coonf - Python Package for Conferences

---

Front Matter

Copyright © 2003, 2008, 2011 Krešimir Kumerički. All rights reserved.

Abstract:

Contents

• • 1 Introduction
  • 2 Requirements
    • 2.1 Software Requirements
  • 3 Installation and basic configuration of coonf
    • 3.1 Installing coonf
    • 3.2 Creating database
    • 3.3 Configuring application form
  • 4 Accessing and modifying data from command line
  • 5 Module web -- automatic WWW site generation
  • 6 Module abstracts -- on-line submission of abstracts
  • 7 Module uploads -- on-line file uploads

1 Introduction

Organizing a conference is usually full of various repetitive tasks which are, of course, best left to the computers. This collection of Python modules can be helpful if you want to automate some of the following tasks

• on-line application of participants and/or inserting their data into a database
• accessing and searching this data from command-line
• creating stylistically consistent WWW conference site with possibility of serving dynamical pages (e.g. listings of participants, various searches) using data from the above mentioned database (module web)
• on-line submission of abstracts (module abstracts)
• on-line uploads of files (e.g. PowerPoint presentations or proceedings contributions) (module uploads)

Some functionality of coonf is offered via WWW interface, some via command-line interface, and some by editing and executing python scripts. Some features are relegated to modules which operate independently (unless stated otherwise) and can be ignored if you don't need corresponding features. Note that these are not modules in the sense of Python modules/packages but just some convenient blocks into which various components of coonf are divided.

2 Requirements

To get everything to work you should have some basic knowledge about CGI scripting (not necessarily in Python, any CGI experience may be sufficient) and relational databases. Therefore, before even trying to put this package to work you should at least be able to:

• create and install simple CGI script (``Hello World'' will do) on your machine. This can be quite involved when you do it for the first time. Consult relevant documentation and HOWTOs on the Web.
• create a simple table in your database and try inserting some data into it.

For more advanced usage and for some of the modules, some knowledge of Python programming language is necessary. The more you know about Python and databases the more useful this package becomes.

2.1 Software Requirements

For minimal installation you need to have the following software installed (for modules 'web' and 'abstracts' some additional stuff is needed, as specified below):

• Python (http://www.python.org), at least version 2.2
• MySQL (http://www.mysql.com)
• Web server, e.g. Apache (http://www.apache.org)
• Additional Python packages:
  • MySQLdb Python database interface (http://sourceforge.net/projects/mysql-python)
  • SQLObject object-relational mapper (http://sqlobject.org)
  • eGenix.com mxDateTime extension (http://http://www.egenix.com/files/python/eGenix-mx-Extensions.html)
  • Python Cryptography Toolkit (http://www.dlitz.net/software/pycrypto/)
  E.g. on Debian Linux additional packages to be installed are python-mysqldb, python-sqlobject, python-egenix-mxdatetime and python-crypto

• If you want automatic emailing of confirmation messages to clients or administrators you also need email server (SMTP daemon) such as sendmail.

3 Installation and basic configuration of coonf

3.1 Installing coonf

python setup.py install

Move everything from HTML and CGI directories of coonf source distribution into directories where your web server can find them. In my case it goes like this:

shell> mv coonf-0.??/HTML /var/www/conference
shell> mv coonf-0.??/CGI /usr/lib/cgi-bin/conference

Python scripts in directory scripts will be needed later for accomplishing various tasks. You can put them to some convenient place.

Edit config.py from CGI directory carefully. In principle, it is a Python source file, but the syntax is very simple and you should have no problems modifying it to suit your needs even if you don't know Python language. Comments there should lead you.

In particular, variables HTML_ROOT and CGI_ROOT should point to the directories where your have just put stuff from HTML and DGI directories.

This file contains database password, so it should be readable only to WWW server and people who will have administrative privileges to edit participants' data.

shell> chown -R www-data:confadmins /usr/lib/cgi-bin/conference
shell> chmod 660 /usr/lib/cgi-bin/conference/config.py

Also in the beginning of the files CGI_ROOT/cgi_init.py and, if you need abstracts, CGI_ROOT/abstracts/pdfview.py, you have to specify absolute path to config.py!

3.2 Creating database

For example, on MySQL, after installing MySQL software you should login as root, and then type:

shell> mysql --user=root mysql
mysql> create database <DB_NAME> DEFAULT CHARACTER SET utf8;
mysql> GRANT ALL PRIVILEGES ON <DB_NAME>.* TO <DB_USER>@localhost
           IDENTIFIED BY '<DB_PASSWD>';
mysql> GRANT ALL PRIVILEGES ON <DB_NAME>.* TO <DB_USER>@"%"
           IDENTIFIED BY '<DB_PASSWD>';

If needed, you can similarly grant access to other users.

Also, you can have more databases for e.g. more conferences. Than you just repeat the procedure and create separate databases and corresponding HTML_ROOT and CGI_ROOT direcotories, containing corresponding config.py.

3.3 Configuring application form

The idea is to set up a HTML page with HTML application form so that entered data get inserted into a single row of table People in the database. This can be the usual data about conference participant (name, address, title of his talk etc.) but this could also be used in non-conference application to set up a simple WWW interface for inserting any kind of data into database. Configuration is done like this:

1. Create page with HTML form somewhere on your site. (You can probably just use publish_fancy.py script from web module. In that case you should install module web, see Sect. 5) This HTML form should point to register.py as its corresponding CGI script (ACTION attribute of HTML form).
2. Edit register.conf (found in CGI_ROOT directory) and specify data that will be entered in form. Syntax is specified in file itself. Basically, to each item in HTML form there should be a matching FIELD item in register.conf.
3. Edit register.py (also found in CGI_ROOT directory) and edit some obvious strings there to specify content of the confirmation email and WWW page that will be returned to clients after they fill out the form.
4. Create database table People by single run of create_table_people.py script with coonf configuration file as an argument.

   shell> create_table_people.py /usr/lib/cgi-bin/conference/config.py
   

Now you are all set up and you should be able to make a test by pointing your WWW browser to application HTML page (application.html in HTML_ROOT by default) and submit some data to the database.

