Re: [Tutor] Changing the Attribute of a Variable

[wesley chun]

>> Initally, a variable. self.stop_time is created as type datetime.time,
>> with the default value 06:00:00, a time stamp, during entry into the
>> mainloop. self.stop_time = datetime.time(10,10,10). The user reads his
>> configuration file with the time stamp value of 08:00:00. self.time_stop
>> becomes type string. He then goes to a dialog to change time stop, and it
>> tries to access self.time_stop, and find is needs to have the attribute
>> strftime, as in:
>> set_loc_dict[ "stop_time" ] = self.stop_time.strftime("%H:%M:%S")
>> When reading the configuration file, how do I make sure that
>> self.time_stop is really has the strftime attribute?
>
> You already received an answer to your immediate question, but I wanted to
> clarify: "strftime" is NOT an attribute, it's a method.  Calling the
> strftime method of a time object returns a string, formatted according to
> the pattern you specify - so what you're storing as "stop_time" is not a
> time, but a string.


this is a reply to both:

wayne: the initial setting of self.stop_time as a datetime.time object
seems to be ok, but i'm uncomfortable with the fact that after "[the]
user reads his [config] file," it becomes a str. i think that the code
that processes the config file should be setting self.stop_time as
another datetime.time object, so that way, when you want to set the
value for set_loc_dict, it would not have any problems calling its
strftime() method.

marc: i will slightly disagree with you with regards to strftime *not*
being an attribute. it *is* an attribute, just not a *data
attribute*... i call it a "function attribute," but that's just
terminology. any time you have an object x with an attribute y, the
fact that you can issue x.y means that y is indeed an attribute of x.
if it's a data attribute, you access it with x.y. if it's a function
attribute, i.e., a method, you also access it with x.y, esp. if you
want to pass the function object around, and finally, if you actually
want to *execute* it *and* it's a method, then you add the parens,
x.y().

hope this helps,
-- wesley
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
"Python Web Development with Django", Addison Wesley, (c) 2009
http://withdjango.com

wesley.j.chun :: wescpy-at-gmail.com
python training and technical consulting
cyberweb.consulting : silicon valley, ca
http://cyberwebconsulting.com
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Add readline capabilities to a Python build 2.6 on Ubuntu

[spir]

Le Mon, 16 Feb 2009 22:34:23 -0700,
Eric Dorsey  a écrit :

> Greetings Tutor:
> I've managed to install Python 2.6 on my Ubuntu VM from source, however, it
> looks as though I missed something important along the way. My 2.6
> interpreter does not have readline support (example: I cant hit up arrow to
> repeat the last command) Is there a way to add this functionality now?
> 
> Or alternative, if I just rebuild it, does anyone know the flag or verbage
> to get readline support in?

As far as I know, it's built-in by default. But I have 2.5.2, not 2.6. Have you 
tried to import it, just to check?

s...@o:~/prog/io$ python
Python 2.5.2 (r252:60911, Oct  5 2008, 19:24:49) 
[GCC 4.3.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import readline
>>> readline


denis
--
la vida e estranya
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] exec "self.abc=22" ?

[Andre Engels]

On Mon, Feb 16, 2009 at 10:01 PM, Wayne Watson
 wrote:
> My limited repertoire. Actually, there wasn't much of a traceback. It came
> up in a small OK dialog. I copied what I could.  I see my image I used above
> did make it to the list, so here's the skinny.
>
>
> I see Marc covered it with setattr. How does one do it with a dictionary?
> What else lurks out there that might be useful along these lines?

It all depends on what you will use it for. As said, exec should work,
but it usually is not the way to go - if there's outside input
involved, it's _extremely_ unsafe, if everything comes from inside
your program it's an ugly sledgehammer to crack a nut. So please take
one step back - WHY do you want to do this? Where does this string
"self.abc = 22" come from? What are the values it can have? Can you
create a toy example that shows the problem you want to solve?

