Nabble - Apache POI

Background color for symbol/textpart in HWPFDocument

2009-12-03T21:52:18Z

Hi,
How can i get background color for part of .doc-document?
(need it for doc2html)

Thnks

DO NOT REPLY [Bug 48336] Error reading date

2009-12-03T15:17:29Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

Josh Micich <josh@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|REOPENED |NEEDINFO

--- Comment #4 from Josh Micich <josh@...> 2009-12-03 15:17:26 UTC ---
A short term work around is for you to use methods from DateUtil like
setCalendar to calculate dates with negative 'serial date' values. However it
seems like you might be requesting a real change (to how POI behaves).

Normally such an enhancement request is rejected on the grounds that it
involves POI doing something that Excel considers invalid. However, I agree
that the functionality you desire is very reasonable. I think the restriction
on dates before 1900 is silly. On the other hand, we can't go about
implementing every request for features from other applications that go beyond
the Excel file format spec.

I want to wait for comments from other POI committers before doing anything.
There is also the general question of how we accept/reject/prioritize bugs and
feature requests that raise issues of compliance with other applications. I
don't know of any POI documentation that deals with this.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 46479] [PATCH] evaluateInCell fails if the formula was previously invalid

2009-12-03T14:23:55Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=46479

--- Comment #4 from Josh Micich <josh@...> 2009-12-03 14:23:52 UTC ---
Additional similar fixes applied in svn r886951

Problems occurred when changing a formula cell to type string. These cases
weren't addressed in the previous fix.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: a 'lite' version of ooxml-schemas jar

2009-12-03T13:45:29Z

Hi,

On Thu, Dec 3, 2009 at 5:58 PM, Yegor Kozlov <yegor@...> wrote:
> the problem should be fixed in r886733.
> At least, Tika trunk compiles OK against poi-ooxml-schemas produced from POI
> trunk. JUnits run OK too.

Excellent, thanks!

BR,

Jukka Zitting

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48336] Error reading date

2009-12-03T10:14:08Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

vdhanda@... changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|INVALID |

--- Comment #3 from vdhanda@... 2009-12-03 10:14:08 UTC ---
Can you try opening it in OpenOffice. It shows the correct value as
12/31/1896.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48336] Error reading date

2009-12-03T10:04:24Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

Josh Micich <josh@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |INVALID

--- Comment #2 from Josh Micich <josh@...> 2009-12-03 10:04:23 UTC ---
There is something wrong with cell D1. It displays as
'#######################' in my Excel 2007. Excel has roll-over tip text that
provides more information: "Dates and times that are negative or too large
display as ######."
Indeed the cell value is -1094
HSSFCell.getNumericCellValue() should return -1094.0 and
HSSFCell.getDateCellValue() should return null due to this invalid date value.

The cell has a custom format 'mmm-dd-yyyy' which suggests that the application
that wrote this file intended for the cell to have a date value.

I am marking this bug as invalid because it appears that the problem is with
"SoftArtisans OfficeWriter"

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: Text Substitution for Word 2007 in POI 3.5

2009-12-03T10:01:02Z

Sorry to say Rafael that I am not sure what it is that you are asking. I have just a a very quick - cursory - glance at the javadoc for the XWPFRun class and it is possible to call methods that will recover the text the run contains and to set the text it ought to encapsulate. Would it not be possible to recover the text, use Java's String handling capabilities to search for and replace text and then re-set the runs text?

Yours

Mark B

Rafael Paiva wrote:

I assume it's not available. Any plans for having this back on 3.6? Any
undocumented workaround for 3.5?

Thanks!

Rafael.

From forum: POI - Dev

Re: DO NOT REPLY [Bug 48336] New: Error reading date

2009-12-03T09:56:12Z

Can I ask, what does POI think the value in that cell is? I would guess that it is telling you that the cell holds a numeric value.

You do know that dates are actually stored as numeric values that are then 'converted' into the familair format by the formatting that is applied to the cell? If you are expecting to see a formatted date value recovered from the cell then you will need to use something like the DataFormatter class to convert the numeric value into the form you expect to see.

