Highwire-2.0 translator does not handle empty N1 tags appropriately

[karnesky]

Some highwire journals emit an empty N1 field for articles that lack a DOI, see ​http://forums.zotero.org/discussion/8259/ensuring-dois-and-dois/

Modify the regex on line 100 and/or 101 to look for a valid doi:
https://www.zotero.org/trac/browser/extension/trunk/translators/Highwire%202.0.js?rev=4589#L100

I haven't tried this, but something like this should work:

if (text.match(/N1\s+-\s+10\.\d+\/.+\n/)) {

var doi = text.match(/N1\s+-\s+(10\.\d+\/.+)\n/)[1];

[ajlyon]

This patch seems to address the problem. I've applied a version of it and posted the updated translator: ​http://github.com/ajlyon/zotero-bits/raw/master/Highwire%202.0.js

[karnesky]

Since posting this, I had implemented the proposed patch & it seems to work for me too.

[ajlyon]

(In [6128]) Closes #1539, empty N1 entries cause bad data in DOI for Highwire 2.0