--
André Engels, andreeng...@gmail.com
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Marc Tompkins]

On Tue, Feb 17, 2009 at 12:16 AM, wesley chun  wrote:

> marc: i will slightly disagree with you with regards to strftime *not*
> being an attribute. it *is* an attribute, just not a *data
> attribute*... i call it a "function attribute," but that's just
> terminology. any time you have an object x with an attribute y, the
> fact that you can issue x.y means that y is indeed an attribute of x.
> if it's a data attribute, you access it with x.y. if it's a function
> attribute, i.e., a method, you also access it with x.y, esp. if you
> want to pass the function object around, and finally, if you actually
> want to *execute* it *and* it's a method, then you add the parens,
> x.y().
>

Which follows from the fact that in Python, functions are objects too.
However, in the context of the OP's question, I think that if he referenced
strftime with no parentheses the result would NOT be what he expected or
intended; the distinction between methods and what you call "data
attributes" can be an important one.  Point taken, though.

-- 
www.fsrtechnologies.com
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] urllib unquote

[Sander Sweers]

On Tue, Feb 17, 2009 at 08:54, Norman Khine  wrote:
> Thank you, but is it possible to get the original string from this?

You mean something like this?

>>> urllib.quote('hL/FGNS40fjoTnp2zIqq73reK60=\n')
'hL/FGNS40fjoTnp2zIqq73reK60%3D%0A'

Greets
Sander
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] urllib unquote

[Senthil Kumaran]

On Tue, Feb 17, 2009 at 1:24 PM, Norman Khine  wrote:
> Thank you, but is it possible to get the original string from this?

What do you mean by the original string Norman?
Look at these definitions:

Quoted String:

In the different parts of the URL, there are set of characters, for
e.g. space character in path, that must be quoted, which means
converted to a different form so that url is understood by the
program.
So ' ' is quoted to %20.

Unquoted String:

When %20 comes in the URL, humans need it unquoted so that we can understand it.


What do you mean by original string?
Why are you doing base64 encoding?
And what are you trying to achieve?

Perhaps these can help us to help you better?



-- 
-- 
Senthil
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Add readline capabilities to a Python build 2.6 on Ubuntu

[زياد بن عبدالعزيز البا تلي]

On Mon, 16 Feb 2009 22:34:23 -0700
Eric Dorsey  wrote:

> Greetings Tutor:
> I've managed to install Python 2.6 on my Ubuntu VM from source,
> however, it looks as though I missed something important along the
> way. My 2.6 interpreter does not have readline support (example: I
> cant hit up arrow to repeat the last command) Is there a way to add
> this functionality now?
> 
> Or alternative, if I just rebuild it, does anyone know the flag or
> verbage to get readline support in?
I think you need to recompile it again, this time make sure that
"libreadline5-dev" package is installed.

Hope that help.
Ziyad.
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Add readline capabilities to a Python build 2.6 on Ubuntu

[Genevi�ve DIAGORN]

Bonjour,
Je suis absente jusqu'au 22/02/09 inclus.
Cordialement.

Geneviève 

___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] exec "self.abc=22" ?

[Kent Johnson]

On Mon, Feb 16, 2009 at 4:01 PM, Wayne Watson
 wrote:

> I see Marc covered it with setattr. How does one do it with a dictionary?

Instead of trying to create variables with variable names, use the
names as keys in a dict. So instead of
  exec "self.abc=22"
you might use
  self.values = {}
  self.values['abc'] = 22

Kent
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] urllib unquote

[Kent Johnson]

On Mon, Feb 16, 2009 at 8:12 AM, Norman Khine  wrote:
> Hello,
> Can someone point me in the right direction. I would like to return the
> string for the following:
>
> Type "help", "copyright", "credits" or "license" for more information.
 import base64, urllib
 data = 'hL/FGNS40fjoTnp2zIqq73reK60%3D%0A'
 data = urllib.unquote(data)
 print base64.decodestring(data)
> ???Ը???Nzv̊??z?+?

>
> What am I missing?

How is data created? Since it doesn't decode as you expect, either it
isn't base64 or there is some other processing needed. Do you have an
example of a data string where you know the desired decoded value?

Kent
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Wayne Watson]

Note that the diagnostic output in the image shows attributeError: 'str' 
object has no attribute 'strftime'.


Let me see if I clarify what's really going on by including some of the 
code.


