Yes. Indexing a PDF&other types with '/extract' means that Solr finds words in the document and indexes those in a field 'content'. It does not save the binary contents of the file. You could make a request handler that fetches one document and generates a redirect to the link.
On Thu, Sep 2, 2010 at 7:35 AM, Matteo Moci <mox...@libero.it> wrote: > Thank you for the suggestions, > I just completed the tutorial at http://lucene.apache.org/solr/tutorial.html > and i understood that in the GET parameters I can choose wt=standard (and > obtain an xml structure in the results), > wt=json or wt=php. > > All of them display the results inline, in the sense that they are embedded > and entirely included in the response. > > If I submit pdfs (i think it is also for docs and CSVs) files to solr, I > will get in the results something like this in json: > > [some part of response omitted] > response":{"numFound":1,"start":0,"maxScore":0.34002018,"docs":[ > { > "last_modified":"2010-08-05T14:07:24Z", > "id":"doc1", > "content_type":["application/pdf"], > "score":0.34002018}] > }} > > ( example taken from http://wiki.apache.org/solr/ExtractingRequestHandler ) > > that shows no content at all. > The only way I have to retrieve and download the pdf file is to use the > id=doc1 to access some repository (even a database table ) > that can provide me the content starting from the id. > > Does this look like a common practice? > > Thank you > > > > > > Il 02/09/10 08:47, Lance Norskog ha scritto: >> >> Solr can return the list of results in JSON or php format, so that you >> UI can allow a download. >> >> You can write a UI in the Velocity toolkit- it's pretty easy. >> >> On Wed, Sep 1, 2010 at 8:24 AM, Matteo Moci<mox...@libero.it> wrote: >>> >>> Hello to All, >>> I am a newbie with Solr, and I am trying to understand if I can use it >>> form >>> my purpose, >>> and I was wondering how Solr lists the result documents: do they appear >>> as >>> "downloadable files", >>> just like http://solr.machine.com/path/file.doc, or do I need develop >>> another layer to take care of downloading? >>> Even a link to the docs might work... >>> >>> Thank you, >>> Matteo >>> >>> >> >> > -- Lance Norskog goks...@gmail.com
