PyGTK Notebook button_press_event connection

2006-01-24 Thread Luigi
Hi all!

I have an application that uses a gtk.Notebook to show the content of a
GUI. Each page of it has a gtk.Label with a text that explains the
content. Each page is added to the notebook with the method
append_page(child, tab_label=None), passing a gtk.Label instance as
tab_label variable.

Now I'd like to connect to the button_press_event of the gtk.Label a
function call. I've done something like this:



notebook = gtk.Notebook()
...
child = gtk.Frame()
...
label = gtk.Label('Any text')
label.connect('button_press_event', a_function)
...
notebook.append_page(child, label)



But the button_press_event event is not intercepted (nothing happens
when I click on the tab label).

Any idea to solve this question?

Thanks

Luigi

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


XML-RPC "filter"

2008-09-09 Thread Luigi
Dear all,

I'm writing an XML-RPC server which should be able to modify the
incoming request before dispatching it. In particular I wand to added
two fixed parameters to the method called: one is the client host
address, and the other is the user name provided as for Basic
Authentication (http://[EMAIL PROTECTED]).

To do this, at the present I've overwritten the do_POST method of
SimpleXMLRPCRequestHandler, including at a certain point this code:


data = ''.join(L)

params, method = xmlrpclib.loads(data)
user = "unknown"
if self.headers.has_key('Authorization'):
  # handle Basic authentication
  (enctype, encstr) =  self.headers.get('Authorization').split()
  user, password = base64.standard_b64decode(encstr).split(':')
params = list(params)
params.append(self.address_string())
params.append(user)
params = tuple(params)
data = xmlrpclib.dumps(params, methodname=method)

(I slightly modified it to make it more readable at mail level)

It works, but I don't really like it because it completely overwrites
the do_POST method that in the future Python releases is going to
change (I verified it). Do you know a better way to do this?

Thanks in advance.

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


Re: newbie questions

2006-03-15 Thread Luigi
Hi!

1. Documentation... sorry...
2. There are no standars, it depends on the apidoc tool you want to use
3. I don't know, but when you extract the api documentation (using
tools as epydoc or happydoc) you have all the information you need (I
think)

4. As your python code is a script code, you don't need a main()
function (like C): all the lines you write are directly executed. For
example a code:

---
#!/usr/bin/python
a=5
b=a+3
print b
---

is directly executed.

If you are writing a class:

---
#!/usr/bin/python
class Test:
  _init__(self):
self.name="My test"

t = Test()
print t.name


This code instances a class Test and prints the name attribute.

Note that if you import a file containing the code above as a python
module, the last 2 lines are executed. In order to avoid it you must
write:

---
#!/usr/bin/python
class Test:
  _init__(self):
self.name="My test"

if __name__ == "__main__":
  t = Test()
  print t.name


Like this you execute the last two lines only if you execute directly
your file with the python interpreter.

I hope this can help you.

Luigi

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


Capturing stdout without waiting for the process end

2006-04-03 Thread Luigi
Hi to all!

I'd like to execute an external program capturing the stdout/stderr
messages at "real-time". I mean that I don't want to wait for the end
of the process. If I write a code like this:

import os
import sys

class Runner:

def run(self, arg):
try:
fin, fout = os.popen4(arg)
self.outData = fout.readlines()
self.outStatus = fout.close()

except Exception, err:
print err

def printOut(self):
print "###"
print self.outStatus
print "###"
for line in self.outData:
print line


r = Runner()
r.run("ls /tmp")
r.printOut()

I can print out (in this case in the os.stdout, but it may be
elsewhere) the whole external program output only once it ends. How can
I do to intercept the external program output during the processing?

Thank you in advance

Luigi

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


Re: can I get the index number in for x in y loop?

2006-04-03 Thread Luigi
Try this:

>>> a='String'
>>> i=0
>>> for x in a:
...   print i, x
...   i+=1
...
0 S
1 t
2 r
3 i
4 n
5 g

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


Re: Capturing stdout without waiting for the process end

2006-04-03 Thread Luigi
I use Linux.

So, if I modify my sdtout behaviour, when I execute:

fin, fout = os.popen4(arg)

this is executed asyncronously? And how can I intercept a fout.write()
event?

The question is that I have a C program (by third part) that streams
the logs into the stderr and stdout devices. I need to create an
envelopment that captures the outputs and puts them in a file
generating log events (for a real-time view).

Thanks

Luigi

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


Re: Modeling life on Earth –- an object-oriented (Python?) challenge

2013-07-30 Thread Luigi Ponti
[...forgot to reply to the list...]

Dear David,

Thanks for your feedback -- you got right to the point:

...python would be more of a prototyping language, and later translated
> into another language for faster maneuvering of data
>

exactly! I was hoping that, since the modeling framework is conceptually
well developed (i.e., books, papers, analysis, etc. in 35+ years), most of
the work would be towards getting the code up to the same conceptual (i.e.,
abstraction) level. Hence, I was thinking Python would be a good tool for
that. Performance can be taken care of at a later stage, if needed.

Please do not hesitate to drop a further line.

