Hi,
On 13 Aug 2012, at 12:31, 122jxgcn <ywpar...@gmail.com> wrote:
> Hello,
>
> I'm using Solr's ExtractingRequestHandler to let Tika know the name of the
> file when indexing.
> I'm currently sending HTTP request something like
>
> /update/extract?stream.file=#{filepath}&literal.id=#{filepath}&resource.name=#{resource_name}&commit=true
>
> Will setting the resource.name variable let Tika know the name of the file
> so that it can determine Metadata of the file properly?
> (for example resource_name = "file.custom" then in Tika,
> Metadata.RESOURCE_NAME_KEY becomes "file.custom")
> I'm not sure how can I test this so I'm confused.
>
> Thank you.
If you pass the resource name as you are doing in it will be fed into Tika and
used as a hint for Mime Type detection. I assume it was Mime Type detection
you were looking for.
The best way to see what is happening around metadata - and if what you were
trying to pass has made it - is to look at the results of what happens in
Extract Only mode[1].
Cheers,
Dave
[1] http://wiki.apache.org/solr/ExtractingRequestHandler#Extract_Only
