On Fri, Oct 7, 2011 at 5:52 AM, Ian Bays <ian.b...@ptfs-europe.com> wrote: > Hi Ben, Brian et al, > > We came across this when trying to show images from Amazon. We know from > experience that Amazon mostly only responds to 10-digit ISBNs as the Amazon > ID matched the 10-digit ISBN. Amazon's line has always been that they > provide images keyed on their Amazon ID. > > When loading data there are 10-digit and 13-digit ISBNs often in no > particular order. So we found the script in Evergreen that forms the URL > for the image from Amazon and test if the ISBN is 13-digits, and if so > convert it using Business::ISBN perl module to be a 10-digit ISBN. So no > messing with the data in bib at all. Seems to work well. > > On our system the script to change is: > /openils/lib/perl5/OpenILS/WWW/AddedContent/Amazon.pm > > We're happy to share - it's just a few lines. Maybe this should be on the > DEV list?
Please do share: it's an open-source project and it sounds like a good incremental improvement to Amazon added content. The contribution process is at http://evergreen-ils.org/dokuwiki/doku.php?id=contributing - it would be great to see your name / company added to the contributor list! > Also we noticed that if your ISBNs have hyphens in them then the ISBN only > comes through as the digits up to the first hyphen (in our case usually > "0"). We do get lots of cataloguers who believe ISBNs should have hyphens. > So we are weighing up removing hyphens in data conversion versus finding the > code that writes the ISBN field and getting it to skip/ignore hyphens. > Should make it look real pretty! There are multiple ways of addressing this. I would err on the side of letting cataloguers enter whatever they want and cleaning up the results when you pull the ISBN - it looks like the cleanISBN() function in opac_utils.js is a candidate, and/or in the AddedContent Perl layer (given that requests may come in from places other than rdetail.js / result_common.js).
