|
PhiloMine (release 0.1a) Download, Installation, and Use ARTFL/DLDC University of Chicago |
OK, you were warned. Expect this to change constantly as we are developing PhiloMine on an as need basis to support our own work. We will try to keep releases relatively coherent and hopefully backward compatible.
PhiloMine should run on any OS-X or Linux machine. The more memory and faster machine the better. We have had success using a Mac G5 Quad with 2GB memory.
Current dependencies include (but may not be limited to):
Installation. There is no installer script, so you should untar the latest version and copy all of the contents in cgi-bin/ to the philologic directory in your cgi-bin/. Check permissions and note that we assume #!/usr/bin/perl. Edit the "GLOBALS we are expecting section" in philomine1 and encycmine1 to reflect your installation.
We have supplied a couple of default search forms. You will need to edit these to suit for each database. Also note that you must copy utils/philominesubs.pl to the lib/ directory of every database and edit the two lists of bibliographic variables to suit your search form.
Encycmine is designed to work with div level objects in a database. You will need to generate div level word frequency data. Copy utils/mkcountbyobj.pl to your database directory and run it. This will generate two files: wordfreqbyobject and wordfreqbyobject.idx. These can be fairly large. Note that you must delete these files if they exist before rerunning, otherwise you will append to existing files.
Tarballs.
If you get stuck, please contact us. We're happy to try to help.
Hacking: please, feel free and send us code. PhiloMine is being developed by Russ Horton and Mark Olsen pretty much independently. If the code is nicely structured and well documented, this is certainly the work of Russ. Otherwise, blame Mark for convoluted, ugly and undocumented code.