Page 1 of 1

OCLC PsycInfo import

Posted: Thu Aug 25, 2005 4:28 pm
by jzacks
Hi Jon et al.,

A couple questions about importing from OCLC PsycInfo:

* The OCLC PsycInfo filter distributed with Bookends 8 clips off most of the abstracts. Telling the filter that fields end with Return followed by any valid tag appears to fix the problem. Any reason not to keep this change?

* Using this filter, journal names usually wind up with " Vol" appended to them, because the output format's source field goes "Source: Journal Name Vol XX(X)" [etc]. Is there any way in the source parsing page to tell it to collect until a particular string of characters (in this case, " Vol")? Or is there any other way around this?

* Finally, has anyone tried setting up an import filter to talk to OCLC's z39.50 server. In EndNote, I find I'm able to connect from on campus (where we have a site license) by using EndNote's filter and just leaving the user name and password fields blank. I tried setting up a similar filter in Bookends, but it doesn't seem to allow one to leave the user name field blank. Setting up the filter to not require a password gives an Access Denied message. Any thoughts about how to get around this?

Thanks,

Jeff Zacks

Re: OCLC PsycInfo import

Posted: Thu Aug 25, 2005 5:40 pm
by Jon
jzacks wrote: * The OCLC PsycInfo filter distributed with Bookends 8 clips off most of the abstracts. Telling the filter that fields end with Return followed by any valid tag appears to fix the problem. Any reason not to keep this change?
Well, be careful. "Any valid tag" is dangerous because you must make sure that you account for any possible tag the service provides (and ignore most of them). If you don't you can get LOTS of data in one field.
* Using this filter, journal names usually wind up with " Vol" appended to them, because the output format's source field goes "Source: Journal Name Vol XX(X)" [etc]. Is there any way in the source parsing page to tell it to collect until a particular string of characters (in this case, " Vol")? Or is there any other way around this?
No, the source parser looks for characters, not words.
* Finally, has anyone tried setting up an import filter to talk to OCLC's z39.50 server. In EndNote, I find I'm able to connect from on campus (where we have a site license) by using EndNote's filter and just leaving the user name and password fields blank. I tried setting up a similar filter in Bookends, but it doesn't seem to allow one to leave the user name field blank. Setting up the filter to not require a password gives an Access Denied message. Any thoughts about how to get around this?
Sorry, I can't help here. You might contact your library to see if there is a userid/password combo you can use.

Jon
Sonny Software