Sun, 19 Sep 2010
/Coding/python/Django:
Django "Hello World" Example
(Lifted from [1].)
On my Debian system, first install Django:
apt-get install python-django
cd to where you want to put your new Django web application, and create a Django project:
django-admin.py startproject djangoHello
which will cause a subdirectory djangoHello to be created, containing manage.py, settings.py, and urls.py, among other things.
Now create an application within that project:
cd djangoHello
python manage.py startapp helloworld
which will create a helloworld subdirectory, containing models.py and views.py. Now edit djangoHello/helloworld/views.py
to add the following lines:
<![CDATA[
from django.http import HttpResponse
def helloworld(request):
return HttpResponse('Hello, World!')
]]>
Now edit djangoHello/urls.py, and add the line below after the line above (that will already be in the file):
#(r'^admin/(.*)', admin.site.root),
(r'^$', 'djangoHello.helloworld.views.helloworld'),
Make sure you are in the project root directory (djangoHello) and start the built-in Django server:
python manage.py runserver 0.0.0.0:8000
and point your browser at port 8000 (localhost:8000 if browser and Django are on the same machine) and observer your "Hello, World!".
[1] https://support.eapps.com/index.php?_m=knowledgebase&_a=viewarticle&kbarticleid=178
posted at: 05:01 | path: /Coding/python/Django | permanent link to this entry
Thu, 07 May 2009
/Coding/python:
Python: How to Put Some of Your Code in a Subdirectory
Splitting one's code up into logical sub-directories is a fairly normal
practice. This seems to be a little more complicated then usual with Python.
One uses the "import" command to access functions in other files /
directories. Of course Python automatically searches the system path and
your execution directory to satisfy an import command. But to pull in a
subdirectory one must place an (empty? Mine is....) __init__.py file in any
sub-directory containing Python code that one wants to access. Then the
directory name is added to the Python name space and is treated the same as
files and functions defined within files.
A concrete example is most definitely in order. Suppose I want to place
all my database connect code into a subdirectory called "db":
db$ ll
total 16
-rw------- 1 1489 2009-02-23 15:18 armor-mail-cert.pem
-rw------- 1 3243 2009-02-23 15:18 armor-mail-key.pem
-rw------- 1 1009 2009-02-23 14:48 armorMailSSL-cacert.pem
-rwxr-xr-x 1 436 2009-05-07 15:55 connectArmorMail.py
-rw-r--r-- 1 0 2009-05-07 16:33 __init__.py
My main code is executing one directory up. Any code there that wants
to access the connect function must contain:
import db.connectArmorMail
db = db.connectArmorMail.connect_armor_mail()
Note that we are importing db.connectArmorMail, *not*
db/connectArmorMail, and then the reference in the next line also uses a
".", not a "/". "db" is a directory, "connectArmorMail" is a file name, and
"connect_armor_mail" is a function within a file, but all are treated
equally in Python name space, separated by periods. db.connectArmorMail
contains:
def connect_armor_mail():
import MySQLdb
ssl_settings = {'ca': 'db/armorMailSSL-cacert.pem',
'cert': 'db/armor-mail-cert.pem',
'key': 'db/armor-mail-key.pem'
}
try:
db=MySQLdb.connect(host="www.armor-mail.com",
user="username",
passwd="password",
db="databasename",
ssl=ssl_settings
)
except:
print "ERROR: could not connect to database."
return db
Note that the files referenced in ssl_settings are in the same directory
as db.connectArmorMail, but they are referenced relative to the execution
directory (db/*).
posted at: 09:01 | path: /Coding/python | permanent link to this entry
Thu, 26 Feb 2009
/Coding/python:
Using Python to Connect to MySQL Over SSL
Assuming you have already setup your MySQL server to accept external SSL
connections[1], with the MySQLdb module and the help of a little
documentation[2][3] getting Python to talk to the MySQL server is fairly
straightforward:
<![CDATA[
#!/usr/bin/python
import MySQLdb
ssl_settings = {'ca': 'db/cacert.pem',
'cert': 'db/cert.pem',
'key': 'db/key.pem'
}
try:
db=MySQLdb.connect(host="www.yoursite.com",
user="SSLremote",
passwd="yourpassword",
db="databasename",
ssl=ssl_settings
)
except:
print "<p>ERROR: could not connect to database."
c=db.cursor()
c.execute("""SELECT * FROM thistablename""")
while True:
row = c.fetchone()
if row == None:
break
else:
print "<p>"
print row
]]>
Point your web browser at this piece of code (you probably have to name
it "something.cgi" and put it a folder where your web server is enabled for
cgi). This code will connect to the MySQL server and dump the contents of
table thistablename into your browser window. If you get "ERROR: could not
connect to database." then execute the same code in a terminal on the Python
interpreter command line to see the actual MySQL error returned.
Note that the path specification and permissions of the SSL files
specified in "ssl_settings" are quite sensitive, as the web server must have
access to them. I have stored the files with my code to allow a relative
path. For security, these files should be "chmod 600" with the owner set to
www-data (or whatever user name your webserver runs under....)
[1] http://blog.langex.net/index.cgi/Admin/MySQL/
[2] http://riskable.com/2009/02/12/how-to-use-ssl-with-the-python-mysqldb-module
[3] http://mysql-python.sourceforge.net/MySQLdb.html
posted at: 05:32 | path: /Coding/python | permanent link to this entry
