#!/bin/tcsh -f
#(ie run the tshell on this but don't read the .cshrc or .tcshrc)

echo version = 1.11 of topaper 2006 Aug 02
# 2006 Aug 02, 1.11: duck sound to klaxon
# 2006 Jan 12, 1.10: clean up
# 2006 Jan 12, 1.09: clean up and fix; key for externa link: dx.doi.org
# 2005 Dec 14, 1.08: break fused urls
# 2005 Dec 13, 1.07: fused urls with:
#                    pms 257 286 1982 martin
# 2005 Apr  7, 1.06: crash on "pms vergani gadsby 876" at openline
# 2005 Mar  5, 1.05: give line with PMID if failure
# 2004 May 11, 1.04: identification sounds, duck, wolf
# 2004 Apr 26, 1.03: error: pubmedsearch "Jarzynski C": 'Word too long'
# 2004 Apr  9, 1.01: functional!!!!
# 2004 Apr  9, 1.00: origin 

echo 'Move the browser to a paper based on a pubmed url\!'

if ($#argv != 1) then
  echo 'usage: topaper [pubmed web page]'
  exit
else
  echo pubmed web page given is: "$1"
endif

set baseurl = 'http://www.ncbi.nlm.nih.gov'

set tmp1 = /tmp/`whoami`1.topaper.html
set tmp2 = /tmp/`whoami`2.topaper.html

wget -O $tmp1 "$1"
# cat $tmp1

# The key is the way to identify the relevant URLs on the page!
set key = "dx.doi.org"
cat $tmp1 |\
sed "s/url=/@@/" |\
sed 's/"/@/g' |\
tr '@' '\n' |\
grep "$key" |\
cat > $tmp2
echo ---- $key links:
cat $tmp2

set count = `cat $tmp2 |wc -l` 

if ("$count" > 1) then
   echo "***********************************************"
   echo "* The wolfchorus means that there are multiple links"
#   echo "* $count Multiple links\!"
   sound wolfchorus > /dev/null &
   cat $tmp2 | number
   echo "***********************************************"
endif

if ("$count" == '0') then
   echo "No $key line found\!"
   echo This probably means that there are no links
   sound klaxon > /dev/null &
   exit
else
   sound towerclock > /dev/null &
endif

set url = `head -1 $tmp2`

echo --------------------------------- the next url is:
echo "$url"
echo --------------------------------------------------

# fused urls detection code
# set urls = `echo "$url"|sed 's,http://,@http:,g'|tr '@' '\n'|grep 'http:'|wc -l`
# if ("$urls" != 1) then
#   echo "urls = $urls"
#   echo "url = $url"
#   echo "baseurl = $baseurl"
#   # echo "$url"|sed 's,http://,@,g'|tr '@' '\n'
#   exit
# endif

mozillaurl "$url"

echo "********************************************************************"
exit

# ok, try to find a PDF and move to that!!

set tmp3 = /tmp/`whoami`2.topaper.html

echo the tmp3 file is: "$tmp3"
echo "$tmp3"

echo the url we are trying is:
echo "$url"

echo --- FINAL WGET DOES NOT WORK YET:
wget -O "$tmp3" "$url"
echo ---

exit
********************************************************************************

From toms Fri Apr  9 19:47:58 2004
To: custserv@nlm.nih.gov
Subject: unknown protocol
Content-Length: 1342

Hi:

A pubmed entry is at:

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14755570

The link on the pubmed entry is:

http://www.ncbi.nlm.nih.gov//entrez/utils/lofref.fcgi?PrId=3058&uid=14755570&db=pubmed&url=http://dx.doi.org/10.1002/bip.10525

I can paste this in to my Mozilla browser and it works.

However, If I use wget, it gives me an error:

--19:45:22--  http://www.ncbi.nlm.nih.gov:80/entrez/utils/lofref.fcgi?PrId=3058&uid=14755570&db=pubmed&url=http%3A/dx.doi.org/10.1002/bip.10525
           => `/tmp/toms2.topaper.html'
Connecting to www.ncbi.nlm.nih.gov:80... connected!
HTTP request sent, fetching headers... done.
Location: http%3A/dx.doi.org/10.1002/bip.10525 [following]
http%3A/dx.doi.org/10.1002/bip.10525: Unknown/unsupported protocol.

It should go (through a hop) to

http://www3.interscience.wiley.com/cgi-bin/abstract/106565357/ABSTRACT

In theory the Elink should do this:

http://eutils.ncbi.nlm.nih.gov/entrez/query/static/elink_help.html

But most of the examples fail!

How do I do this?

Tom

  Dr. Thomas D. Schneider
  National Cancer Institute
  Laboratory of Experimental and Computational Biology
  Frederick, Maryland  21702-1201
  toms@ncifcrf.gov
  permanent email: toms@alum.mit.edu (use only if first address fails)
  http://www.lecb.ncifcrf.gov/~toms/