Kind regards,

Luigi


On Tue, Jul 30, 2013 at 1:23 PM, David Hutto  wrote:

> Never used pascal, and python might not be the fastest way to implement a
> program such as this.
>
> In a previous discussion, this was taken place by someone using a predator
> prey brain class..
>
> The simulation will vary, until a full refinement of forecast is above a
> certainty percentage level.
>
> Visualization is needed as well.
>
> Collaboration is, of course
> , the best possible route. However you need to start with certain
> statistics, and know there will be an Uncerrtainty Principle rule applied.
>
> The algorithm for such massive amounts of data analysis in a simulation
> forecast, will involve HD space and RAM
> .
>
> You will also want to collaborate with certain databases in order to
> refine the accuracy of your models.
>
> This is kind of what I would consider being a Dune(Frank Herbert)
> planetary engineer. It also takes in other db data such as tagging marks of
> animals percentiles of bacterias/viruses/etcSO it's not as simple as it
> sounds, and python would be more of a prototyping language, and later
> translated into another language for faster maneuvering of data.
>
>
>
> On Tue, Jul 30, 2013 at 4:57 AM,  wrote:
>
>> Dear List,
>>
>> I have to start this email by saying that I have recently attended
>> EuroPython in Florence, and it was the best and better organized conference
>> I have ever attended in 14 years of international meetings.
>>
>> I apologize if this is off topic, but I read in the list's description
>> that “[p]retty much anything Python-related is fair game for discussion”.
>>
>> Although I am not a Python developer, I decided to attend EuroPython in
>> search for a programmer interested in collaborating in the Python project I
>> briefly describe below.
>>
>> I use ecosystem models implemented with a procedural paradigm in a
>> language different from Python (Pascal, for the records). I would like to
>> migrate these ecosystem models (and code) to an object-oriented paradigm
>> using Python, as I have come to believe its expressiveness would help a lot
>> get the semantics right, rather than simply split procedural code into
>> objects corresponding to ecological elements. What's more, our models use
>> physiological analogies among the different levels of the food chain or
>> web, and this makes them amenable to an even higher level of
>> object-oriented abstraction given adequate expressiveness.
>>
>> The goal is to go beyond the currently (mostly) formal implementation of
>> the object-oriented paradigm in ecological models. To do that, I would need
>> help from an expert Python programmer (who also has some math skills, knows
>> English, and can work in the Rome area, or at least central Italy). I need
>> help because I am a Python beginner with limited programming experience in
>> general, and hence my contribution will mainly be the ecosystem modeling
>> insight.
>>
>> At EuroPython, I gave a lightning talk about the project that can be
>> found on YouTube
>> http://youtu.be/iUNbgNuN0qY?t=31m50s
>>
>> As I already made some very promising contacts at EuroPyton with
>> developers that are interested and willing to help, and many people shared
>> their views and provided useful insight into the issue (thanks!), this post
>> is meant to get further feedback on my idea and possibly reach other
>> interested developers.
>>
>> Kindly contact me if you have any interest in the idea and time to devote
>> it, as it is becoming a funded project.
>>
>> Kind regards, thanks for any hint, and apologies for the many
>> inaccuracies,
>>
>> Luigi
>> --
>> http://mail.python.org/mailman/listinfo/python-list
>>
>
>
>
> --
> Best Regards,
> David Hutto
> *CEO:* *http://www.hitwebdevelopment.com*
>
-- 
http://mail.python.org/mailman/listinfo/python-list


Custom handler for logging

2008-09-03 Thread Luigi Paioro

Dear all,

concerning the logging module, I have written a new Handler which stores 
the logging messages in an SQLite database. It works well, but when I 
try to use it exploiting a configuration file I have a problem since the 
"class" entry does not accept a class which is not within the logging 
name-space, while the class I implemented is coded in my personal module 
(thus another name-space).


Is there a workaround to allow the usage of custom handlers?

Thanks in advance.

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


Re: Custom handler for logging

2008-09-03 Thread Luigi Paioro

Thanks, it works perfectly!

Luigi

Peter Otten ha scritto:

Luigi Paioro wrote:


concerning the logging module, I have written a new Handler which stores
the logging messages in an SQLite database. It works well, but when I
try to use it exploiting a configuration file I have a problem since the
"class" entry does not accept a class which is not within the logging
name-space, while the class I implemented is coded in my personal module
(thus another name-space).

Is there a workaround to allow the usage of custom handlers?


Maybe

http://mail.python.org/pipermail/python-list/2003-October/232762.html

Peter

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


Re: XML-RPC "filter"

2008-09-10 Thread luigi . paioro
On 9 Set, 17:55, "Diez B. Roggisch" <[EMAIL PROTECTED]> wrote:
> I would go for a slightly different approach: make your server have a
> dispatch-method that delegates the calls to the underlying actual
> implementation. But *before* that happens, extract the information as
> above, and either
>
>  - prepend it to the argument list
>
>  - stuff it into threadlocal variables, and only access these if needed in
> your implementation.
>
> Diez

Are you suggesting me to overwrite the _dispatch(self, method, params)
method of SimpleXMLRPCDispatcher? I thought to this possibility, but
it only accepts "method" and "params" as arguments, so, as far as I
know, I have no way to get the user and host address to append.

Perhaps I've misunderstood your suggestion... in that case can you
post a short example?

Thank you very much!

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


Re: XML-RPC "filter"

2008-09-12 Thread luigi . paioro
On 11 Set, 18:45, Richard Levasseur <[EMAIL PROTECTED]> wrote:
> Because he wants to insert parameters at the very start, he can
> probably get away with modifying the xml directly.  Just find the
> position of the  (i think thats the tag) and insert the xml
> you need after it.  Its pretty dirty, but would work.  The wire format
> isn't that complicated.

I think this is exactly what I do... isn't it?
--
http://mail.python.org/mailman/listinfo/python-list