Comet project complete codebase
----------------------------
---------------------------
Ed Chamberlain, Cambridge University Library 2011 - licensed under the GPL. 
Code is provided 'as is' with no firm assurance of support or functionality. May need substantial alteration to work


Perl based Marc21 analysis tool
-----------------------------------------------------
- Found in /analysis. A standalone tool designed to analyse Marc21 records and assess origin through identifier field codes. See the readme file in that directory for more details.

Perl based Marc21 to RDF publishing tool
---------------------------------------------------
 -Found in /conversion. features extensive CSV based customization for vocal and graph output. See the readme file in that directory for more details. Work by Huw Jones


/ PHP/MYSQL Comet RDF data service for libraries - bare bones setup instructions
--------------------------------------------------------------------
Built with the ARC2 library https://github.com/semsol/arc2/wiki and ARC2 starter pack https://github.com/tuukka/arc2-starter-pack
(both included)

Knowledge of PHP, Apache and MYSQL is required. Some understanding of RDF, RDF stores and the ARC2 codebase is also helpful. 


Overview
-----------------
/arc/ - arc2 codebase
/comet/comet.php - container class with functionality to query datastore and return requested content. Also a few other tricks
/includes/ - css header and footer helpers
index.php - home page with sample query and code
endpoint.php - generic SPARQL endpoint
label.php, record.php, results.php - Takes URI's or labels as parameters (from mod_rewrite, passes to comet.php function and returns corresponding data)
cli.php - enables command line querying and control of rdfstore
enricher.php a sample script to find subject URI's and match their labels against id.loc.gov. Corresponding triples are then written back to the RDf store!

I - Install
-----------
1) Copy all PHP in this directory into a PHP enabled web server. Ensure Mod_rewrite is enabled on the apache server for the directory and sub driectories.
PHP command line interface may need to be installed


II - Database setup
------------------
1) Create a database and corresponding user for MYSQL. Enter server, user and database details details in config.php.
2) Don't create and tables yet, although SQL to do so is in /comet directory. 
3) Load up index.php, tables should be generated automatically. 


III- Load some data
-------------------
1) Load some RDF. IF you don't have any, try using the perl script in  / analysis to create some from MARC. Remember to alter base URIs to match your intended site. 
2) At the server command line, go to the web directory and type using a sample of triples or other RDF and a suitable name for the graph representing the dataset:

./cli.php "LOAD <file:////var/www/data/sample.nt> INTO <http://your.lib.ac.uk/context/dataset/bib>" >> load_log.txt

3) revisit index.php - all being well used properties should display

4) Go to endpoint.php and try a sample SPARQL query suggested in the box

Alternatively, the site can use an external or existing datastore with an Endpoint. Replace the store config array and object in  config.php with a remoteStore
object. Further details on the Arc2 documentation. 

https://github.com/semsol/arc2/wiki/Remote-Stores-and-Endpoints


IV. - Other useful data management commands
--------------------------------------------
delete a graph
./cli.php "DELETE FROM <http://your.lib.ac.uk/context/dataset/bib>"

delete all triples attached to a graph
./cli.php "DELETE FROM <http://your.lib.ac.uk/context/dataset/bib>  {?s ?p ?o.}"

delete triples matching a specific SPARQL query
./cli.php "DELETE FROM <http://your.lib.ac.uk/context/dataset/bib>  {?s  <http://www.w3.org/2004/02/skos/core#broader> ?o.}"

More commands detailed here:
https://github.com/semsol/arc2/wiki/Using-ARC%27s-RDF-Store

These can all be run at Command line. They can also be enabled for use on the public sparql endpoint by editing config.php


V) Edit .htaccess file to match the URI's in your RDF
------------------------------------------------------
1) This is required to pass URI's to the comet.php application classes, which can in turn render the RDF in the required format
2) This may require some knowledge of mod_rewrite - See this blog post for some pointers
http://cul-comet.blogspot.com/2011/05/small-but-fiddly-win-for-uris.html

VI) - Test other stuff
----------------------
1) Go to a record URI - ensure HTML displays ok. Try following links and seeing what is attached via the 'View entries related to this' link.
2) With a record page, try appending the URI with .xml, .json, .rdf in your browser
3) Try a content specific request  curl -H "Accept: application/rdf+xml" http://data.lib.cam.ac.uk/id/entry/cul_comet_pddl_4589705
4) Try a label match at /label/Dogs - see if your page is redirected to the right URI. mod_rewrite may need some work if not.
5) Modify and run enricher.php to add library of congress subject links to subject nodes