In Sentinel_GUI, the mainloop, the code shows (hard coded default values):
  ...
   self.slowdown = 1
   self.stop_time = datetime.time(6,0,0) <<---
   self.start_time  = datetime.time(18,0,0)

The code for OpenConfigFile is below, which reads the text file 
containing the configuration variables and values.


===Sample of config file==
Sentinel NC Configuration File Sentinel User 3 - 1/3/2009 (Meteor Software)
config_file_name=Initial.sen
mask_file_name=*none*
gray_scale=True
post_event_stack=False
post_event_format=Tiff 2
show_real_time=False
hourly_rate=12
slowdown=1
start_time=22:00:00
stop_time=06:00:00
lat=40.0
...
===end of sample==
Note below that I'm trying to pick off "user" variables in the file that 
are dates. self.config_var_list contains DAT (user) type for stop_time. 
Note the use of setattr at the bottom. I may have gone wrong here, when 
the code handles the DAT differently than the other user variables. See 
the code after this, where the error is raised when the dialog begins to 
be activated. There are a few more comments below this.

==OpenConfigFile==
   def OpenConfigFile(self):
   def time_tuple(tstring):
   t = tstring.split(':')
   print 'here is t', t
   tnum = ()
   for j in range(0,len(t)):
   tnum = tnum+(string.atoi(t[j]),)   
   return tnum


   print "OCFile entered"
   print
   config_file_name = askopenfilename( title="Open Configuration File",
   filetypes=CEN_FILE_TYPES )
   config_file=open(config_file_name,'r')
   first_line = config_file.readline()   # skip first line
   for (j,newline) in enumerate(config_file):
   aline = newline[:-1]
   aline.rstrip()
   (config_var, config_value) = aline.split('=')
   config_type = self.config_var_list[j][1][1]
   self_var = "self." + config_var
   print "ocf: ",config_var,config_value,"self_var=",self_var
   if config_var == DAT: # Date type, expecting hh:mm:ss
   t_ntup = time_tuple(config_value)
   stime = datetime.time(t_ntup[0],t_ntup[1],t_ntup[2])
   print "type stime: ",type(stime)
   # Date vars should be type datetime.time
   config_value = str(stime)
   else:
   self_var_assignment = self_var +'='+ config_value
   print "self_var_assignment", self_var_assignment
   abc = self
   # self_var_assignment
   setattr(self, config_var, config_value)
   config_file.close()
=End of OpenConfigFile===
When the program brings up the dialog, this piece of code gets me into 
trouble, as marked.

===OperationalSettings===
   def OperationalSettings(self):
   print "OSett self = ", self, "type =", type(self)
   print
   set_loc_dict = {}
   set_loc_dict[ "ok" ] = False
   set_loc_dict[ "color" ] = 2
   if self.gray_scale:
   set_loc_dict[ "color"] = 1
   print "gray scale now--wtw: ", self.gray_scale
   set_loc_dict[ "hourly_rate" ] = self.hourly_rate
   print "wtw self.stop_time", self.stop_time, type(self.stop_time)
   # set in GUI as datetime.time(6,0,0)
   # HEY wtw self.stop_time.strftime("%H:%M:%S")
   set_loc_dict[ "stop_time" ] = 
self.stop_time.strftime("%H:%M:%S")   <

Re: [Tutor] urllib unquote

[Norman Khine]

it is my error, the data is a sha string and it is not possible to get 
the string back, unless you use rainbowtables or something of the sort.


Kent Johnson wrote:

On Mon, Feb 16, 2009 at 8:12 AM, Norman Khine  wrote:

Hello,
Can someone point me in the right direction. I would like to return the
string for the following:

Type "help", "copyright", "credits" or "license" for more information.

import base64, urllib
data = 'hL/FGNS40fjoTnp2zIqq73reK60%3D%0A'
data = urllib.unquote(data)
print base64.decodestring(data)

???Ը???Nzv̊??z?+?
What am I missing?


How is data created? Since it doesn't decode as you expect, either it
isn't base64 or there is some other processing needed. Do you have an
example of a data string where you know the desired decoded value?

Kent


___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

[Tutor] Possible to change values of scalar function parameters?

