README

This repository is for my meagre contribution to zotero, streamlining import of files from EndNote.

================================
To use this RIS.js:
================================
- To correctly import File Attachments, you must edit RIS.js and place the
    absolute path to your EndNote PDF folder:
<path_to_folder_where_EndNote_library_lives>/<EndNote_library_name>.Data/PDF/
    in the variable "RIS_internalPDFPath."
- If you want to get attached PDFs for Journal Articles, you need to go into
    the Refman Output Style in style editor (in EndNote X14 for Mac, you get
    there via Edit->Output Styles->Open Style Manager) and change L1 from
    outputting "URL" to outputting "File Attachments".  So the line will go
    from looking like this:

L1  - `URL|`

to looking like this:

L1  - `File Attachments|`

I'd probably make a copy of the style, too, before you change anything.

--------------------------------
Notes on how this RIS.js works:
--------------------------------
- If you have lots of files attached to a given reference, Zotero will
    sometimes freeze up.  3 or 4 usually works just fine.  This code contains
    a variable, "RIS_maxImports", that defines how many import files it will
    process per reference before storing subsequent ones in notes, such that you
    can import them by hand later.  It will also attach a note and log debug
    messages each time it encounters a file that it doesn't import, so if you
    want to know which files you need to update, turn on debugging before you
    run your import, then search for "RIS.js" in the debug output.

- * If you disable all addons other than zotero and python when you run a large
    import, it makes it much less likely you will break Firefox, even if you
    have lots of attachments on some references.

- the variable "RIS_risFieldToImportFieldMap", defined around line 1065, is the
    master mapping of RIS tags to their processing instructions.  RIS tags are
    either mapped directly to an EndNote field (CN is dumped into "callNumber" field):

    CN : new ImportField( ImportField.IN_TYPE_DIRECT, "callNumber", null, false ),

    or passed to a function for processing (A3 is passed to "processCreator" function):

	A3 : new ImportField( ImportField.IN_TYPE_FUNCTION, "", processCreator, true ),
	
	Any tag that isn't in the map will be appended to the reference as a note.

- processing functions are defined above "RIS_risFieldToImportFieldMap", in
    alphabetical order.  They should accept the following arguments:
    - item_IN - Zotero Item instance that we are populating.
    - tag_IN - String name of tag we are currently processing.
    - value_IN - String value of current tag (if multiple lines, will be a space-delimited concatenation of values from each line).
    - valueArray_IN - optional Array of values if multiple values for a given tag.

- To add a new processing function, define the function, then either update the
    mapping for the field you will process to reflect that it is now being
    passed to a function instead of being placed directly into a field or add
    a new row for the mapping.

- Example of turning a direct mapping into a function (pass CN to "processCN"
    instead of just putting it in the "callNumber" zotero field):

    CN : new ImportField( ImportField.IN_TYPE_DIRECT, "callNumber", null, false ),

    becomes (after defining processCN()):

    CN : new ImportField( ImportField.IN_TYPE_FUNCTION, "", processCN, false ),

- processTY() is where the itemType is set, based on the reference type in the
    TY tag.

- Regular Expression seem to make Firefox more likely to go out to lunch when
    combined with PDF import and processing, so I removed all from my code in
    this file.