There is a problem when using PDFPlugin with -convert_to html and
-complex options with collection containing duplicate file names. The
image extracted from all file of the same name is the same when
displayed. I think this is probably due to some caching or
no-overwriting in conversion process.
Affect Greenstone 2.83 on Windows.
Step to reproduce:
1. Prepare 3 image-based PDF documents, put them in 3 different folders
but name the 3 files the same, say doc.pdf.
2. Add those 3 folders to a Greenstone collection.
3. Add PDFPlugin, choose option -covert-to html, -complex and
4. Build the collection
5. Browse the 3 documents in Greenstone, all of them will show up the
same. That is the page 1 image of all documents will be the page 1 of
one of the document.
There seems to be no problem with -convert_to pagedimg, though.
Asian Institute of Technology
PS. Sorry for reporting bugs here. I could not login to Trac. The login
registered at http://www.greenstone.org/developers-register doesn't work