Yours

Mark B

Bugzilla from bugzilla@apache.org wrote:

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

Summary: Error reading date
Product: POI
Version: 3.5-FINAL
Platform: Other
OS/Version: Linux
Status: NEW
Severity: critical
Priority: P2
Component: HSSF
AssignedTo: dev@poi.apache.org
ReportedBy: vdhanda@epistemic.com

Created an attachment (id=24665)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24665)
Excel file

The attached Excel has a date value in row 1, column D. POI doesn't read it as
a date. The file was created by SoftArtisans OfficeWriter for Excel 7.0.0.1642
(http://officewriter.softartisans.com).

Thanks.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@poi.apache.org
For additional commands, e-mail: dev-help@poi.apache.org

From forum: POI - Dev

DO NOT REPLY [Bug 48336] Error reading date

2009-12-03T09:47:10Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

--- Comment #1 from vdhanda@... 2009-12-03 09:47:09 UTC ---
Created an attachment (id=24666)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24666)
BiffViewer dump

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48336] Error reading date

2009-12-03T09:46:19Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

vdhanda@... changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |vdhanda@...

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48336] New: Error reading date

2009-12-03T09:46:10Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48336

Summary: Error reading date
Product: POI
Version: 3.5-FINAL
Platform: Other
OS/Version: Linux
Status: NEW
Severity: critical
Priority: P2
Component: HSSF
AssignedTo: dev@...
ReportedBy: vdhanda@...

Created an attachment (id=24665)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24665)
Excel file

The attached Excel has a date value in row 1, column D. POI doesn't read it as
a date. The file was created by SoftArtisans OfficeWriter for Excel 7.0.0.1642
(http://officewriter.softartisans.com).

Thanks.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: a 'lite' version of ooxml-schemas jar

2009-12-03T08:58:04Z

the problem should be fixed in r886733.
At least, Tika trunk compiles OK against poi-ooxml-schemas produced from POI trunk. JUnits run OK too.

Regards,
Yegor

> Hi,
>
> On Wed, Dec 2, 2009 at 7:58 PM, Yegor Kozlov <yegor@...> wrote:
>> Can you point me at the place in Tika where getBookmarkStartArray() is used?
>
> See line 78 of o.a.tika.parser.microsoft.ooxml.XWPFWordExtractorDecorator [1].
>
> [1] http://svn.apache.org/viewvc/lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/microsoft/ooxml/XWPFWordExtractorDecorator.java?revision=820962&view=markup
>
> BR,
>
> Jukka Zitting
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@...
> For additional commands, e-mail: dev-help@...
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: HSSFCell short/int

2009-12-03T05:48:27Z

Hi Josh,

thanks for that thorough explanation.
My formulation was rather ambiguous.

The need for migrating to int is clear.

My issue is that for Rows it makes sense in the HSSF world to
convert to "unsigned short" by AND'ing with 65535, as that is
the maximum no. of rows, but for columns the max is 256, so it
is really quite misleading, because the signed range of the
short is sufficient.

The HSSFCell.getCell(short) code...
int ushortCellNum = cellnum & 0x0000FFFF; // avoid sign extension
return getCell(ushortCellNum);
...raises an issue with the beholder that is not relevant.

Maybe I can illustrate that with an example:
I actually noticed this while migrating code from the deprecated
method to the int method. My first shot was to AND my short with
65535 before calling the int method, so you see, I was forced to
deal with an issue that was irrelevant. My code would of course
have worked with the AND, but its not necessary.

I would replace the 2 lines with...
return getCell((int)cellnum);

In any case, I would think its better to give an out-of-range error
that mentions the value supplied & not its unsigned equivalent.

Regards,
DaveLaw

----- original message --------

Subject: Re: HSSFCell short/int
Sent: Wed, 02 Dec 2009
From: Josh Micich<josh.micich@...>

