If we replaced that, and stored the document as Java objects only, we could then conceivably write the document as either XLS or XLSX, or whatever format comes down the pike simply by selecting a write format. Someone would have to create the writers. Expanding on that, we could also read whatever format we wanted by creating an appropriate reader. However that would lose the ability for a user to just use the CT classes to support a bit of the standard that hasn't yet been surfaced by POI. But, it would make swapping out XMLBeans for something else easier.
Still thousands of hours of work. -----Original Message----- From: Javen O'Neal [mailto:[email protected]] Sent: Friday, December 02, 2016 2:21 PM To: POI Users List <[email protected]> Subject: Re: Too much memory is used when reading a xlsx-file whose size is just 7.3M Those numbers sound about right. I'm used to 4 MB balloning to 1 GB. We could significantly reduce memory consumption if we didn't maintain the XML DOM in memory, but replacing that requires thousands of hours of work. On Dec 2, 2016 09:44, "Blake Watson" <[email protected]> wrote: > Well, this is interesting: I've also seen a ballooning of RAM usage > (though I wouldn't call it "violent" =P) when loading a spreadsheet > and I didn't know about the input-stream vs. file distinction. > Although, come to think of it, I'm pulling out of a database so that might > not help me. > > The excess usage is only temporary, though, right? My small (<100K) > spreadsheets seem to take about 50MB, and my large ones (5MB) seem to > take about 500MB. > > On Fri, Dec 2, 2016 at 7:54 AM, Gordon Pettey <[email protected]> > wrote: > > > On Fri, Dec 2, 2016 at 2:41 AM, Javen O'Neal <[email protected]> > wrote: > > > > > CPU pegged due the the Java garbage collector trying to clean up > > > unused memory. The out of memory error is raised when the garbage > > > collector > > cannot > > > free the necessary memory for new objects to be allocated. > > > > > > Have you read the FAQ yet? https://urldefense.proofpoint. > > com/v2/url?u=https-3A__poi.apache.org_faq.html-23faq- > 2DN10165&d=DgIFaQ&c= > > dmLomitc30UP5j2qU8E1rg&r=p42pHJHEwFZOHtVFHKJUdL2fYbroN3 > > 3stXXb3Psthjw&m=K-mZqLKcau3v-2mO1e7Es8pmREXI9_WkDMs3UxiVEsE&s=_ > > gk9WyMdApng32mJ3IDsTgjct7dpw_BenPZP-wdE0fA&e= > > > > > > On Dec 2, 2016 00:11, "[email protected]" <[email protected]> wrote: > > > > > > > Hi! > > > > > > > > I am trying to read a xlsx-file whose size is 7.3M with POI > > > > version > > 3.15, > > > > here is the code： > > > > > > > > > > > > *InputStream is = new FileInputStream(new File(path));//line > 1Workbook > > > book = new XSSFWorkbook(in);// line 2* > > > > When running to line 2，the memory of javaw rised violently from > > > > 700M to 1200M，cpu from 10% to 90%，the following is the > > > > screenshot of memory dump: > > > > > > > > > > > See https://urldefense.proofpoint.com/v2/url?u=http-3A__poi. > > apache.org_spreadsheet_quick-2Dguide.html-23Files-2Bvs- > > 2BInputStreams&d=DgIFaQ&c=dmLomitc30UP5j2qU8E1rg&r= > > p42pHJHEwFZOHtVFHKJUdL2fYbroN33stXXb3Psthjw&m=K-mZqLKcau3v- > > 2mO1e7Es8pmREXI9_WkDMs3UxiVEsE&s=Q2dr3Xc2iSpx8HHAAj0kQkGXZYpdBK > > 9yZFifR-mQo64&e= , > > also. > > > > > > -- > > *Blake Watson* > > *PNMAC* > Application Development Manager > 5898 Condor Drive > Moorpark, CA 93021 > (805) 330.4911 x7742 > [email protected] <[email protected]> > www.PennyMacUSA.com <http://www.pennymacusa.com/> >
