[Subversion] / PEAK / README.txt  

Diff of /PEAK/README.txt

Parent Directory | Revision Log

version 276, Wed Feb 28 22:03:10 2001 UTC version 525, Fri Oct 4 15:48:10 2002 UTC
Line 1 
Line 1 
 TransWarp Preview Release 0.1  PEAK Release 0.5 alpha 1
   
  Copyright (C) 2001 Phillip J. Eby, All rights reserved.   Copyright (C) 1996-2002 by Phillip J. Eby and Tyler C. Sarna.
  This software may be used under the same terms as Zope or Python.   All rights reserved.  This software may be used under the same terms
    as Zope or Python.  THERE ARE ABSOLUTELY NO WARRANTIES OF ANY KIND.
    Code quality varies between modules, from "beta" to "experimental
    pre-alpha".  :)
   
   
    Package Description
   
       PEAK is the "Python Enterprise Application Kit". If you develop
       "enterprise" applications with Python, or indeed almost any sort of
       application with Python, PEAK may help you do it faster, easier, on a
       larger scale, and with fewer defects than ever before. The key is
       component-based development, on a reliable infrastructure.
   
       PEAK is an application kit, and applications are made from components.
       PEAK provides you with a component architecture, component infrastructure,
       and various general-purpose components and component frameworks for
       building applications.  As with J2EE, the idea is to let you stop
       reinventing architectural and infrastructure wheels, so you can put more
       time into your actual application.
   
       But PEAK is different from J2EE: it's a single, free implementation of
       simpler API's based on an easier-to-use language that can nonetheless
       scale with better performance than J2EE.
   
       PEAK is the successor to TransWarp, an experimental toolkit for software
       automation in Python.  PEAK takes the best of the techniques and ideas
       from TransWarp, and repackages them as an enterprise software toolkit.
       Where TransWarp emphasized techniques like generative programming and
       aspect-oriented programming, PEAK emphasizes enterprise applications,
       and hides the computer science stuff "under the hood", so you can focus
       on building your application.
   
       PEAK tools can be used with other "Python Enterprise" frameworks such as
       Zope and the Python DBAPI to construct web-based, GUI, or command-line
       applications, interacting with any kind of storage, or with no storage
       at all.  Whatever the application type, PEAK can help you put it together.
   
  Please see http://www.zope.org/Members/pje/Wikis/TransWarp for tutorials,   Installation Instructions
  FAQs, package layout, etc.  Selected pages from the Wiki are included  
  in the docs/ directory for your convenience.  
   
  At this time, the 'Features', 'Aspects', 'SOX', and 'tests' modules  
  and packages are usable, if not necessarily full-featured.  All other  
  modules/packages (except as imported by the above) are under heavy  
  construction - don't enter without a hard hat!  (That is,  
  use them at your own risk.  Although, there's no warranty that any of  
  the other stuff works, beyond the fact that the tests run on my  
  home computer.)  
   
   
  INSTALLATION INSTRUCTIONS  
   
   To use this package, you will need to install it by placing the TW/ directory  
   inside a directory which is listed in your Python path.  You will also need  
   to install Aaron Watters' "kjbuckets" library, and Jim Fulton's "Scarecrow"  
   Interfaces package.  
   
   Later versions will hopefully automate some of this as we learn to use the  
   Python distutils, but for now, you have to do everything yourself...  
   
   Getting and Installing kjbuckets  
   
    Unix Platforms  
   
     You can download the C source code for kjbuckets at:  
   
      http://www.chordate.com/kjbuckets/  
   
     And build as you would any other Python module.  
   
    Windows Platforms  
   
     If you're using Python 1.5, you can download a pre-built kjbuckets.pyd at:  
   
      http://www.chordate.com/kwParsing/kjbuckets.pyd  
   
     And then place it in your Python path.  If you're using Python 2.0,  
     or can't get this to work, see "If You Can't Compile kjbuckets" below.  
   
    If You Can't Compile kjbuckets  
   
     If for whatever reason you can't get the C version of kjbuckets to work  
     on your system, download this file:  
   
      http://www.chordate.com/kwParsing/kjbuckets0.py  
   
     Rename it to "kjbuckets.py", and place it in your Python path.  This runs  
     slower than the C version, but it'll do in a pinch.  
   
   Getting and Installing the Interface package (aka "The Scarecrow")  
   
    The easiest place to find a copy of the Interface package is in the  
    lib/python directory of a Zope installation.  Just copy it over to your  
    main Python path, or add the lib/python directory to your Python path.  
   
    Failing that, you can look at this URL for downloads:  
   
     http://www.zope.org/Members/michel/Products/Interfaces/  
   
    This may not be the best place for an up-to-date copy, but the current  
    TransWarp code doesn't do much with interfaces yet, so it probably doesn't  
    matter right now.  
   
   
  TESTING YOUR INSTALLATION  
   
   TransWarp comes with a fairly hefty built-in test suite.  If you have  
   the Python "unittest" module installed in your Python path, you can use  
   it to run the test suites, like this::  
   
    python unittest.py TW.tests.suite  
   
   This will run about 114 tests on various parts of TransWarp.  If you have  
   installed everything correctly, 100% of the tests should succeed.  If  
   you're missing any needed parts, you will probably experience a massive  
   number of failures and errors.  
   
       See the INSTALL.txt file.


Generate output suitable for use with a patch program
Legend:
Removed from v.276  
changed lines
  Added in v.525

cvs-admin@eby-sarna.com

Powered by ViewCVS 1.0-dev

ViewCVS and CVS Help