DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-08-15 18:37 --- I was away from online for a last week. :( The work on this bug is depends on bug #11010. As I have wrote, I think there is no bug in the patch you have rolled back. Please, look at the last posts on this bug. About more Unicode support: while offline I have done unicode support for FormatRecord. Also I have some questions on NameRecord. I have made the implemetation for it for Page Titles, but I am not sure in the way I have done it. Please look at the bug #11010, because the Unicode changes depend on it. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-08-15 19:57 --- sorry I missed the copy of the factory file... (*nudge* *nudge* probably because it wasn't a patch).. . Looks like you nailed it. Thanks! -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976
[Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
--- Additional Comments From [EMAIL PROTECTED] 2002-07-29 08:59 ---
Hi Andy, Sergei and all
I have checked the following unit test failure:
Class org.apache.poi.hssf.record.TestBoundSheetRecord
Name Tests Errors Failures Time(s)
TestBoundSheetRecord 2 0 1 0.581
Tests
Name Status Type Time(s)
testRecordLength Success 0.020
testWideRecordLength Failure 2 + 2 + 4 + 2 + 1 + 1 + len(str) * 2 expected:24
but was:18
junit.framework.AssertionFailedError: 2 + 2 + 4 + 2 + 1 + 1 + len(str) * 2
expected:24 but was:18
at
org.apache.poi.hssf.record.TestBoundSheetRecord.testWideRecordLength(TestBoundSheetRecord.java:93)
0.020
The test is as follow:
public void testWideRecordLength()
throws Exception
{
BoundSheetRecord record = new BoundSheetRecord();
record.setCompressedUnicodeFlag((byte)0x01);
record.setSheetname(Sheet1);
record.setSheetnameLength((byte)6);
assertEquals( 2 + 2 + 4 + 2 +1 +1+ len(str) *
2, 24,
record.getRecordSize());
}
The setSheetname of BoundSheetRecord is as follow:
public void setSheetname(String sheetname)
{
boolean is16bit = is16bitString(sheetname);
setSheetnameLength((byte) sheetname.length() );
setCompressedUnicodeFlag((byte ) (is16bit?1:0));
field_5_sheetname = sheetname;
}
The unit test failed because the setSheetname use autodetection in deciding
whether the sheetname can be represented in excel compressed unicode format. If
yes, than it set the CompressedUnicodeFlag to 0 i.e. 8 bit representation. It
simply ignore the following statement in testWideRecordLength():
record.setCompressedUnicodeFlag((byte)0x01);
When I submitted the patch, the test was not there yet. So now, the question is
whether we want autodetection in the setSheetname( and later, in the cell
setCellValue). The following are some pro and cons :
It seems for the current situation, the programmer (the user of the POI API)has
to do more work by setting
record.setCompressedUnicodeFlag((byte)0x01);
for sheetname or
cell.setEncoding(org.apache.poi.hssf.usermodel.HSSFCell.ENCODING_UTF_16);
for cell value.
And in case the programmer make a inconsistence mistake, ex. set for
cell.setEncoding(org.apache.poi.hssf.usermodel.HSSFCell.ENCODING_COMPRESSED_UNICODE
);
and then
setCellValue(\u0422\u0435\u0441\u0442\u043E\u0432\u0430\u044F);
What exception should be thrown?
If autodetection is used, the programmer code less, and consistence checking is
avoided.
Please conside each alternative and I will change the code accordingly
Patrick Lee
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-29 11:48 --- I think we decided against autodetection because it doesn't work with russian and other langauges anyhow. The default should be 8-bit and using 16-bit should be optional. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-29 19:23 --- I have a succes and wrote it back to you vai e-mail, but juxt forgot, that it succeded on my changes. %) And have took a look on my changes. As I see it is a very good idea to put the code of using in the Records UnicodeString class. Let's merge our work on the unicode getting and putting we have into the UnicodeString. Could you see at the code at BoundSheetRecord, when I have refined you step: using only StringUtil without SSTSerializer? What do you think about putting it into the UnicodeSrting, so we can benefit on using this class and instantiating it in any record we need Unicode? -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-30 04:03 --- Hi Andy, Sergei and all I have a succes and wrote it back to you vai e-mail, but juxt forgot, that it succeded on my changes. %) And have took a look on my changes. As I see it is a very good idea to put the code of using in the Records UnicodeString class. Great! can you send me a copy of the final code that works for you. Let's merge our work on the unicode getting and putting we have into the UnicodeString. Sure. Let fix this bug together. I have some planning on futher refactor/complete the UnicodeString class. Lets collaborate on this issue. Could you see at the code at BoundSheetRecord, when I have refined you step: using only StringUtil without SSTSerializer? Please check http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 for [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class as it refactor code regarding BIFF8 format from SSTDeserializer to UnicodeString class. What do you think of this idea. What do you think about putting it into the UnicodeSrting, so we can benefit on using this class and instantiating it in any record we need Unicode? Do you mean code described in? http://nagoya.apache.org/bugzilla/showattachment.cgi?attach_id=2422 I am really interesting in working on this together. Thanks Patrick Lee -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-28 23:24 --- Class org.apache.poi.hssf.record.TestBoundSheetRecord Name Tests Errors Failures Time(s) TestBoundSheetRecord 2 0 1 0.581 Tests Name Status Type Time(s) testRecordLength Success 0.020 testWideRecordLength Failure 2 + 2 + 4 + 2 + 1 + 1 + len(str) * 2 expected:24 but was:18 junit.framework.AssertionFailedError: 2 + 2 + 4 + 2 + 1 + 1 + len(str) * 2 expected:24 but was:18 at org.apache.poi.hssf.record.TestBoundSheetRecord.testWideRecordLength(TestBoundSheetRecord.java:93) 0.020 Fix this and I'll apply it. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-27 10:34 --- Could it be due to the newly commited changed to the TestBoundSheetRecord.java glens 2002/07/26 18:45:44 Added: src/testcases/org/apache/poi/hssf/record TestBoundSheetRecord.java Log: Test case for bound sheet record... it seems okay. Revision ChangesPath 1.1 jakarta- poi/src/testcases/org/apache/poi/hssf/record/TestBoundSheetRecord.java Thanks for looking at it Patrick Lee -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class --- Additional Comments From [EMAIL PROTECTED] 2002-07-27 01:24 --- Hi. I tried to apply this but the unit tests were failing so I'm going to look at it later along with that bug. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
DO NOT REPLY [Bug 10976] - [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10976 [Patch] Unicode Support for sheetname , refactor SSTDeserializer UnicodeString class [EMAIL PROTECTED] changed: What|Removed |Added Status|RESOLVED|REOPENED Resolution|INVALID | --- Additional Comments From [EMAIL PROTECTED] 2002-07-25 12:23 --- Sorry I didn't notice it. You didn't reopen the bug. (I may sound silly or pedantic, I'm just busy, I scan the list of bugs for [PATCH] and open status when reviewing). I'll review this shortly. Right now, I have to go to work. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
