Sounds reasonable. Can you offer any insights into the shared string
table? It looks like the shared string ID in a sheet(n).xml file is
simply the index of the string in the table. That suggests to me a
simple Java vector might be a suitable implementation with not only low
memory use but also constant lookup time; don't even need a map.
I was hoping the SharedStringTable type might be an interface but no luck.
chris...
Nick Burch wrote:
On Wed, 6 May 2009, Chris Lott wrote:
So my next question is, is there something especially memory hungry in
the shared strings table builder? Is there anything I can do here?
Would you like me to write a bug?
The SharedStringsTable helper on XSSFReader is a full XMLBeans powered
class. This made it much quicker to write, but does mean it'll use more
memory. Also, because it's a read-write class (used by XSSFWorkbook
too), it keeps many more references than are needed for just reading in
a file
If you have the time, I'd suggest writing a purely sax, read only shared
strings table helper for use with XSSFReader. One of those ought to be
able to process the strings in a much lower memory footprint, and ought
to be fairly quick to write as the shared strings part is such a simple
format
Nick
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