[python]

Is there a way to change values of scalar function parameters? I
know you can change the values of parameters if the parameter is
a mutable type like a list, but is there a way to update the
value of scalar parameters of type integer or string?
Simple example: Is there a way to have the following function
update its changeme parameter in a 'call by reference' manner?
>>> def test1( changeme ):
changeme = 'Changed!'
>>> ref_value = 'blah'
>>> test1( ref_value )
>>> ref_value
'blah'
Thanks!
Malcolm
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Possible to change values of scalar function parameters?

[Kent Johnson]

On Tue, Feb 17, 2009 at 11:57 AM,   wrote:
> Is there a way to change values of scalar function parameters? I know you
> can change the values of parameters if the parameter is a mutable type like
> a list, but is there a way to update the value of scalar parameters of type
> integer or string?
>
> Simple example: Is there a way to have the following function update its
> changeme parameter in a 'call by reference' manner?
>
 def test1( changeme ):
> changeme = 'Changed!'

No, not a simple way at least. Possibly you can do it with hackery
involving stack frames but I wouldn't recommend that. Either pass the
values in some kind of container (list, dict, class instance) or
return the new value and assign it in the caller.

Kent
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Possible to change values of scalar function parameters?

[python]

Kent,

> No, not a simple way at least. Possibly you can do it with hackery
involving stack frames but I wouldn't recommend that. Either pass the
values in some kind of container (list, dict, class instance) or
return the new value and assign it in the caller.

That's what I thought. Thank you!

Malcolm
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Marc Tompkins]

On Tue, Feb 17, 2009 at 4:44 AM, Wayne Watson
wrote:

> Note that the diagnostic output in the image shows attributeError: 'str'
> object has no attribute 'strftime'.
>
> Let me see if I clarify what's really going on by including some of the
> code.
>

Everything in Python - both variables and code - is an object.  Objects have
attributes - data, basically - and methods - functions - associated with
them.  (As Wesley pointed out, since pieces of code are also objects,
methods are attributes too.)

However, not all objects have the same attributes or methods associated with
them!  datetime.time objects have a "strftime" method, which, when called,
returns a string representation of the time.  String objects do not have any
such method or data attribute, hence the error.

You're showing us both too much code and too little -
  print "wtw self.stop_time", self.stop_time, type(self.stop_time)
  # set in GUI as datetime.time(6,0,0)
  # HEY wtw self.stop_time.strftime("%H:%M:%S")
  set_loc_dict[ "stop_time" ] = self.stop_time.strftime("%H:%M:%S")
< from validate import Validator

...

> cfgFileName = os.getcwd() + os.sep + 'fsr_1500.ini'
> tmpStr = """
> npiXMLFile = string(default="npiMap.XML")
> UCFformLength = integer(min=50, max=80, default=66)
> FIformLength = integer(min=50, max=80, default=64)
> OutformLength = integer(min=50, max=80, default=64)
> IncludeLegacy = boolean(default=False)
> TopLeft = int_list(min=2, max=2)
> BottomRight = int_list(min=2, max=2)
> FIHist = string_list(default=None)
> UCFHist = string_list(default=None)
> OutHist = string_list(default=None)
> LastRunUCF = boolean(default=True)
> LastRunPrinter = boolean(default=False)
> detailLeft = integer(min=0, max=80, default=0)
> detailTo = integer(min=0, max=80, default=9)
> detailPOS = integer(min=0, max=80, default=19)
> detailCode = integer(min=0, max=80, default=25)
> detailMods = integer(min=0, max=80, default=32)
> detailDiags = integer(min=0, max=80, default=44)
> detailCharge = integer(min=0, max=80, default=49)
> detailUnits = integer(min=0, max=80, default=58)
> detailEPSDT = integer(min=0, max=80, default=62)
> detailEMG = integer(min=0, max=80, default=22)
> detailID = integer(min=0, max=80, default=67)
> bodyLeftBlock = integer(min=0, max=80, default=0)
> bodyMidBlock = integer(min=0, max=80, default=24)
> bodyRightBlock = integer(min=0, max=80, default=49)
> bodyLabelEdge = integer(min=0, max=80, default=40)
> ConfirmSuccess = boolean(default=True)
> """
> cfgSpec = StringIO.StringIO(tmpStr)
> cfgFile = ConfigObj(cfgFileName,
> configspec=cfgSpec, raise_errors=True,
> write_empty_values=True,
> create_empty=True, indent_type='', list_values=True)
> vtor = Validator()

...