> Hello Dave,
>
> It's not clear if you are questioning the merit of creating the newer
> method "getCell(int)", or whether the old method "getCell(short)"
> should have been deprecated at all, so I'll attempt to answer both
> concerns.
>
>
> The most visible reason why POI is moving away from bytes and shorts
> is the need for typecasts when using literal integer constants, even
> when the value is quite obviously within range. For example without
> the new method "row.getCell(5)" would not compile. We would have to
> write "row.getCell((short)5)".
>
> There have been several POI bugs involving representation of unsigned
> (ushort, ubyte) quantities with java datatypes (signed) of the same
> width (short, byte). When shorts and bytes are used, the corrected
> code generally needs more typecasts and bit mask expressions. If
> those datatypes are changed to int, the code is usually simpler and
> easier to read.
>
> The new method "getCell(int) method may in the first place have been
> introduced as a result of the XSSF common interface work (a separate
> reason, as suggested by Mark B).
>
>
> That sort-of explains the rationale for creating the new method
> "getCell(int)". The reason for deprecating the old is to eliminate
> overloading (of "getCell"). Method overloading is sometimes very
> handy but has quite a few pitfalls, and for that reason POI tries to
> use overloading sparingly.
>
> As far as the exact meaning of the '@deprecated' tag, in POI (unlike
> in the java runtime library) it means "This method will be removed in
> a future POI version". An alternative is always given. A deprecation
> date has been provided in most places, to help POI users prioritize
> clean-up of their own code.
>
>
> You mentioned the comment from the the first line of the deprecated method:
> int ushortCellNum = cellnum & 0x0000FFFF; // avoid sign extension
> The bit-mask is correct for 16-bit unsigned conversion. Masking with
> 0xFF would be wrong because it would silently convert every possible
> input to a valid column index. For example: is it sensible for
> "getCell((short)300)" to succeed? Maybe it looks pointless to
> convert values in the range (-32768...-1) to (32768...65535) is
> because the maximum column index is 255. This conversion to unsigned
> 16-bit has been done because it's more likely that the caller intends
> an unsigned quantity (column indexes are never negative). So if a
> negative value *does* get into this method, it's probably better to
> format the error message in terms of the unsigned 16 bit quantity.
> Similar results could have been achieved with bounds checking the
> argument, but would involve duplicating all the MissingCellPolicy
> logic too.
> So - that line of code is there for making better error messages, and
> has little to do with the reasons for the deprecation.
>
> regards,
> Josh
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@...
> For additional commands, e-mail: user-help@...
>
>

--- original message end ----

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@...
For additional commands, e-mail: user-help@...

From forum: POI - User

DO NOT REPLY [Bug 48327] cloneSheet() gives an "Error file: data may have been lost" in Excel 2003 or higher if sheet contains commented cells

2009-12-03T02:29:10Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48327

Manuela <manuela.munaretto@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Summary|cloneSheet() gives an |cloneSheet() gives an
|"Error file: data may have |"Error file: data may have
|been lost" in Excel 2003 or |been lost" in Excel 2003 or
|higher if sheet commented |higher if sheet contains
|cells |commented cells

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48327] cloneSheet() gives an "Error file: data may have been lost" in Excel 2003 or higher if sheet commented cells

2009-12-03T02:28:36Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48327

Manuela <manuela.munaretto@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Summary|cloneSheet() gives an |cloneSheet() gives an
|"Error file: data may have |"Error file: data may have
|been lost" in Excel 2003 or |been lost" in Excel 2003 or
|higher |higher if sheet commented
| |cells

--- Comment #1 from Manuela <manuela.munaretto@...> 2009-12-03 02:28:35 UTC ---
I tried to cancel all comments and Excel opens file without pop-up-ing error
message: "Error file: data may have been lost".

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] Unable to read Excel file

2009-12-02T13:30:16Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

Josh Micich <josh@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED

--- Comment #4 from Josh Micich <josh@...> 2009-12-02 13:30:14 UTC ---
Fixed in svn r886311

junits added

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: a 'lite' version of ooxml-schemas jar

2009-12-02T12:35:03Z

Hi,

On Wed, Dec 2, 2009 at 7:58 PM, Yegor Kozlov <yegor@...> wrote:
> Can you point me at the place in Tika where getBookmarkStartArray() is used?

See line 78 of o.a.tika.parser.microsoft.ooxml.XWPFWordExtractorDecorator [1].

[1] http://svn.apache.org/viewvc/lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/microsoft/ooxml/XWPFWordExtractorDecorator.java?revision=820962&view=markup

BR,

Jukka Zitting

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] Unable to read Excel file

2009-12-02T11:59:37Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

vdhanda@... changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEEDINFO |NEW

--- Comment #3 from vdhanda@... 2009-12-02 11:59:36 UTC ---
Hello Josh,

Yes, if I open and save the file in Excel and/or OpenOffice, the POI can read
it.

The stack trace from BiffViewer is at the bottom of the BiffViewer output.
Here's the stack trace from my application.

