Hello Josh / David, I tried the same conditional formatting code with POI 3.5 and it works properly. However, the generated excel file has a strange behaviour. The condition was that if cell A1 had "A" then B1 cell should have a red fill and strikeout text. The code generates an excel file accordingly and when I set A1 value to "A", B1 cell still has a white fill. But as soon as I double click on the B1 cell, the red background fill shows up.
has anyone else noticed this odd behaviour with conditional formatting using POI 3.5? Also, could anyone point me to the list of patches that would provide me with conditional formatting with POI 3.2? I tried searching bugzilla for all possible classes mentioned by Josh (and a few more) but I was not able to locate any other patches. Is it possible for me to recompile POI 3.6 with JAVA 1.4 compliance setting in Eclipse? On Wed, Dec 30, 2009 at 4:54 PM, Mahesh Ganapathy <[email protected]> wrote: > Can you suggest a hex editor for me? I have never tried using a hex > editor before. Also, I haven't been able to try this out using 3.6 > since I will not be able to install a 1.5 or higher version of JDK in > my dev box. I will try this out at home later this week and will let > you guys know before the end of this year ;) > > Also as an FYI, I noticed an interesting behaviour with DataValidation > API's, the error message box and ignore blank option appears only if > the API's are used in the following order: > > CellReference indirectRef = new CellReference(...); > DVConstraint dvConstraint = DVConstraint.createFormulaListConstraint(formula); > dataValidation = new HSSFDataValidation(new CellRangeAddressList(...), > dvConstraint); > dataValidation.setErrorStyle(HSSFDataValidation.ErrorStyle.STOP); > dataValidation.createErrorBox("Title", "Message"); > m_dataSheet.addValidationData(dataValidation); > dataValidation.setEmptyCellAllowed(false); > > On Wed, Dec 30, 2009 at 4:45 PM, Josh Micich <[email protected]> wrote: >>> The difference is that the code does not set the .needRecalc to true; >>> while excel sets it to true. Any idea which patch would fix this? I >>> rememer seeing a discussion on formula (re)evaluation on excel files >>> generated by POI. Could it possibly be related to this? >> >> It's worth investigating, but not every anomaly you observe with >> BiffViewer corresponds to a bug. Why don't you try setting this bit >> with a hex editor and see if it makes a difference? >> >> BTW - You still haven't said whether you see the same problem with POI 3.6 >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >> > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
