[EMBOSS] Seqret fetch by URL

[Peter Rice]

Steve Taylor wrote:
> 
> Hi,
> 
> I am trying to fetch a sequence using seqret. In my EMBOSS defaults I
> have the following entry set up. I am using EMBOSS 6.0.1 both on Linux
> and Solaris.
> 
> DB NCBI_prot  [  method: url  format: fasta  type: P
>   url:
> "http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?db=protein&qty=1&c_start=1&list_uids=%s&uids=&dopt=fasta&dispmax=5&sendto=t&fr
> 
> om=begin&to=end"
>   comment: "NCBI protein - FASTA seqs only"
> ]
> 
> If I try:
> 
> seqret NCBI_prot:ACA69081
> 
> I get:
> Reads and writes (returns) sequences
> Error: Unable to read sequence 'NCBI_prot:ACA69081'
> Died: seqret terminated: Bad value for '-sequence' and no prompt

They have gone away .... but not very far.

seqret -debug gives the actual HTML returned in the seqret.dbg file

> === File Buffer: Before ajFileBuffStripHtml ===
> * 952daa0 HTTP/1.1 301 Moved permanently
>   951b540 Date: Wed, 24 Jun 2009 11:07:02 GMT
>   951b570 Server: Apache
>   951b358 Location: /sviewer/viewer.fcgi?db=protein&qty=1&c_start=1&list_uids=AC
> A69081&uids=&dopt=fasta&dispmax=5&sendto=t&fr
>   951b388 Content-Type: text/plain
>   952fd10 Vary: Accept-Encoding
>   952fd40 Content-Length: 0
>   952fd70 Connection: close
>   951c2e8 


If you change /entrez to /sviewer in your URL it should work.

Your web browser is able to follow these HTML redirects, EMBOSS is a
little more basic. We gave up on trying to interpret HTML error messages
in the very early days when one site responded to failed requests with a
nice picture of a flower (for those new to the net ... it was EBI :-)

Thanks for letting us know,

Peter