Hi,
There’s a parseSheet method in XSSFWorkbook (the relevant class if you’d like
to read/update xlsx). You could possibly subclass XSSFWorkbook and then
override parseSheet to not call onDocumentRead() for the sheets you are not
interested in based on their names. I haven’t tried it myself, and the
community possibly have a much neater solution, but you might want to spend a
few minutes to see if you can get it to work.
See code for the (default) implementation of parseSheet:
public void parseSheet(Map<String, XSSFSheet> shIdMap, CTSheet ctSheet)
{
XSSFSheet sh = shIdMap.get(ctSheet.getId());
if(sh == null) {
logger.log(POILogger.WARN, "Sheet with name " +
ctSheet.getName() + " and r:id " + ctSheet.getId()+ " was defined, but didn't
exist in package, skipping");
return;
}
sh.sheet = ctSheet;
sh.onDocumentRead();
sheets.add(sh);
}
Best of luck,
Markus
> On 07 Jul 2016, at 10:41, Marek Slama <msl...@email.cz> wrote:
>
> Hi,
>
> I have workbook with ~ 10 sheets. Some of them are quite big. I need to
> update just one. Is it possible to load given sheet on request ie. not to
> load whole workbook model? (provided I have no links/dependency between
> sheets). Is it possible in principle? I want to ask before I try to dig into
> code. Then on write I would save only sheets which are loaded. (main reason
> for this is memoru usage and also performance as creating complete model
> takes long).
>
> I read some questions/discussions regarding this but it was lazy loading on
> row level. I think on sheet level it might be easier as sheets are separated
> in xlsx file.
>
> Thanks
>
> Marek
>
