Create Content from ISBN

Hi Mike,

This is already being done for DOI's and PubMed ID's so there is no reason I couldn't add an interface to isbndb as well, it would be a fairly trivial addition. The 6.x-2.x branch has modularized all the import export features out of the core biblio module, so one would just have to clone the DOI or PubMed module and modify it for the isbndb api.

Ron.

Upgrading would be highly recommended, since other than security updates, the 5.x branch hasn't seen any new features in years and is quite primitive compared to the 6.x series.

http://manas.tungare.name/software/isbn-to-bibtex/

Ron,

Let me re-open this issue as a very much desired feature request, particularly for the individual/small group user group audience.

By way of background, my biblio data includes about 150 books on one site and maybe 50 on the other. I am faced with the prospect as an individual user with the daunting prospect of having to enter quite a bit of data regarding more than 100 books via the biblio module interface. I also average getting 4-6 books per week from the local libraries and would like to record them in the biblio module as well for future reference. Obviously having a technology solution is better than (mis)typing in much information.

I spent a bit of time this weekend looking at what is available to individual users through the WorldCat and OCLC organizations and I see that one can now (as individual user) perform up to 1,000 requests per day and get back RSS data files with basic information about an ISBN. Information about this service is here:

http://www.oclc.org/developer/sites/default/files/handouts/USB_Web%20Ser....

The above information returned is quite basic biblio information. Hence, I suspect that a more complete and ideal solution would include something like the following:

1) On the initial display of the node/add/biblio form, there should be added a generic identifier field.
2) If the user enters data in this identifier field (and has the appropriate external access parameters set via the settings page), a client browser function sends the value back to the Drupal server. The server code then determines if this is a Library of Congress, WorldCat, or linked-library identifier and constructs the appropriate query URL.
3) The server then uses something like CURL to read the resulting Atom/RSS/MARC record into memory where it is parsed and values are mapped to the Biblio fields.
4) The server then bundles up the Biblio fields into a JSON record which is fed back to the client Javascript which then populates the fields of the form with the retrieved field data.
5) The user makes whatever manual changes and additions to the data and then saves the form as if it had all been hand entered.

Does approach make sense? I am curious to hear your and others' thoughts.

I also am happy to help with the coding of this functionality (with the proviso that I am still working off of the Drupal 6 branch and purposely have done nothing with Drupal 7 yet).

Cheers!! Lars

subscribing. i would enjoy to have ths option too.

Some interesting ideas here, which I will revisit when Biblio 7.x is finished.

Needed quick solution for this and created sub-module for Biblio 6.x-1.18. Bibliography ISBN
Not really sure if module should be promoted to full project and ported to 6.x-2.x and 7.x-1.x.

Thanks for the "heads up". I'll look into incorporating your code into the Biblio module.

I'm assuming this feature is supported in the D7 release but I have a practicality question. My original intention is to pull-down a CSV list or something using MARC of current textbooks (not my collection, just a list of textbooks on the market) and then use this module to import them and format them accordingly to a custom "book" content type.

In terms of the database and site storage requirements/performance, would it be better to pull them into a content type as nodes or just store the data, as lines in a table and pull data accordingly via Views. My gut tells me that importing potentially thousands of lines from a CSV and creating nodes would be ridiculously resource intensive.

In other words, is creating nodes from a CSV overkill?

Suggestions much appreciated.

Correct me if I'm wrong, but I think this is a little off the ISBN topic unless you are talking about pulling a list of ISBN numbers from MARC and using that list to look up the rest of the data?

FYI, Biblio can read MARC files directly so no need for CSV.

As for resource intensiveness... You are only creating the nodes once, so (in my opinion) it doesn't really matter if it takes 1 minute or 10. I regularly populate biblio databases with thousands of records and it while it can take a while, we are still only talking minutes or 10s of minutes, not days.

This version is no longer maintained. If this issue is still relevant to the Drupal 7 version, please re-open and provide details.