POI-HWPF - A Quick Guide
HWPF is still in early development. It is in the scratchpad section of the SVN. You will need to ensure you either have a recent SVN checkout, or a recent SVN nightly build (including the scratchpad jar!)
Basic Text Extraction
For basic text extraction, make use of org.apache.poi.hwpf.extractor.WordExtractor. It accepts an input stream or a HWPFDocument. The getText() method can be used to get the text from all the paragraphs, or getParagraphText() can be used to fetch the text from each paragraph in turn. The other option is getTextFromPieces(), which is very fast, but tends to return things that aren't text from the page. YMMV.
Specific Text Extraction
To get specific bits of text, first create a org.apache.poi.hwpf.HWPFDocument. Fetch the range with getRange(), then get paragraphs from that. You can then get text and other properties.
Headers and Footers
To get at the headers and footers of a word document, first create a org.apache.poi.hwpf.HWPFDocument. Next, you need to create a org.apache.poi.hwpf.usermodel.HeaderStores, passing it your HWPFDocument. Finally, the HeaderStores gives you access to the headers and footers, including first / even / odd page ones if defined in your document. Additionally, HeaderStores provides a method for removing any macros in the text, which is helpful as many headers and footers do end up with macros in them.
It is possible to change the text via insertBefore() and insertAfter() on a Range object (either a Range, Paragraph or CharacterRun). It is also possible to delete a Range. This code will work in many, but not all cases, and patches to improve it are gratefully received!
For now, the best source of additional examples is in the unit tests. Browse the HWPF unit tests.
by Nick Burch