It isn't entirely clear to me what you are asking.

Presumably you are asking about citation formatting (and mean apa.csl, rather than apa.xml).

The Zotero extension (to Firefox) includes a javascript implementation of citeproc. Citeproc takes (1) bibliographic information and (2) a .csl style file to create formatted citations.

For (1), the Zotero implementation of citeproc uses zotero item objects (stored in your database) directly. Other implementations uses MODS XML and/or other bibliographic files/databases.

For (2), the CSL style file is an XML document that adheres to the csl.rnc schema (on sourceforge, as part of the xbiblio project). The actual code that zotero uses to follow the instructions for formatting given in a CSL style file are part of zotero in e.g. /chrome/content/zotero/xpcom/csl.js

Part1: I mean definitely apa.xml.

I might be a bit thick, but what do you mean by apa.xml? CSL files are XML files & do have an info block & apa.csl would refer to the American Psychological Association citation style. I can't think of what you might mean by apa.xml.

But where are the data

The data on authors, etc. in Zotero are stored in Zotero's SQLite database (mozstorage in Firefox). Each individual reference can be put into a Zotero item object, with various javascript methods to do things like get individual fields.

I would like to test it with a small javascript (outside of zotero) to understand the procedure.

To use citeproc outside of zotero you have a few options:

• Make Zotero item objects and leverage the pre-existing citeproc code as-is
• Modify the citeproc code in Zotero to accept e.g. MODS.xml instead of just Zotero items (the place to do this is clearly marked in the code)
• Use a different (non-javascript) implementation of citeproc.

(I suspect Harald is clicking on the file on the styles page and going to File, Save As, which currently saves the style as apa.xml by default.)

Harald: You mean apa.csl.

If I select "Save page as..." in Firefox I now get proposed file names like ama.csl.xml, which probably isn't the aimed-for result.

OK, sorry, I mean apa.csl. To be quite sure, here are the
first lines:

<?xml version="1.0" encoding="UTF-8"?>
<style xmlns="http://purl.org/net/xbiblio/csl" class="in-text" xml:lang="en">
<info>
<title>American Medical Association</title>
<id>http://www.zotero.org/styles/ama</id>;
<link href="http://www.zotero.org/styles/ama"/>
<author>

Later it goes this way:

<macro name="author-short">
<names variable="author">
<name form="short" and="symbol" delimiter=", " initialize-with="."/>
<substitute>
<names variable="editor"/>
<names variable="translator"/>
<text macro="anon"/>
</substitute>
</names>
</macro>

Compared to normal xml: <name>Miller</name> (the data is 'miller') my question is: where or how come the data (Name or Author Miller) into that csl/xml file, where there is the 'author' in a macro as a variable and not between tags. I do not want citeproc to do it for me nor zotero, I just want to do it myself in a simple javascript (as I said I would do it in a normal xml with DOM etc.)
Perhaps my question is too simple but I would like to understand the basic procedure.

Thanks a lot
harald

Please refer to my first post and to Bruce's schematic of citeproc (mentally replacing 'XSLT' with a more general implementation (Zotero's uses javascript transformations, for example)).

The CSL file only provides instructions on how to format a citation. It does not store raw data.

The raw data that is used in the citation ("Miller", in your example) has nothing to do with CSL. It is part (1) in my previous description and is labeled "BibDB" in the figure. As above, this raw data is stored in a database for Zotero (but other implementations of citeproc use different storage).