Unlike records, whose header contains the record size, a PTG's size has to be
inferred from what kind of a ptg it is. therefore, if we do not understand a
ptg, we cannot process the subsequent stream. The commented out code is
therefore completely incorrect, and will result in a scrambled output in some
way.
the best way to solve this is therefore to implement the ptg that is missing.
If it is the percentage ptg, should be simple to implement.
Regards
-
Avik
On Tuesday 28 March 2006 11:02, Daniel Noll wrote:
> I was getting an unexpected UnsupportedOperationException thrown when
> creating a workbook from a certain document.
>
> Digging through a debugger I eventually tracked it down to being thrown
> from Ptg.java, line 330:
>
> //retval = new UnknownPtg();
> throw new java.lang.UnsupportedOperationException(
> Integer.toHexString(( int ) id) + " (" + ( int ) id + ")");
>
> What I want to know is why that code which is commented out wasn't left
> in. It seems bad to throw an exception just because some kind of
> formula isn't understood, so I'm tempted to change it back, but is there
> a risk if I revert that code back to the original?
>
> I suppose the other alternative is implementing a Ptg class for ID 0x14.
> This appears to be the formula element used when you have a percentage
> in a cell formula (e.g. =D4 * 9%).
>
> Daniel
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
Mailing List: http://jakarta.apache.org/site/mail2.html#poi
The Apache Jakarta Poi Project: http://jakarta.apache.org/poi/