4 Accessing and modifying data from command line

Data entered via application form can be accessed via WWW with help of CGI scripts people.py, person.py and search.py. This is seen in default setup.

Access, and, additionally, editing of data is possible also via command line to users who have read permission on coonf configuration file (config.py in CGI_ROOT directory), which containts database password. You should preferably go to CGI_ROOT directory (or put it in your PYTYONPATH environment variable). Then start Python interpreter (or, better, IPython) and import coonf package, configuration data, and make this data available to coonf python module:

shell> python

>>> from config import *
>>> import __builtin__; __builtin__.cfg = cfg
>>> from coonf import *

Then connect to database and use dbconnect() to create a class of participants, which will be used for all access.

>>> People = dbconnect()

All data corresponding to a person with id=5 can be printed out using command print asciiRecord(People, 5), or shorter print aR(People, 5)

>>> print aR(People, 5)

================================[    5 ]================================
         name =>  albert
      surname =>  zweistein
        email =>  alby@princeton.edu
          tel =>  21342134
  institution =>  IAS
       street =>  N/A
         city =>  princeton
          zip =>  N/A
      country =>  US
         pres =>  O
    prestitle =>  Theory of everything
     banketYN =>  Y
         diet =>  N/A
========================================================================

Then you can change any of these attributes like this:

>>> People.get(5).surname = "dreistein"

Syntax is obvious but for people who know object-oriented programming it can be said here that each person is an instance of class People.

For searching you have two functions available: asciiSearchAll() or aSA() that searches all attributes, and asciiSearchName() or aSN() that searches only names and surnames. Arguments of these functions are strings (don't forget quotation marks!) and search is case-insensitive.

>>> asciiSearchName(People, 'ei')
===========================
|  5 | albert | dreistein |
| 20 | steven |  weinberg |
===========================

Deleting a person with id=15 is simple:

>>> People.delete(15)

You can also do more complicated things. E.g. to find emails of all participants from Croatia:

>>> croats = People.select(People.q.country=='Croatia')
>>> print asciiTable(croats, 'id', 'name', 'surname', 'email')
================================================
| 1 |  prvi |      covjek |       njegov@email |
| 2 | drugi | participant |          his@email |
| 7 | again |    somebody | address@hidden.gov |
================================================

Of course, all this can also in principle be done by direct access to the database using SQL language.

5 Module web -- automatic WWW site generation

This module requires python package HTMLgen. (On Debian: python-htmlgen.)

All pages are defined either in publish_simple.py (simple pages that anyone with some knowledge of HTML can edit) or in publish_fancy.py (sofisticated scripts that generate timetables, etc.)

First you should edit template file (TEMPLATE_FILE in config.py) and CSS file (CSS_FILE in config.py) to design general appearance of your WWW pages (and to correct URLs there!)

Then edit scripts publish_simple.py and publish_fancy.py to create actual pages with some content. Each page is defined by four strings as described in table

Running these two scripts (they take coonf configuration file as an obligatory argument) will publish pages to HTML_ROOT.

shell> publish_fancy.py /usr/lib/cgi-bin/conference/config.py
shell> publish_simple.py /usr/lib/cgi-bin/conference/config.py

This method of specifying all pages in a single file (or two) is convenient only for small sites with not much statical content. If you have pages with lot of content you will be better off by specifying content strings in external files and importing them.

6 Module abstracts -- on-line submission of abstracts

This module requires pdflatex. (On Debian: tetex-base, tetex-bin and probably tetex-extra)

Installation:

1. Create database table for abstracts by single run of create_table_abstracts.py script.

   shell> create_table_abstracts.py /usr/lib/cgi-bin/conference/config.py
   

2. Create directory abstracts/PDF (in CGI_ROOT dir) with permissions 777
3. Turn the module on by setting MODULE_ABSTRACTS and EMAILING in config.py to 1.

Then (consult register.py) every client will get an email confirmation of application with URL of his private page where he can enter and modify his abstract. At the moment abstracts are set in LaTeXand resulting abstract is set as a nice PDF page.

7 Module uploads -- on-line file uploads

Installation:

1. Create directory uploads/FILES (in CGI_ROOT dir) with permissions 777
2. Turn the module on by setting MODULE_UPLOADS and EMAILING in config.py to 1.

Then (consult register.py) every client will get an email confirmation of application with URL of his private page where he can upload file of his electronic presentation. Files will end up in CGI_ROOT/uploads/FILES and they will have name of the form <id>-<surname>-v<number>-<original filename> where <number> is version number incremented by each submission of file with same name by the same person. Thus no files are overwritten.

About this document ...

This document was generated using the LaTeX2HTML translator.

LaTeX2HTML is Copyright © 1993, 1994, 1995, 1996, 1997, Nikos Drakos, Computer Based Learning Unit, University of Leeds, and Copyright © 1997, 1998, Ross Moore, Mathematics Department, Macquarie University, Sydney.

The application of LaTeX2HTML to the Python documentation has been heavily tailored by Fred L. Drake, Jr. Original navigation icons were contributed by Christopher Petrilli.

---

coonf - Python Package for Conferences

---

Release 0.9.11 (beta release), documentation updated on 27 Feb 2012.