Re: [libextractor] Using libextractor from Java on Windows

[Christian Grothoff]

On Tuesday 27 September 2005 02:50, Lukas Mathis wrote:
> Hi,
>
> I'm currently attempting to read some metadata from .doc files written by
> Microsoft Word. My application runs on Java on a Windows system. Frankly, I
> don't know a whole lot about Windows, as I normally use *nix systems. I've
> downloaded the Windows binary, but running it from the DOS shell gives me a
> list of similar errors, starting with:
>
>   Loading 'libextractor.exiv2' plugin failed: can't open the module

Hmm. the "." in the error message looks odd (should be "_", at least on Unix) 
-- but I don't know anything about these things for Win32 systems.  Maybe 
Nils can help.  In general, yes, the first step should be to get the 
"extract" binary to work in a shell on Windows.

> Basically, I guess I have two questions:
>
> 1) What do I have to do to get libextractor running on Windows?
> 2) How do I access libextractor from Java under Windows?

For Java, download the Extractor-java source 
(http://gnunet.org/libextractor/download/libextractor-java-0.5.1.tar.gz is 
the latest; don't worry about the version-mismatch).  This contains a shared 
C library and some Java source code.  Integrate the Java code with your app, 
and compile the C code using the configure-script into a DLL which should be 
linked against the main libextractor library (./configure 
--with-extractor=PATH --prefix=PFX) and then installed in a location PFX that 
is found by the OS and hence by the JVM with "System.loadLibrary()".  At 
least, this is how it works on Linux -- again, I don't run Windows, so I 
can't tell you more than these 'theoretical' steps.

> Reading through the list archive, I haven't found anyone with this
> particular problem.

Well, sorry I can't be of more help, but maybe someone else here can :-).  
Also, if you find a solution yourself later, it would be great if you could 
post some details here so that others can learn from it.

Happy hacking

Christian