.isError()) {
if (cr.isOverflow()) {
break;
}
} else if (UTF16.isSurrogate((char)sourceChar)) {

4
icu4j/src/com/ibm/icu/dev/data/testdata.jar
View file

@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:145bae6106ca7dee045f126cd5e9f21145a2cf93bcb8ce2969a0f5d37efe103f
size 764988
oid sha256:5a0b40590ed99edc5b53d8ad2bd7a975840f281d2285760c020bef50e5492de4
size 765310

4
icu4j/src/com/ibm/icu/dev/test/charset/TestCharset.java
View file

@ -3432,8 +3432,8 @@ public class TestCharset extends TestFmwk {
ccus.clear();
//test ascii overflow buffer
ccus.put((char)0x0A);
ccbs.put((byte)0x00);
ccus.put((char)0x0A); ccus.put((char)0x0043);
ccbs.put((byte)0x00); ccbs.put((byte)0x00);
ccus.limit(ccus.position());
ccus.position(0);