> cfgFile['TopLeft'] = data.GetMarginTopLeft()  # writing a couple of
> values
> cfgFile['BottomRight'] = data.GetMarginBottomRight()
>
 ...

> test = cfgFile.validate(Global.vtor, copy=True)
> cfgFile.write()
>

Looking at that, I see a few things I want to clean up.  That's the danger
(and advantage) of exposing your own code to public scrutiny...

-- 
www.fsrtechnologies.com
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[wesley chun]

>   # Date vars should be type datetime.time
>   config_value = str(stime)

this is pretty much the offending code right here... in fact, the
comment contradicts the assignment. it says that date vars should be
of type datetime.time, yet it assigns a *string* to config_value,
which then gets set later using setattr().

if you just remove that assignment, things should work better..., at
least from what i can see.

good luck!
-- wesley
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
"Core Python Programming", Prentice Hall, (c)2007,2001
"Python Fundamentals", Prentice Hall, (c)2009
http://corepython.com

wesley.j.chun :: wescpy-at-gmail.com
python training and technical consulting
cyberweb.consulting : silicon valley, ca
http://cyberwebconsulting.com
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Possible to change values of scalar function parameters?

[Alan Gauld]

 wrote


No, not a simple way at least. Possibly you can do it with hackery
involving stack frames but I wouldn't recommend that. Either pass 
the

values in some kind of container (list, dict, class instance) or
return the new value and assign it in the caller.

That's what I thought. Thank you!


But don't forget that in python you can return multiple values
from a function...

def f():
return 42, 66, 101

a,b,c = f()

HTH,

--
Alan G
Author of the Learn to Program web site
http://www.alan-g.me.uk/


___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Possible to change values of scalar function parameters?

[python]

Alan,

> But don't forget that in python you can return multiple values from a 
> function.

Yes. Thank you!

Malcolm
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Wayne Watson]

Title: Signature.html




Yes, about public scrutiny of code and your paragraphs on
"Everything.." and "However, not all..."-- both  understood. 

Here's the print from the code line below.

Second line from the top.

Regarding, ConfigObj, I was aware of it when I decided to go this
route. That's the one that uses has an init file like Windows?  Rather
than have to go through a learning process on it (the Win init
module/object), and some uncertainty about it's acceptability in my
situation, I thought I'd be more exploratory and proceed as I have. So
far it has paid off in many learning dividends. ConfigObj, if not the
Win stuff, may be attractive. Anyway, I'd like to proceed for the
moment with this effort. 

Marc Tompkins wrote:

  On Tue, Feb 17, 2009 at 4:44 AM, Wayne
Watson 
wrote:
  Note
that the diagnostic output in the image shows attributeError: 'str'
object has no attribute 'strftime'.

Let me see if I clarify what's really going on by including some of the
code.
  
  
Everything in Python - both variables and code - is an object.  Objects
have attributes - data, basically - and methods - functions -
associated with them.  (As Wesley pointed out, since pieces of code are
also objects, methods are attributes too.)
  
However, not all objects have the same attributes or methods associated
with them!  datetime.time objects have a "strftime" method, which, when
called, returns a string representation of the time.  String objects do
not have any such method or data attribute, hence the error.
  
You're showing us both too much code and too little - 
      print "wtw self.stop_time", self.stop_time, type(self.stop_time)
      # set in GUI as datetime.time(6,0,0)
      # HEY wtw self.stop_time.strftime("%H:%M:%S")
      set_loc_dict[ "stop_time" ] = self.stop_time.strftime("%H:%M:%S")
  <

Re: [Tutor] Changing the Attribute of a Variable

[Kent Johnson]

On Tue, Feb 17, 2009 at 7:44 AM, Wayne Watson
 wrote:

