Help with database planning

[Juliano]

Hello, everybody.

I'm a linguist with practical skills on computers/programming.

We've been working with an ontology at my department, and now I need
to create a GUI viewer for the flat file we have.
I tried to write an Ontology class which manages the data read and
parsed from the flat file, but it takes a relatively long time.
Besides, we have plans to set up a website for online display of said
ontology. So, I have been being pushed towards changing the basic plan
and build a DB so that data access will be faster and easier for both
the desktop GUI and the web app. Right now, I'm trying to work with
sqlite, since it can be used as a separate file for the GUI and as a
DB for Django (which may be the choice for the web interface).

I have been redaing some books on DBs but I kind of get stuck when it
comes to the normalization and the planning of the tables. The problem
is that we have basically four fields that can be arranged in a tree-
like structure. Eg:

Concept
  |- Slot
  |`-- Facet
  |  `-- Filler
  `- Slot
   `-- Facet
 `-- Filler
 `-- Filler
...

So, for ONE *concept*, we have, usually, MANY *slots*, each *slot* has
ONE *facet*, and each *facet* can have MORE THAN ONE *filler*.
Besides, some *slots* and *fillers* are themselves *concepts*,
creating a sort of recursive reference.


line_no conceptsslots   facets  fillers
--
0   ABANDON DEFINITION  VALUE   "to leave or desert something or
someone"
1   ABANDON IS-AVALUE   EXIT
2   ABANDON LEXEMAP-LEX "leave behind-V1"
3   ABANDON LEXEMAP-LEX abandon-V1
(...)
97420   ZULUDEFINITION  VALUE   "a language or dialect spoken in south
africa and others"
97421   ZULUINSTANCE-OF VALUE   OTHER-NIGER-KORDOFANIAN-LANGUAGE
97422   ZULULANGUAGE-OF INV LESOTHO
97423   ZULULANGUAGE-OF INV SOUTH-AFRICA


I tried to create index tables for concepts, slots, facets and
fillers, which gave me the following table:


line_no conceptsslots   facets  fillers
--
0   cn_0sl_00048fc_7fl_07349
1   cn_0cn_02605fc_7cn_01768
2   cn_0sl_00121fc_2fl_04329
3   cn_0sl_00121fc_2fl_15009
(...)
97420   cn_05429sl_00048fc_7fl_01340
97421   cn_05429cn_02493fc_7cn_03526
97422   cn_05429cn_02750fc_1cn_02816
97423   cn_05429cn_02750fc_1cn_04580


(cn_X from concept index, sl_X from slot index,
fc_X from facet index, fl_X from filler index.)

As you can see, only concepts and facets are populated by their own
type of data.
Whereas slots and fillers can be populated by their own types or by
concepts.

What would be a good way to create tables for this situation?
In fact, this is the first time I've ever tried to create a DB, so I'm
completely lost.

I'm looking forward to a reply...

Thank you very much,
Juliano
-- 
http://mail.python.org/mailman/listinfo/python-list

using os

[Juliano Freitas]

how can i get just the directories in other directorie without a files
using the os module in Python??


Juliano



-- 
http://mail.python.org/mailman/listinfo/python-list

remove child in DOM??

[Juliano Freitas]

Hello!!


how can i remove a child in DOM ?

for example, i have thi xml below:

  
  juliano 
  


and i want to remove the  element:

   
  



how can  i do this??

Juliano Freitas

-- 
http://mail.python.org/mailman/listinfo/python-list

decode php base64

[Peter juliano]

 Not long ago, my friend resignation of the company. He left a few app's in
php, some of those files are hashed with base64 code, he told me that those
files handle http request on the server and some security rules. Because
i'm migrating all these applications on django. I do not really care about
those files, but today just for playing around, i try to see the code, and
in a less of an hour can see all the code. I want to share the solution
with this mail list

http://pastebin.com/e6KJhAW8
-- 
http://mail.python.org/mailman/listinfo/python-list