Exception in thread "main" org.apache.poi.hssf.record.RecordFormatException:
Unable to construct record instance
at
org.apache.poi.hssf.record.RecordFactory$ReflectionRecordCreator.create(RecordFactory.java:64)
at
org.apache.poi.hssf.record.RecordFactory.createSingleRecord(RecordFactory.java:263)
at
org.apache.poi.hssf.record.RecordFactoryInputStream.readNextRecord(RecordFactoryInputStream.java:270)
at
org.apache.poi.hssf.record.RecordFactoryInputStream.nextRecord(RecordFactoryInputStream.java:236)
at
org.apache.poi.hssf.record.RecordFactory.createRecords(RecordFactory.java:377)
at org.apache.poi.hssf.usermodel.HSSFWorkbook.<init>(HSSFWorkbook.java:275)
at org.apache.poi.hssf.usermodel.HSSFWorkbook.<init>(HSSFWorkbook.java:200)
at org.apache.poi.hssf.usermodel.HSSFWorkbook.<init>(HSSFWorkbook.java:182)
at
com.epistemic.http.ExcelHandler.handleResponseBody(ExcelHandler.java:126)
at com.aexp.pl.Tester.main(Tester.java:17)
Caused by: java.lang.RuntimeException: Unusual record size remaining=(0)
at
org.apache.poi.hssf.record.ColumnInfoRecord.<init>(ColumnInfoRecord.java:72)
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
at
sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
at
org.apache.poi.hssf.record.RecordFactory$ReflectionRecordCreator.create(RecordFactory.java:56)
... 9 more

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] Unable to read Excel file

2009-12-02T11:55:07Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

Josh Micich <josh@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO

--- Comment #2 from Josh Micich <josh@...> 2009-12-02 11:55:06 UTC ---
Can you please attach the stack-trace of your error? Also, can you investigate
whether re-saving the file with Excel allows POI to read it?

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] Unable to read Excel file

2009-12-02T11:52:29Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

--- Comment #1 from vdhanda@... 2009-12-02 11:52:28 UTC ---
Created an attachment (id=24662)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24662)
The Excel

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] Unable to read Excel file

2009-12-02T11:51:16Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

vdhanda@... changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |vdhanda@...

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48332] New: Unable to read Excel file

2009-12-02T11:50:55Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48332

Summary: Unable to read Excel file
Product: POI
Version: 3.5-FINAL
Platform: Other
OS/Version: Linux
Status: NEW
Severity: blocker
Priority: P2
Component: HSSF
AssignedTo: dev@...
ReportedBy: vdhanda@...

Created an attachment (id=24661)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24661)
BiffViewer output

I have an Excel file that POI is not able to read. MS Excel and OpenOffice can
read it. Attached is the BiffViewer output. The Excel was created by another
process using SoftArtisans OfficeWriter for Excel 7.0.0.1642
(http://officewriter.softartisans.com).

I really need to read this file via POI. Please help.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: HSSFCell short/int

2009-12-02T11:27:06Z

Hello Dave,

It's not clear if you are questioning the merit of creating the newer
method "getCell(int)", or whether the old method "getCell(short)"
should have been deprecated at all, so I'll attempt to answer both
concerns.

The most visible reason why POI is moving away from bytes and shorts
is the need for typecasts when using literal integer constants, even
when the value is quite obviously within range. For example without
the new method "row.getCell(5)" would not compile. We would have to
write "row.getCell((short)5)".

There have been several POI bugs involving representation of unsigned
(ushort, ubyte) quantities with java datatypes (signed) of the same
width (short, byte). When shorts and bytes are used, the corrected
code generally needs more typecasts and bit mask expressions. If
those datatypes are changed to int, the code is usually simpler and
easier to read.

The new method "getCell(int) method may in the first place have been
introduced as a result of the XSSF common interface work (a separate
reason, as suggested by Mark B).

That sort-of explains the rationale for creating the new method
"getCell(int)". The reason for deprecating the old is to eliminate
overloading (of "getCell"). Method overloading is sometimes very
handy but has quite a few pitfalls, and for that reason POI tries to
use overloading sparingly.

As far as the exact meaning of the '@deprecated' tag, in POI (unlike
in the java runtime library) it means "This method will be removed in
a future POI version". An alternative is always given. A deprecation
date has been provided in most places, to help POI users prioritize
clean-up of their own code.

