examples = {
'example1': {
    'categories': u'case study, library',
    'description':
        u'Outline\r\n=======\r\n\r\n.. contents:: :local:\r\n\r\n'
        u'Background on project\r\n---------------------\r\n\r\n'
        u'* Reminder/calendar web application\r\n* Primarily handles '
        u'yearly holiday, birthday, and anniversary reminders\r\n'
        u'* Tasked to replace a 3rd-party Perl+ASP system\r\n'
        u'* Serving over 3 million users on AmericanGreetings.com '
        u'and BlueMountain.com\r\n* Business wanted new capabilities, '
        u'less maintenance\r\n\r\nProblems to solve\r\n'
        u'-----------------\r\n\r\n'
        u'* Event recurrence without intervention\r\n\r\n'
        u'  * No manual intervention (shouldn\'t have to set up new '
        u'holiday dates every year)\r\n'
        u'  * No sentinel process to fix data\r\n\r\n'
        u'* Time zone support\r\n\r\n  * Sending cell phone reminders '
        u'in the middle of the night is rude\r\n\r\n'
        u'What\'s dateutil?\r\n----------------\r\n\r\n'
        u'* Extensions for Python\'s datetime module\r\n'
        u'* Provides:\r\n\r\n  * Recurrence for a superset of RFC '
        u'2445 (iCalendar)\r\n  * Exhaustive timezone support\r\n'
        u'  * Date parsing\r\n  * Relative timedeltas\r\n\r\n'
        u'Why dateutil?\r\n-------------\r\n\r\n'
        u'* Compatible with our production environment '
        u'(Python 2.3)\r\n* No dependencies on other projects such '
        u'as Zope\'s interfaces\r\n* Highly Pythonic, clean use\r\n'
        u'* Recurrence even handles Easter\r\n'
        u'* Timezone mangling more reliable than other projects I '
        u'evaluated\r\n* One-stop shopping for my datetime '
        u'needs\r\n\r\nBasic recurrence\r\n----------------\r\n\r\n'
        u'* with ``dateutil.rrule``\r\n* how to create and use\r\n'
        u'* example code\r\n\r\nMore sophisticated recurrence\r\n'
        u'-----------------------------\r\n\r\n'
        u'* with ``dateutil.rruleset``\r\n* why, how to create and '
        u'use\r\n* example code\r\n\r\nRecurrence and February 29, '
        u'or the "Pirates of Penzance" problem\r\n'
        u'----------------------------------------------------------'
        u'------\r\n\r\n* an event that recurs yearly but begins on '
        u'February 29 only generates recurrences in leap years\r\n'
        u'* this is bad for things like birthdays and anniversaries '
        u'(worst case scenario, you are indentured to the Pirate King '
        u'for a lot longer than you expected)\r\n* solution: add '
        u'extra February 28 rdates to the rruleset on the fly\r\n'
        u'* example code\r\n\r\nTime zones\r\n----------\r\n\r\n'
        u'* how to get tzinfo instances\r\n* adjusting for the '
        u'user\'s preferred timezone\r\n* example code\r\n\r\n'
        u'Other capabilities of dateutil\r\n'
        u'------------------------------\r\n\r\n'
        u'* ``relativedelta``, often easier and more convenient '
        u'than ``timedelta``\r\n\r\n  * example code\r\n\r\n'
        u'* Parsing date strings into datetime instances\r\n\r\n  '
        u'* example code\r\n\r\n* Much more with time zones\r\n'
        u'  \r\n  * basic tzinfo classes for UTC and offset-based '
        u'time zones\r\n  * OS-local time zones\r\n  * parsing time '
        u'zone strings\r\n  * parsing ical time zone definitions\r\n'
        u'\r\nConclusion\r\n----------\r\n\r\n* dateutil saved my '
        u'project -- huge time savings, allowed me to focus on doing '
        u'things rather than rebuilding the wheel\r\n* A great '
        u'complement to datetime\r\n* Would be good to see as part of '
        u'the standard library some day\r\n',
    'duration': 30,
    'example_author': u'Mike Pirnat (AG Interactive)',
    'level': u'I',
    'name': u'Example 1',
    'summary': u"A practical look at dateutil, a set of powerful extensions "
        u"for Python's datetime module.\r\n\r\n    When I was asked to "
        u"replace a vendor-supplied reminder web application with a more "
        u"robust and sophisticated Python service, two of the most  fearsome "
        u"requirements involved better handling of event recurrence and time "
        u"zone support.  Luckily, I happened across Gustavo Niemeyer's "
        u"dateutil (http://labix.org/python-dateutil) package at exactly "
        u"the right time; it quickly became the indispensable core of AG "
        u"Interactive's new calendaring service, catering to over three "
        u"million users on AmericanGreetings.com and BlueMountain.com .\r\n"
        u"\r\nI'll illustrate dateutil's capabilities, how I used it to "
        u"solve some of my tricky problems, and share some interesting "
        u"adventures I had with it along the way.\r\n",
    'title': u'Dateutil to the Rescue!'},
'example2': {
    'categories': u'implementations',
    'description':
        u'Over the past few years the Perl community has been designing '
        u'and\r\ndeveloping Parrot, a virtual machine interpreter for '
        u'dynamic languages.\r\nWhile this virtual machine is obviously '
        u'intended to support compilation\r\nin Perl, a fundamental design '
        u'goal of Parrot is also to provide language\r\ninteroperability at '
        u'the bytecode level with many scripting languages,\r\nincluding '
        u'(especially) Python, Ruby, Tcl, and others.\r\n\r\nWithin the past '
