PyCon DC 2005 / Presentations - Python Wiki
Python Wiki Python Wiki / Python Wiki / Conferences / PyCon / PyCon dc 2005 / PyCon DC 2005 / Presentations View as Markdown PyCon DC 2005 / Presentations ¶ Legacy Wiki Page This page was migrated from the old MoinMoin-based wiki. Information may be outdated or no longer applicable. For current documentation, see python.org . #1. Python in the IT Field Greg Lindstrom / NovaSys Health ( http://www.novasyshealth.com ) In business, customer service is the name of the game.
In the Information Technology field, this service manifests itself in the ability to accept, format, store, and analyze data. At NovaSys Health, we selected Python for our programming needs because it’s easy to use, plays well with a variety of platforms (not only different operating systems, but a wide variety of databases) and the Python community is as helpful as it gets. This talk will discuss how Python is allowing us to list technology as a competitive advantage (accessing MS SQL database, teaching other techies, using Python’s standard library). #4.
MissionEngine : Multi-system integration using Python in the Tactical Language Project Prasan Samtani ( http://www.isi.edu/~samtani ), Hannes Vilhjalmsson ( http://www.isi.edu/~hannes ) University of Southern California - Information Sciences Institute (USC/ISI) Project homepage URL: http://www.tacticallanguage.com The Tactical Language Training system is designed to help learners acquire communicative competence in spoken Arabic and other languages.
Learners practice vocabulary items and learn gestures, following which they apply them in a simulated Mission Environment. In the simulation environment, users interact with the systemís onscreen characters using a speech recognition interface. The simulation environment is built on the Unreal tournament Game Engine. In the past several months, a major transition has been made from the original system in Java, to a new Python-centric architecture.
This presentation traces the evolution of the Tactical Language Training System from a set of separate programs to a single integrated system. The various different system components will be introduced, and the underlying technology behind will be briefly discussed. We will go over what features of the language we found especially useful for integration (latent-typing, lambda functions, pickle), and some hurdles we found particularly tricky coming from a C++/Java background.
The talk will also address the way we tackled the problem of talking to incompatible languages. For example, Unreal Tournament uses a proprietary scripting language - UnrealScript - and in order to communicate with Unreal we had to make use of an earlier ISI technology called GAMEBOTS, which is implemented both in Python and UnrealScript , and uses sockets as a method of communication. Other topics discussed in the context of Tactical Language will adress the performance issue (writing expensive routines - the speech recognition and NLP algorithms - in C and authoring extension modules), the importance of using a powerful language for build scripts, using algorithmic optimizations over language micro-optimizations and the importance of Python’s standard library in speeding up the development process. #7.
Happy Hooking: Designing Software for Extensibility and Customization Ed Leafe / The Dabo Project It is rare for anyone to create any non-trivial software project that will not need customization at some point. It is therefore wise to plan for this when designing your code by inserting hook methods that are designed to be customized to add additional functionality later on. By default these methods do nothing, so they can be freely overridden by others without the risk of losing a critical behavior because a superclass method had been overridden.
This is especially useful for creators of frameworks and tools that can be extended by their users. As one of the two primary authors of the Dabo framework, I am especially aware that in order to be useful, the code I create needs to be adapted and augmented by developers who build their apps with Dabo. I will show several code examples from Dabo to illustrate my points. Paper #9. Dabo, the 3-tier Database Application Framework) Ed Leafe / The Dabo Project There are several tools available that help a developer create applications for working with data: Visual FoxPro , Visual Basic, Delphi, etc.
They make some of the tedious parts of the app, such as ensuring that a change to a value in a text box gets propagated back to the correct column in the correct record of the correct table in the correct database on the correct host. These tools all have their strengths and weaknesses, but none are Open Source, and none are in Python! In March 2004, two long-time Visual FoxPro developers, Ed Leafe and Paul McNett , had both become convinced that Python and cross-platform apps represented the future.
Trouble was that there wasn’t a tool available that could do what we wanted; so as a result, we decided to create our own. Dabo is designed to create true 3-tier apps, in which the UI, the business rules, and the data storage are cleanly separated, using a Chain of Responsibility pattern for the messaging between tiers. We’ve chosen wxPython as our initial UI toolkit, but have written our UI classes so that their interfaces are not wxPython-specific.
We currently support MySQL, Firebird and PostgreSQL on the backend, with more to follow. Paper #10. Agile Testing with Python Test Frameworks Grig Gheorghiu / Avamar I present various test frameworks that are either written in Python or use Python/Jython as their scripting language. I discuss these frameworks in the context of agile testing. Here is a high-level view of the talk I’ll give: 1. What is agile testing? “agile” as in “agile methodologies”, (aka Extreme Programming) 2.
Why Python? eminently “agile” language 3. MAIN SECTION: Python-based test frameworks unit testing (unittest, doctest, py.test) continuous performance testing (pyUnitPerf) functional/system/integration testing (homegrown framework) test distribution, execution and reporting (STAF/STAX) acceptance testing (pyFIT/FitNesse) Web application testing (Selenium, MaxQ, mechanize, webunit, PBP, Twill) 4. Jython and Dynamic Java Recent discussions on Java and scripting languages Jython is leading the pack in terms of scripting the JVM Java-based test frameworks can be used within Python frameworks via Jython Examples of Java- and Jython-based frameworks (The Grinder, TestMaker , Marathon) Using Java-based tools with Jython (example: HTTPUnit) #11.
Localized Type-Inference in Python Brett Cannon / Cal Poly SLO This talk will present research into type inferencing in Python. The main goal of this work is to extract a performance gain from type-specific opcodes added to Python’s VM. The type information needed to use these new opcodes during code generation comes about from a simple, localized type inferencing algorithm. The design of this algorithm required that no changed be made to the semantics of Python.
Because of this requirement the type inference algorithm is hindered by Python’s dynamic nature. This talk will discuss what in Python’s semantics makes type inferencing difficult, the algorithm used, the added opcodes, and the results of adding the new opcodes. #12. The Roundup Issue Tracker Richard Jones / Common Ground Roundup is a simple-to-use and -install issue-tracking system with command-line, web and e-mail interfaces.
It is based on the winning design from Ka-Ping Yee in the Software Carpentry “Track” design competition. This presentation will cover: An introduction to Roundup A quick tour of the Web, e-mail and command-line interfaces What’s involved in setting up a tracker The hyperdatabase at the core of Roundup An indication of where Roundup’s going A simple customisation of the classic tracker #14. Documentation Costs Avoided using Python and other Open Standards Andrew Fine
The rest of this page is on the live site.