You mentioned the comment from the the first line of the deprecated method:
int ushortCellNum = cellnum & 0x0000FFFF; // avoid sign extension
The bit-mask is correct for 16-bit unsigned conversion. Masking with
0xFF would be wrong because it would silently convert every possible
input to a valid column index. For example: is it sensible for
"getCell((short)300)" to succeed? Maybe it looks pointless to
convert values in the range (-32768...-1) to (32768...65535) is
because the maximum column index is 255. This conversion to unsigned
16-bit has been done because it's more likely that the caller intends
an unsigned quantity (column indexes are never negative). So if a
negative value *does* get into this method, it's probably better to
format the error message in terms of the unsigned 16 bit quantity.
Similar results could have been achieved with bounds checking the
argument, but would involve duplicating all the MissingCellPolicy
logic too.
So - that line of code is there for making better error messages, and
has little to do with the reasons for the deprecation.

regards,
Josh

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@...
For additional commands, e-mail: user-help@...

From forum: POI - User

Re: a 'lite' version of ooxml-schemas jar

2009-12-02T10:58:55Z

it for the next POI release.
>
> The only odd thing about the upgrade was that I needed to comment out
> a piece of Tika extraction code that uses the
> org.openxmlformats.schemas.wordprocessingml.x2006.main.CTBookmark
> class as returned from XWPFParagraph.getCTP().getBookmarkStartArray().
> It looks like that class is not included in the poi-ooxml-schemas jar
> even though the CTP class with the getBookmarkStartArray() method is
> there.
>

Can you point me at the place in Tika where getBookmarkStartArray() is used?
ooxml-lite only includes classes called during execution of junits. getBookmarkStartArray is not covered by the tests
and it explains why the CTBookmark class is missing.

Yegor

> BR,
>
> Jukka Zitting
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@...
> For additional commands, e-mail: dev-help@...
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: release of POI-3.6

2009-12-02T10:36:50Z

Hi,

On Tue, Dec 1, 2009 at 3:50 PM, Nick Burch <nick@...> wrote:
> I'd probably lean towards doing a beta now (eg roll it tomorrow or thurs, 72
> hour vote, on the mirrors by monday), then release final just into the new
> year.

+1 I'd really love that!

I'm hoping to have a Tika 0.6 release done next week, and being able
to base it on a POI 3.6 beta (or even an alpha if there's some stuff
that's not ready for beta status) would be perfect.

BR,

Jukka Zitting

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Re: a 'lite' version of ooxml-schemas jar

2009-12-02T10:28:31Z

Hi,

On Tue, Nov 24, 2009 at 11:02 AM, Yegor Kozlov <yegor@...> wrote:
> For Maven this change is transparent - POM for the poi-ooxml module depends
> on poi-ooxml-schemas instead of ooxml-schemas, this means Maven users will
> only need to update the version of POI from 3.5-FINAL to 3.6, the rest will
> be handled by Maven automatically.

I just had a chance to test this with Tika, and it works beautifully.
After upgrading to a POI 3.6-beta1-20091202 snapshot the size of the
tika-app jar dropped from 25MB to 15MB. That's a major improvement,
thanks! I can't wait for the next POI release.

The only odd thing about the upgrade was that I needed to comment out
a piece of Tika extraction code that uses the
org.openxmlformats.schemas.wordprocessingml.x2006.main.CTBookmark
class as returned from XWPFParagraph.getCTP().getBookmarkStartArray().
It looks like that class is not included in the poi-ooxml-schemas jar
even though the CTP class with the getBookmarkStartArray() method is
there.

BR,

Jukka Zitting

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

Text Substitution for Word 2007 in POI 3.5

2009-12-02T10:26:45Z

I assume it's not available. Any plans for having this back on 3.6? Any
undocumented workaround for 3.5?

Thanks!

Rafael.

From forum: POI - Dev

Re: HSSFCell short/int

2009-12-02T08:42:48Z

I am only guessing David but the change my be the result of introducing the SS model to accomodate both the binary and OpenXML based file formats 'invisibly'. If you look at the org.apache.poi.ss.usermodel package, you will find there are classes called Workbook, Sheet, Cell, etc and these are used by those who may have to work with both types of file format and do not want to maintain two different code bases. Well, both HSSFCell and XSSFCell both implement the Cell interface and I would guess that the decision was taken to include just the single method getCell(int) in the interface and - as a result - both concrete classes because the OpenXML file format does not impose the 256 row limit on the user.

