PDF files handling

[Ken Ray]

On Wed, 04 Jul 2007 18:39:01 -0400, Bryan McCormick wrote:

> Can Rev do such a thing (easily) or am I destined to pick up the 
> light version of Google's indexing server?

No, Rev an do it so long as it's consistent. If the files aren't that 
large, read them from disk into a variable as if it was a text file. 
The title of the document may be in the text. For example, here's the 
first part of one PDF (version 1.2) for the AportisDoc Reader:

%PDF-1.2
%‚„œ”
1 0 obj
<<
/CreationDate (D:19980420172555)
/Producer (Acrobat Distiller 3.01 for Power Macintosh)
/Keywords ()
/Subject (AprotisDoc Reader)
/Creator (Adobe PageMaker 6.52)
/Title (AportisDoc Reader)
/Author (Aportis Technologies Corp.)
>>
endobj

I tested a bunch of PDFs, and they all had a /Title "property" that you 
can search for, although some were plain (like the one above), and some 
were encrypted, like the Read Me from QuickBooks:

/CreationDate (`Êã:°-òD˚ÑÙ%æ≈“_9]¶)
/Creator (s≥Àn)
/ModDate (`Êã:°-òD˚ÑÒ/ø¬“_9]¶)
/Producer (iΩ⁄*fiKâ-¯Ö
Ú3≤÷Æ\\lQ°Œíi∆ÈF„B†)
/Title (u©–i˙Z∆C∫ïÒ-Ω÷ô O> J‚æÑJ‰ß„)

That might give you a start if they're all created in the same way...


Ken Ray
Sons of Thunder Software, Inc.
Email: kray at sonsothunder.com
Web Site: http://www.sonsothunder.com/