        u'year, Parrot has developed a number of advanced tools\r\nfor '
        u'implementing language translators.  The parser grammar engine '
        u'(PGE)\r\nmakes it simple to develop language parsers for Parrot, '
        u'while the\r\ntree grammar engine (TGE) provides a '
        u'attribute-grammar-based structure\r\nfor translating parse trees '
        u'into executable bytecode.  While both PGE\r\nand TGE are '
        u'continuing development, they are being successfully used\r\nin '
        u'the development of compilers and translators for Perl 6,\r\nTcl, '
        u'Ruby, and APL, and work is being performed for other languages\r\n'
        u'such as Python and JavaScript.\r\n\r\nThis talk continues from the '
        u'previous talk ("Python on \r\nParrot -- overview") by providing the '
        u'implementation details \r\nfor a Python compiler in Parrot.  The '
        u'talk gives a detailed \r\nlook at the Parrot compiler toolkits, '
        u'and walks through the \r\ngrammar, parser, and translation into '
        u'executable code.\r\n\r\nThe talk is aimed at people who are '
        u'interested in virtual machines,\r\nare somewhat aware of the '
        u'Parrot virtual machine (e.g., from the\r\nfirst talk), and want '
        u'to know more of the technical details of\r\nhow things work.\r\n'
        u'\r\nI\'ve indicated this talk as a 45-minute talk, because there '
        u'is\r\na lot of material to be covered here (and I\'d like to '
        u'leave\r\nplenty of time for questions).  If necessary for '
        u'scheduling\r\nreasons I can likely squeeze the talk down to a '
        u'30 minute\r\npresentation.\r\n\r\nTalk outline\r\n  - status of '
        u'the compiler (with demos)\r\n  - detailed introduction to PGE and '
        u'TGE (with demos)\r\n  - Parrot\'s AST\r\n  - degree of '
        u'compatibility with existing Python interpreters\r\n  - handling '
        u'C-based modules\r\n  - integrating Python with other languages and '
        u'modules\r\n  - current performance\r\n  - next steps in '
        u'development\r\n\r\n',
    'duration': 45,
    'example_author': u'Dr. Patrick R. Michaud',
    'level': u'I',
    'name': u'Example 2',
    'summary':
        u'Parrot is a virtual machine environment that is being '
        u'developed to support dynamic languages such as Perl, Python, '
        u'Ruby, JavaScript, and many others.  This talk provides a '
        u'detailed description of the architecture of Parrot, its '
        u'compiler tools, and the implementation of a Python compiler '
        u'for Parrot.',
    'title': u'Python on Parrot -- under the hood'},
'example3': {
    'categories': u'panel, testing',
    'description':
        u"The panel will allow testing tool authors to interact with the "
        u"audience\r\nand take questions on various topics such as:\r\n"
        u"\r\n* what need prompted the creation of the testing tool\r\n"
        u"* what type of testing does the tool belong to \r\n\r\n   (unit, "
        u"functional, acceptance, system, performance)\r\n\r\n* what specific "
        u"features does the tool offer that other tools lack\r\n* what are "
        u"the most common testing scenarios you have seen in your user base"
        u"\r\n* are there any OS/platform specific gotchas related to the "
        u"tool\r\n* how extensible is the tool (plugins etc.)\r\n* how "
        u"easy to learn is the tool\r\n* how well tested is the tool\r\n* "
        u"how well documented is the tool\r\n* others\r\n\r\nI have "
        u"contacted a number of testing tool creators. \r\nSome of them "
        u"have yet to respond to my email, and some already accepted:\r\n"
        u"\r\n* Titus Brown (author of twill)\r\n* Chad Whitacre (author "
        u"of testosterone)\r\n* Martin Taylor (Texas Instruments, author of "
        u"Python test framework used internally by TI's Calculator "
        u"division)\r\n* Kumar McMillan (author of testtools)\r\n* Holger "
        u"Krekel (author of py.test) -- does not know if he can make it"
        u" to PyCon, but will participate in case he will make it\r\n"
        u"* Neal Norwitz (author of PyChecker) -- has not confirmed yet, "
        u"but expressed interest\r\n",
    'duration': 45,
    'example_author': u'Grig Gheorghiu',
    'level': u'B',
    'name': u'Example 3',
    'summary':
        u'I maintain a "Python Testing Tools Taxonomy" Wiki page at'
        u'\r\n <http://pycheesecake.org/wiki/PythonTestingToolsTaxonomy>. '
        u'\r\n\r\nA panel where authors of some of these tools \r\nwould '
        u'discuss and take questions on topics such as:\r\n\r\n * what need '
        u'prompted the creation of the testing tool\r\n * what type of '
        u'testing does the tool belong to \r\n   (unit, functional, '
        u'acceptance, system, performance)\r\n * what specific features '
        u'does the tool offer that other tools lack\r\n * what are the most '
        u'common testing scenarios you have seen \r\n    in your user base'
        u'\r\n * are there any OS/platform specific gotchas related to '
        u'the tool\r\n * how extensible is the tool (plugins etc.)\r\n * '
        u'how easy to learn is the tool\r\n * how well tested is the tool'
        u'\r\n * how well documented is the tool\r\n * others\r\n\r\nDetails '
        u'and Questions: http://us.pycon.org/TX2007/TestingToolsPanel\r\n',
    'title': u'Panel: Testing Tools'}
}