As I said, this is only a guess.

Yours

Mark B

David Law-2 wrote:

Could it be, the HSSFCell getCell(short cellnum)
method has been rather confusingly deprecated?

Arguably, its laudable to "avoid sign extension",
but as there can only be 256 columns anyway,
I find it really rather misleading.

Regards,
DaveLaw

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@poi.apache.org
For additional commands, e-mail: user-help@poi.apache.org

From forum: POI - User

Re: Determine version of excel file

2009-12-02T08:36:21Z

Hello again,

If you code to the 'SS' model then you will not need to check the type returned by the WorkbookFactory at all. By this. I mean that you will use objects from the org.apache.poi.ss.usermodel package such as Workbook, Sheet and Cell in your program. That way, you need have no concerns at all about the actual type - and therefore the file format - that you are dealing with. Before deciding to move your code in this direction however, it would be wise to look closely at the methods defined on the various interfaces just to make sure that you can accomplish everything you require. Just as am example, on HSSFSheet you can add a data validation whuilst this methgod has not yet been declared within the Sheet interface. Should you need to extended functionality then of course you can just test the type returned to you by the WorkbookFactory and then 'direct' program flow accordingly.

Yours

Mark B

PS Sorry again for that 'just open the file and look at the header' bit this morning. I was over-excited about the work we had on today and seriously deprived of tea at that moment. Hope you did not waste any time pursuing that fruitless/pointless direction.

Sparecreative wrote:

Thanks Mark,

That’s exactly what I was after. And, just so I’m clear, with the Workbook
factory, I just check if the Workbook is an instance of either HSSF or XSSF
and process accordingly.

On a related topic it would be great if the specifics for the two event
based readers (for HSSF and XSSF) could be abstracted away so the we could
use one set of methods regardless of the document type. Much like the user
model.

Z.
>
> So, to be clear, all you want to do is identify which files use the OpenXML
> file format and which are binary?
>
> If so, then take a look at the org.apache.poi.ss.usermodel.WorkbookFactory
> (http://poi.eu.apache.org/apidocs/org/apache/poi/ss/usermodel/WorkbookFactory.
> html)
> class. All you need to do is call the static create method pasing an
> InputStream and you will receive back an instance of either the HSSF or
> XSSFWorkbook class depending upon the type of the file. Myself, I have never
> tried using it with files that lack extensions but it ought to work and I
> would certainly suggest giving it a try. Alternatively, you can simply catch
> exceptions; i.e. try to open the file as an HSSFWorkbook, catch the
> exception if the format is not correct and try to open it as an XSSFWorkbook
> then catch and handle the exception thrown if the format is again invalid.
> Finally, you could open an InputStream onto the file and examine the first
> few bytes - I think it is safe to assume that the xml header would be the
> first thing you read from an OpenXML based file.
>
> Yours
>
> Mark B
>
>
> Sparecreative wrote:
>> >
>> > We¹re currently successfully using poi in a webapp to read from uploaded
>> > excel files.
>> >
>> > At present we do a check on the filename extension. If xls we use HSSF and
>> > if xslx we use XSSF and this is working reasonably well.
>> >
>> > The problem we have is that some of our users are uploading files with no
>> > extension. Is there a way to determine the type of excel file that we are
>> > dealing with programmatically?
>> >
>> > I¹d appreciate any pointers as I haven¹t had a lot of luck finding
>> > anything
>> > on the web.
>> >
>> > Z.
>> >
>> >

From forum: POI - User

Re: Determine version of excel file

2009-12-02T08:27:56Z

Thanks for that Chris. I realised what I had said just as we pulled onto site this morning and have spent hours today calling myself all of the names under the sun and bewailing (good word) the fact that I could not get to a PC!! Of course you are correct, the xml is zipped so there will be no xml header at the start of the file. That will teach me to show off before I have had two cups of tea in the morning; my apologies to all.

Yours

Mark B

ChrisLott wrote:

MSB wrote:
> ..
> Finally, you could open an InputStream onto the file and examine the first
> few bytes - I think it is safe to assume that the xml header would be the
> first thing you read from an OpenXML based file.

