There seem to exist a lot of different options with very dfferent complexities, paired with different word-formats.

If it's a plain old word 2000-2003 file and you already know what your tables look like and you need only some data from within some cells, you could do simply a:

$> abiword --to=rtf myworddocument.doc
[download]

$> perl extract-table-cells.pl myworddocument.rtf
[download]

in the latter (extract-table-cells.pl), you would simply search for:

[pseudo] 
 ...
 # table content part already extracted to $tablecontent
 @cells = $tablecontent =~ /} ([^}]*) }\\cell{/xgs;
 ...
[download]

which might give you the cells in @cells.

But it depends on your problem. Of what scale and purpose is your attempt?

Regards

mwa

What type of Word document? There are many different versions of Word and each version can generate a variety of different document formats. Some formats (rtf for example) are well documented and have some degree of CPAN support. The native Word formats however are much more troublesome to manipulate on a non-Windows system. If you have any control over how the Word documents are generated ask for HTML or RTF.