Hi,<br><br>We are developing a product and interested to buy
aspose word/presentation for java as our
support for analyzing those documents. I downloaded the trial
version and did a comparison between Aspose & Apache POI(Just
extracting the plain text from those documents), but the result seemed
not favorable for your product. Neither the usage of processing time
nor the robust for different .doc version.
Aspose is
about 2-4 times slower than apache POI at the processing speed on
average, and it doesn’t support word6 & word95(POI doesn’t support, either) As
a commercial product, I believe your product should be better than
those open source libraries. I’d like to hear from you on our findings.
I think it boils down to the simple question of comparing apples with apples.
Most likely Aspose.Words is slower because it reads and parses a lot more info from DOC files than Apache POI. Aspose.Words does so much more than Apache POI and that surely has its price in the form of processing and memory overheads.
Why don't you compare Apache POI to Microsoft Word for example? Microsoft Word can also be used to extract plain text from DOC files. Yet it does a lot more than Apache POI so the speed could be slower and that will be reasonable to expect.
The thing is, if you are after the fastest plain text only extraction from DOC files, then I might not be able to convince you to use Aspose.Words if you have found a suitable library.
But say if you needed to at least extract text from DOC, DOCX, WordML, HTML (and RTF, ODT soon to be available in Aspose.Words for Java) all from one library all using two lines of code - then we would not have this discussion probably.
Having said all that, we will be happy to look at your test documents and test results. Maybe we can improve. Please feel free to attache the docs here.