Goodness, no! :-) An xlsx file (like a docx file and I suppose a pptx
file) is actually a zip archive. Try opening it with winzip or your
favorite zip-file reader and you'll see ("zip -T sheet.xlsx"). Inside
you'll see XML files, each of which should have a nice XML header. I
suppose you could reimplement the magic-number check for a zip file done
by a unix/linux machine's "file" program, maybe that's what POI's
WorkbookFactory does under the covers.

chris...

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@poi.apache.org
For additional commands, e-mail: user-help@poi.apache.org

From forum: POI - User

DO NOT REPLY [Bug 48292] [PATCH] Support of array formulas

2009-12-02T05:55:21Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48292

--- Comment #5 from Petr.Udalau <Petr.Udalau@...> 2009-12-02 05:55:21 UTC ---
Created an attachment (id=24659)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24659)
Tests for array formulas

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48292] [PATCH] Support of array formulas

2009-12-02T05:54:26Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48292

Petr.Udalau <Petr.Udalau@...> changed:

What |Removed |Added
----------------------------------------------------------------------------
Attachment #24623|0 |1
is obsolete| |

--- Comment #4 from Petr.Udalau <Petr.Udalau@...> 2009-12-02 05:54:26 UTC ---
Created an attachment (id=24658)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=24658)
Final patch

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

DO NOT REPLY [Bug 48292] [PATCH] Support of array formulas

2009-12-02T05:53:14Z

https://issues.apache.org/bugzilla/show_bug.cgi?id=48292

--- Comment #3 from Petr.Udalau <Petr.Udalau@...> 2009-12-02 05:53:11 UTC ---
Thanks for reworked patch.
We have made final patch of array formula support and we will waiting for your
review.

Notes:

1) About "if (isPartOfArrayFormulaGroup() && f == null)" in XSSFCell:
I saved Excel document with array formula and when I had opened the xml of
sheet I saw that formula was set only for the first cell of array formula range
and other cells contained only calculated result.

2) About parsing of array formulas:
I changed OperandClassTransformer. But may be It is not solution for this
problem.

3) About CellRangeAddress.valueOf():
This is the case when the array formula is set into one cell. Then the Excel
save the "ref" param of formula like a "A1"(It is only for one-cell array
formula, when we set array formula into several cells "ref" param will contain
range like "A1:B1").

4) About evaluation of array formulas:
There are two types of array formulas:
1.Some formulas (for example "SQRT({1,2;3,4})") can take array arguments and
have to be evaluated in loop with different args like a simple function.
2.Functions that have to be evaluated in special mode when it is in array
formula.

5) About optimized CHOOSE and IF:
Optimization IF/ Choose was made by analyze of first argument and evaluated
only expression, represented by second argument in true case or by third
argument in false case.
In Array Formula Context first argument may be Boolean array which contains:
-only true values
-only false values
-mixed true and false values.
Method checkBooleanContent() provides analyze, which case we have.
Fist two cases can be optimized in the same manner as usual IF/Choose function.
Third case does not allow such optimization and requires evaluating both
expression and only then choosing right value from evaluated arrays according
Boolean array value. It means that we need to restore “non optimized” way to
evaluate such IF/Choose function. Restore of “non optimized” way lead to
ignoring of AttrPtg elements, which provide skipping no needed evaluation
during optimized way.

Due to that IF/Choose function may be nested, decision of optimized/non
optimized is concern only for current IF/Choose function, need to keep in such
function stack and be restored after finishing of evaluating current
IF/Choose.

6) There will suitable to have common interface for Function and
FreeRefFunction.

Do you have any ideas and proposals?

P.S. "test/resources" must be a source folder to run tests. TestArrays - main
test.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

From forum: POI - Dev

HSSFCell short/int

2009-12-02T04:01:52Z

Could it be, the HSSFCell getCell(short cellnum)
method has been rather confusingly deprecated?

Arguably, its laudable to "avoid sign extension",
but as there can only be 256 columns anyway,
I find it really rather misleading.

Regards,
DaveLaw

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@...
For additional commands, e-mail: user-help@...

From forum: POI - User

Re: New hera

2009-12-02T03:37:16Z

I'm very new too. I found a lot of good information on the POI website (
http://poi.eu.apache.org/) and in the examples and unit tests that you get
in the source distribution.

Regards

Mark

2009/12/2 Emeka <emekamicro@...>

> Hello All,
>
> I am pretty new here, and would like to know where I can get good tutorials
> and other relevant materials to get started.
>
> Regards,
> Janus
>

From forum: POI - User