You are viewing a plain text version of this content. The canonical link for it is here.
Posted to agent@nutch.apache.org by "SKUHRA, Milan" <Mi...@orange.sk> on 2006/07/20 09:24:04 UTC

How can I influence a Content-Type checking?

Hello,

 

I am trying use Nutch to look up at specify URLs like this are:

    http://server.domain/appname/get?id=34&view=content

    http://server.domain/appname/get?id=35&view=content

    http://server.domain/appname/get?id=36&view=content

 

(So I use "-depth 1" option for running crawl.)

Some sites are HTML pages, but some return files.

After running the crawl I found at the log file this record:

 

060719 160036 fetch okay, but can't parse
http://server.domain/appname/get?id=36&view=content, reason:
failed(2,203): Content-Type not text/html: application/msword

 

It looks like crawl infers Content-Type from URL and than compares it
with received Content-Type.

Is possible to use received Content-Type without checking Content-Type
implied from URL?

How can I resolve this problem?

 

Tank you for reply.

Milan Skuhra

 


Re: How can I influence a Content-Type checking?

Posted by Jayant Kumar Gandhi <ja...@gmail.com>.
Have you enabled the msword parse plugin in nutch-site.xml . You will have
to enable that plugin for msword parsing to work.

Cheers,
Jayant

On 7/20/06, SKUHRA, Milan <Mi...@orange.sk> wrote:
>
> Hello,
>
>
>
> I am trying use Nutch to look up at specify URLs like this are:
>
>     http://server.domain/appname/get?id=34&view=content
>
>     http://server.domain/appname/get?id=35&view=content
>
>     http://server.domain/appname/get?id=36&view=content
>
>
>
> (So I use "-depth 1" option for running crawl.)
>
> Some sites are HTML pages, but some return files.
>
> After running the crawl I found at the log file this record:
>
>
>
> 060719 160036 fetch okay, but can't parse
> http://server.domain/appname/get?id=36&view=content, reason:
> failed(2,203): Content-Type not text/html: application/msword
>
>
>
> It looks like crawl infers Content-Type from URL and than compares it
> with received Content-Type.
>
> Is possible to use received Content-Type without checking Content-Type
> implied from URL?
>
> How can I resolve this problem?
>
>
>
> Tank you for reply.
>
> Milan Skuhra
>
>
>
>
>


-- 
www.jkg.in | http://www.jkg.in/contact-me/
Jayant Kr. Gandhi
M.Tech. Computer Tech. Class of 2007,
IIT Delhi