I updated the J-Stage translator. Now it could catch the abstract and
keywords. Also, it works well for English and Japanese webpage.
Bugs:
1. It works fine under Zotero 1.0 but not so well under 2.0.
2. It always return error message, however, It really collected all
contents.
3. Some articles has only PDF url but no abstract webpage, this
translator will omit them.
http://groups.google.com/group/zotero-dev/web/J-Stage.js