Xapers - personal journal article management system
===================================================

Xapers is a personal document indexing system, geared towards academic
journal articles.  Think of it as your own personal document search
engine, or a local cache of online libraries.  It provides fast search
of document text and bibliographic data and simple document and bibtex
retrieval.

Document files (in PDF format) and source identifiers (e.g. DOI) are
parsed and indexed into a Xapian search engine [0].  Document text is
extracted from the PDF and fully indexed.  Bibliographic information
downloaded from online libraries is indexed as prefixed search terms.
Existing bibtex databases can be easily imported as well, including
import of pdf files specified in Jabref/Mendeley format.  Documents
can be arbitrarily tagged.  Original document files are easily
retrievable from a simple curses search UI.  The command line
interface allows for exporting bibtex [1] from arbitrary searches,
allowing seamless integration into LaTeX work flows.

Xapers provides source modules for some common online libraries:

  * DOI: http://www.doi.org/
  * arXiv: http://arxiv.org/

Contributions of additional library interface modules is highly
encouraged.

Xapers is heavily inspired by the notmuch mail indexing system [2].

[0] http://www.xapian.org/
[1] http://www.bibtex.org/
[2] http://notmuchmail.org/


Contact
=======

Xapers was written by:

    Jameson Graef Rollins <jrollins@finestructure.net>

Xapers has a mailing list:

    xapers@lists.mayfirst.org
    https://lists.mayfirst.org/mailman/listinfo/xapers

We also hang out on IRC:

    channel: #xapers
    server:  irc.oftc.net


Getting Xapers
==============

Source
------

Clone the repo:

    $ git clone git://finestructure.net/xapers
    $ cd xapers

Dependencies :
  * python (>= 2.6)
  * python-xapian - Python Xapian search engine bindings
  * poppler-utils - PDF processing tools
  * pycurl - Python bindings to libcurl
  * pybtex - Python bibtex parser

Recommends (for curses UI) :
  * python-urwid - Python Urwid curses library
  * xdg-utils - Desktop tools for opening files and URLs
  * xclip - X clipboard support for copying document fields

On Debian:

    $ sudo apt-get install python-xapian poppler-utils python-pycurl pybtex python-urwid xdg-utils xclip

Run the tests:

    $ make test

Debian
------

Debian/Ubuntu snapshot packages can be easily made from the git
source.  You can build the package from any branch but it requires an
up-to-date local branch of origin/debian, e.g.:

    $ git branch debian origin/debian

Then:

    $ sudo apt-get install build-essential devscripts pkg-config python-all-dev python-setuptools debhelper dpkg-dev fakeroot
    $ make debian-snapshot
    $ sudo dpkg -i build/xapers_0.1_amd64.deb


Using Xapers
============

See the included xapers(1) man page for detailed usage and information
on source modules and searching.

Command line interface
----------------------

The main interface to Xapers is the xapers command line utility.  From
this interface you can import documents, search, tag, etc.

The "add" command allows importing or updating single documents with
sources.  The "import" command allows importing an entire bibtex
databases (.bib file).  If the bibtex entries include "file" fields
(ala. Mendeley or Jabref), then those files are retrieved, indexed,
and imported as well.

Curses interface
----------------

The curses interface (accessed through 'xapers show ...') provides a
simple way to search the database and retrieve files.  Documents
matching searches are displayed with their bibliographic information
and a short text summary.  Document tags can be manipulated, files can
be viewed, and source URLs can be opened in a browser.

xapers-adder
------------

xapers-adder is a simple script that helps the adding of individual
documents to your Xapers database.  It can be used e.g. as a PDF
handler in your favorite browser.  It displays the PDF then presents
the user with the option to import the document into Xapers.  The user
is prompted for any sources to retrieve and any initial tags to add.
If the source is known, bibtex is retrieved and indexed.  The
resulting xapers entry for the document is displayed.

Development of more clever import methods is highly encouraged.

Python library
--------------

Xapers is really a python library interface under the hood:

    >>> import xapers
    >>> db = xapers.Database('~/.xapers/docs')
    >>> docs = db.search('tag:new')
    >>> for doc in docs:
            doc.add_tags(['foo'])
            ...
    >>> 

Development of new interfaces to the underlying library is highly
encouraged.