it absolutely makes sense to write translators for sites with DOI - you can add more info (e.g. CrossRef doesn't provide keywords and abstracts), you can attach a link to the article PDF (for people with access) and you can import search results.

Have a look at the documentation and see if this looks like something you'd like to try to do - there are a number of translators for sites with very similar structure, so I could point you to an example.

It looks like you might also just be able to do this with the translator framework:
http://www.zotero.org/support/dev/translators/framework

using mainly //following-sibling:: type xpaths
http://www.velocityreviews.com/forums/t685244-xpath-to-get-the-first-sibling.html

Thank you for your replies. I did not know that there are different "agencies" for DOIs. The answer that I got from CrossRef is
"The DOI you have reported is registered with some other agency. This is not a CrossRef DOI." ... as aurimas said.

So I would like to create my own translator. Do you know an up to date tutorial, that works with Firefox 15.0.1 ?

The following tutorial does not work:
http://www.zotero.org/support/dev/how_to_write_a_zotero_translator_plusplus
(XPather not available for Firefox 15.0.1)

((((Also, if I hit the button "Test Regex" in Scaffold, I get no result at the TestFrame.))))
Edit: Works with scaffold 3:
https://github.com/downloads/zotero/scaffold/scaffold-3.0.xpi


@adamsmith: Thank you for the hint to the translator framework. If you know an example for a conference paper /similar page please tell me.
(The example at http://www.zotero.org/support/dev/translators/framework
does not seem to work with Target = http://www.eurasianet.org/
What is the right target for the eurasianet example ? Does the example work with an existng web page at all?)

Ok, the current code of my translator is here:

http://matameko.wikispaces.com/file/view/EU+PVSEC+Proceedings.js

It seems to work in scaffold 3. I have still some questions:

- How can I use the translator in zotero? (I am missing a button "Use custom translator for current web page" or so)

- How can I add several text objects to combine them in one entry?
The following line does not work:
extra: "info1: " + FW.Xpath('blabla_1').text() + "\n" + "info2: " FW.Xpath('blabla_2').text()
(I did not manage to use the commands "append" or "prepend" for this purpose)

- How can I remove the string "Abstract/Summary:" from my abstractNode? The follwing line does not work:
abstractNote : FW.Xpath('/html/body/div/div[8]/table/tbody/tr/td[3]').text().remove("Abstract/Summary:")

1. How to use translator: -------------------------------------------
Reloading the page did the trick. The icon in the firefox adress bar changed from a blank white paper(DOI) to the conferencePaper icon.

2. Adding strings: --------------------------------------------------
The example on the google groop uses metadata. Following line works:
extra : FW.Xpath('concat("author: ", string(//meta[@name="author"]/@content))'),

However, the strings that I want to concat are not available as meta data. Following code works for me:
extra : FW.Xpath('concat("topic: ", string(//html/body/div/div[8]/table/tbody/tr[4]/td[2]), "\n",' +
' "subtopic: ", string(//html/body/div/div[8]/table/tbody/tr[5]/td[2]), "\n",' +
' "price: ", string(//html/body/div/div[8]/table/tbody/tr[11]/td[2])' +
' )'
),

3. Remove substring: ------------------------------------------------
I had to use slash (\) instead of quotation marks (" or ') for the string, as shown in your example. The \/ removes the / in Abstract/Summary. I use a . to remove the space. Finally it is:

.remove(/Abstract\/Summary:./)

( instead of .remove("Abstract/Summary: ") )

4. No extra code for translator framework: --------------------------
Works great. The "Uses translator framework" checkbox is located at the Metadata tab in Scaffold 3

5. More stable Xpaths: ----------------------------------------------
I did not manage to improve the XPaths so far. Here are two snippets from the source code of the EU PVSEC example page:
=====================
<tr>
<td><b>DOI:</b></td>
<td>10.4229/25thEUPVSEC2010-2DV.1.19</td>
</tr>
...
======================
<tr>
<td> <br><b>Title:</b></td>
<td> <br><b>Open Source Graphical User Interface
...
=======================

Following lines did not work for me:

... tr[contains(text(), "DOI:")]/following-sibling).text()
... td[contains(text(), "DOI:")]/following-sibling).text()
... td[contains(text(), "<b>DOI:<b>")]/following-sibling).text()

I guess the problem is, that there are <b> or <br> tags. What should be returned by the text() command? Maybe it is an empty string.

6. How to access a single item of a string list?: -------------------
I would like to extract the place "Valencia, Spain" from following string:
"25th European Photovoltaic Solar Energy Conference and Exhibition / 5th World Conference on Photovoltaic Energy Conversion, 6-10 September 2010, Valencia, Spain"

(I might use a regular expression to remove the preceeding text. Hoewever, splitting and using the last item (last two items) would be more comfortable.)

I am able to split the string with the split command (like for the tags). How can I acces a single item (or several items) of the resulting array?

Following lines dont work:

place: FW.Xpath('/html/body/div/div[8]/table/tbody/tr[6]/td[2]').text().split(',',3)
place: FW.Xpath('/html/body/div/div[8]/table/tbody/tr[6]/td[2]').text().split(',')[3]
place: FW.Xpath('/html/body/div/div[8]/table/tbody/tr[6]/td[2]').text().split(',').pop()
place : FW.Xpath('concat("", string(//html/body/div/div[8]/table/tbody/tr[6]/td[2]).split(",")[3] )'),


=====================================================================
Here is my current translator code:
http://matameko.wikispaces.com/file/view/Zotero_translator_EU_PVSEC.txt

//div[@id="detailsTable"]//td[./b[text()="DOI:"]]/following-sibling::td
That should do the trick