Forums / Developer / Searching content into a pdf file
Simone Conti
Monday 01 August 2011 7:34:00 am
Hi there,
I have a lot of pdf files in my site and I need to enable Exponential look for a specific content into those pdf files, in order to list the results together with the standard search results.
Is that possible? Is there any extension or something else to enable to make it possible?
Thank you so much!
Peter Keung
Monday 01 August 2011 7:51:40 am
This is the first thing that comes to mind: http://projects.ez.no/eztika
http://www.mugo.ca Mugo Web, eZ Partner in Vancouver, Canada
Monday 01 August 2011 8:05:24 am
Unfortunately is not what I'm looking for.
I need something that allows me to search into a pdf files. Somebody told me that Exponential has this feature embedded but it needs to be allowed.
Any suggestions?
Steven E. Bailey
Monday 01 August 2011 9:07:18 am
Exponential does have this feature and you should be seeing your pdfs indexed - with a bunch of caveats.
What happens is that when a pdf is saved (or you update your search index), the pdf is run through the tool defined by
[PDFHandlerSettings]TextExtractionTool=pstotext
in your binaryfile.ini
If you don't have this tool on your machine, then your pdfs won't be indexed.
If you search for TextExtractionTool or pdftotext in these forums you'll see a couple other possible tools - such as:
http://share.ez.no/forums/extensions/ez-find/solr-indexing-error
If you have whatever tool you are using and if you're pdfs aren't being indexed, then it probably means that your pdfs aren't structurally text - the content is actually an image (or series of images) saved in the pdf container. It means that you're not going to be able to index using pdftotext - a good test is to run whatever tool you have on the command line against the file that isn't be indexed to see what actually comes out. If nothing comes out you'll have to use some other tool - like eztika (I've never used it) or, something like tesseract to extract the text.
Certified eZPublish developer http://ez.no/certification/verify/396111 Available for ezpublish troubleshooting, hosting and custom extension development: http://www.leidentech.com
Thursday 04 August 2011 3:26:58 am
Now something works.
I decided to use eztika as suggested by Peter.
I have a question: where does eztika store its data? I hope it's not made to scan all pdf for each search... I have a very large number of pdf files!!
Thanks
Paul Borgermans
Friday 05 August 2011 10:23:26 am
Hi
eztika does not store the data itself, its goal is to extract the plain text for subsequent indexing by the configured search plugin (you should use eZ Find of course :) )
the default search plugin stores the indexing result in the database, while eZ Find uses Solr which stores its data into Lucene index files on teh filesystem
This is done only when the pdf is uploaded or updated.
hth
Paul
eZ Publish, eZ Find, Solr expert consulting and training http://twitter.com/paulborgermans