Go to ISI Link

Will do, what field does the ISI link end up in when you import it?

What would be the correct URL for that link?

Don't you need a subscription to access the ISI web of knowledge?

Ron.

As ISI is staunchly anti-open source/access, I'm afraid this link is now outdated and incompatible. You have to buy their new endnote :-)
Yes you do need a subscription to WOS but most big universities will have it. If the user has these kinds of accessions, it is logical to assume they have access to WOS. I think this is a very useful feature.

Here is how I implemented it:

the new link is
http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&ac...

Where X = an integer as accesion record. It happens to be 15 digits long. I tried it with new articles from WOS and it works, but it doesn't work with old articles from my old endnote (<=2006 2008) libraries which are shorted than 15 digits (and padding doesn't help)

Also when importing endnote XML, the URL element is fake: e.g. Go%20to%20ISI%3E%3A/%252F000245099300013 which breaks biblio as it links the title to this fake URL.

So (sorry, I edited it for some more checking to remove instances of the fake URL...)
biblio_theme.inc (it will need some drupal friendly editing...)

157c157,164
<   if ($node->biblio_accession_number)
---
>   if ($node->biblio_accession_number){
>     $accession_num =  check_plain($node->biblio_accession_number);
>     if (preg_match ('/^ISI:/',$accession_num)){
>      $accession_num = str_replace("ISI:","",$accession_num);
>      $accession_link = 'http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&action=retrieve&SrcApp=EndNote&Init=Yes&SrcAuth=ResearchSoft&mode=FullRecord&UT='.$accession_num;
>      $output .= '<h3>'. t("Accession Number") .':</h3><a href="'.$accession_link.'" target="_blank">'. check_plain($node->biblio_accession_number) ."</a>\n";
>     }
>     else {
158a166,167
>     }
>   }
162,165c171,183
<     $attrib = (variable_get('biblio_links_target_new_window', null)) ? array(
<       'target' => '_blank'
<     ) : array();
<     $output .= '<h3>'. t("URL") .':</h3>'. l($node->biblio_url, $node->biblio_url, $attrib) ."\n";
---
>     $biblio_url=$node->biblio_url;
>     # if there is ISI URL, it is a fake link
>     if (preg_match ('/Go\s*to\s*ISI/',$biblio_url)){
>      $biblio_url= str_replace('<Go to ISI>://',"",$biblio_url);
>      $accession_link = 'http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&action=retrieve&SrcApp=EndNote&Init=Yes&SrcAuth=ResearchSoft&mode=FullRecord&UT='.$biblio_url;
>      $output .= '<h3>'. t("URL") .':</h3><a href="'.$accession_link.'" target="_blank">'. check_plain($node->biblio_accession_number) ."</a>\n";
>     }
>     else {
>      $attrib = (variable_get('biblio_links_target_new_window', null)) ? array(
>        'target' => '_blank'
>      ) : array();
>      $output .= '<h3>'. t("URL") .':</h3>'. l($node->biblio_url, $node->biblio_url, $attrib) ."\n";
>     }
211,212c229,233
<     $attrib = (variable_get('biblio_links_target_new_window', null)) ? array('target' => '_blank') : array();
<     $node->biblio_url = l($node->biblio_url, $node->biblio_url, $attrib);
---
>     # if there is ISI URL, it is a fake link
>     if (!preg_match ('/Go\s*to\s*ISI/',$node->biblio_url)){
>      $attrib = (variable_get('biblio_links_target_new_window', null)) ? array('target' => '_blank') : array();
>      $node->biblio_url = l($node->biblio_url, $node->biblio_url, $attrib);
>     }

One thing I have not found out (and is a serious bug) is that when listing all the articles (/biblio page) the title links out to the URL field. In the case of Endnote /WOS integration, the URL field is this fake URL so should be disabled it. In my site, I disabled the option "Hyperlink titles using supplied URL if available" in order to avoid this (but still, it should be checked to see if it is a valid URL before linking out...)

thanks
a

Hi Alexie,

I've implemented this, but in a slightly different way. Rather than fix the links in the theme file, I do it loading them from the database, thus the URL will be correct regardless of where it is used. I put a new function in the biblio.module file called biblio_fix_isi_links() and a new section related to this on the /admin/settings/biblio page.

Try it out (it's in the -dev version) and let me know if it works for you.

Ron.

Hi Ron

I've just tried the fix in the -dev. Sadly there was smth wrong with the default URL (I copy pasted mine and it works but I can't remember what the default was). The one that works is shown below.

Also smth cosmetic:
Page before activating it:
Accession Number:
ISI:000245099300013
URL:
://000245099300013

After
Accession Number:
http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&ac...
URL:
http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&ac...

we probably don't want the URL to be written...

Can we also have target=_blank ?

thanks for a fantastic module... Now I just have to figure out how to make openurl work... (i'll submit an issue if i see it's not an error from my end...)

many thanks
alexie

Hi Alexie,

Strange since the default URL is exactly what you have above, so I don't know why it didn't work. You can see the default right near the end of the biblio.module file on line 2214 (8 lines from the end).

Rather that having the whole URL visible, what about leaving the text there and making it a link to the correct URL, probably what I should have done in the first place.

Ron.

Hmm, maybe something else was wrong. But do leave the default, as this thread started with people not knowing what the URL is.

speaking of url
I added a OpenURL bug report...

Hi,

I have ISILink API. I just wanted to know how much will it cost me for someone ti integrate it in my website.

Sorry for asking you out of box question. Hope you will help me

Thanks
Cris

Hi,

I am new to ISILink API. I have ISILink API, can you tell me how much will it cost me for someone to integrate the API in my website.

Please help.

Thanks
Cris