RE: pdf to html errors

From Marie-Jose Quintard
DateMon, 13 Jan 2003 15:12:06 +1100
Subject RE: pdf to html errors

Greenstone 2.38 on WIn2000 PRO.


When I started to use Greenstone, I had some problems with pdf file that could not be included.
M.Jared Potter gave the idea to include a cover page which indeed works fine
and I have been able in doing so to include most of the .pdf files I had.
Only one up to now can't be included(11Mb may the size is one of the reason?).

Because most of the HTML file generated were unusable(too much garbage), I put a
CLVlist format VList "<td>...</td>" in the collect.cfg (thanks to John R. McPherson)
Now the list of available books give only the PDF file(no more the HTML version) which is neat.

But now, I have another problem:
when I do a search, I get a list of the books which contain the word(s) searched
but when I open the pdf document(to see where these words are located in the document), it is the "image"
of the pdf document which is opened, not the original pdf I included.
The "image" of the pdf file=> I can't use the "find"  which is available in Acrobat to look for the occurrence of the words

within the pdf file.
The original document I scanned in pdf is not a simple "image" but already a file in which I can search for words(I captured all the pages)

my question:
how can I display as a result of a search my original pdf file ?

Please let me know if something is not clear in this message,
Thanks in advance,

Marie-José Quintard (ICT)
Secretariat of the Pacific Community
SPC Private Mail Bag
Suva - Fiji Islands
Tel: (679) 370 733
Fax: (679) 370 021
Email: MarieJoQ@spc.int
http://www.spc.int





-----Original Message-----
From: Jared Potter [mailto:jpotter@mercycorps.org]
Sent: Wednesday, October 30, 2002 11:36 AM
To: greenstone@tripath.colosys.net
Subject: RE: pdf to html errors


An easy work around to the Images of text that John has described is to insert a coverpage that has a title and perhaps the author.  That give greenstone enough text to at least include the file, even if it won't do a full text index of it.  To do that, you will need a copy of adobe acrobat 5, or some other similar software.

-jared

Jared Potter
Mercy Corps
Portland, Oregon

-----Original Message-----
From: John R. McPherson [mailto:jrm21@cs.waikato.ac.nz]
Sent: Tuesday, October 29, 2002 12:52 PM
To: greenstone@tripath.colosys.net
Subject: Re: pdf to html errors


Bhakti P Beke wrote:
>
> Hi,
>
> I have also come across similar problem. I am working on GSDL 2.38
> version. Though I have submitted a query from Greenstone Support
> website I am still waiting for the rely.
>
> Bhakti

> On Tue, 29 Oct 2002 Marie-Jose Quintard wrote :
> >Goo dafternoon,
> >
> >I have created a new collection and have imported a few books
> >which are
> >available in pdf format (scanner used = HP9100C digital sender)
> >When I add these books to the new collection,  a lot of error are
> >generated
> >when the pdftohtml is executed
> >and most of these books can't be converted to HTML.
> >
> >Any solutions, suggestions to solve this problem would be
> >welcome.
> >Thanks,

Hi,
it would help if you told us things such as which operating system you
are using, and what kind of error messages you are getting. Otherwise
it is too difficult for us to know what is going wrong.

Having said that, we are aware of some problems when converting pdf
files on some versions of microsoft windows, although I'm not sure if
we have yet come up with a 100% fool-proof work-around yet. Windows NT,
2000 and XP seem to not see these problems as often as windows 95 and 98
do.

Also remember that some pdf files don't actually contain text, but contain
images of text, and these pdf files cannot have the text easily extracted
from them.

John McPherson