> ==OpenConfigFile==
>   def OpenConfigFile(self):
>   def time_tuple(tstring):
>   t = tstring.split(':')
>   print 'here is t', t
>   tnum = ()
>   for j in range(0,len(t)):
>   tnum = tnum+(string.atoi(t[j]),) return tnum
>
>   print "OCFile entered"
>   print
>   config_file_name = askopenfilename( title="Open Configuration File",
>   filetypes=CEN_FILE_TYPES )
>   config_file=open(config_file_name,'r')
>   first_line = config_file.readline()   # skip first line
>   for (j,newline) in enumerate(config_file):
>   aline = newline[:-1]
>   aline.rstrip()
>   (config_var, config_value) = aline.split('=')
>   config_type = self.config_var_list[j][1][1]
>   self_var = "self." + config_var
>   print "ocf: ",config_var,config_value,"self_var=",self_var
>   if config_var == DAT: # Date type, expecting hh:mm:ss

What is DAT? Where is it defined? Why should the config variable name
(config_var) be equal to DAT? Is this code block being executed at
all?

Kent

>   t_ntup = time_tuple(config_value)
>   stime = datetime.time(t_ntup[0],t_ntup[1],t_ntup[2])
>   print "type stime: ",type(stime)
>   # Date vars should be type datetime.time
>   config_value = str(stime)
>   else:
>   self_var_assignment = self_var +'='+ config_value
>   print "self_var_assignment", self_var_assignment
>   abc = self
>   # self_var_assignment
>   setattr(self, config_var, config_value)
>   config_file.close()
___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Marc Tompkins]

On Tue, Feb 17, 2009 at 11:01 AM, Wayne Watson  wrote:

>  Here's the print from the code line below.
>
> Second line from the top.
>

There it is - stop_time is a str at this point in the program, so has no
strftime attribute or method.  Step back through your code and see why...  I
find Control-F very helpful in situations like this!


>
> Regarding, ConfigObj, I was aware of it when I decided to go this route.
> That's the one that uses has an init file like Windows?  Rather than have to
> go through a learning process on it (the Win init module/object), and some
> uncertainty about it's acceptability in my situation, I thought I'd be more
> exploratory and proceed as I have. So far it has paid off in many learning
> dividends. ConfigObj, if not the Win stuff, may be attractive. Anyway, I'd
> like to proceed for the moment with this effort.
>

There's nothing sacred about the ".ini" extension - any text file that
contains "variable = value" pairs is game.  ConfigObj is not
Windows-centric.  The advantage - which is what I was trying to show when I
posted that gosh-awful hunk of code - is that you can define the format of
the file in one central section of your program - it could even be a
separate module if you wanted - and in one swell foop you tell ConfigObj the
name of the variable, its type, acceptable values or range, and a default
value in case it's missing.  Opening, reading, closing, writing, validating
- all handled.

I certainly wouldn't want to dissuade you from writing your own as a
learning exercise - I'm glad I did - but each time I need to update one of
my old programs that still uses my homegrown ConfigFile, that's the first
thing I re-factor.

-- 
www.fsrtechnologies.com
<>___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Add readline capabilities to a Python build 2.6 on Ubuntu

[Lie Ryan]

On Mon, 16 Feb 2009 22:34:23 -0700, Eric Dorsey wrote:

> Greetings Tutor:
> I've managed to install Python 2.6 on my Ubuntu VM from source, however,
> it looks as though I missed something important along the way. My 2.6
> interpreter does not have readline support (example: I cant hit up arrow
> to repeat the last command) Is there a way to add this functionality
> now?

WORKSFORME
I have Ubuntu and python2.6 and the up arrow history works fine.

___
Tutor maillist  -  Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor

Re: [Tutor] Changing the Attribute of a Variable

[Wayne Watson]

Yes, on F3 (ctrl-F). I use it a lot.  After the dust clears on how to
correct this implementation, I'll give ConfigObj more consideration. 
(I think a subsequent post I made with two images in it showing the
layout of the program in action got side tracked here. I got a msg from
the list, saying the moderator had to approve something--probably it. 
Whatever I said in it might be somewhat irrelevant here. So I continue.
See whoops at end!)

Ok, let's see how this works. I've defined this function and the
config_var_list. stop time is the last entry shown. It is part of the
constructor for mainloop, Sentinel_GUI. and loads sdict as shown.
    def Set_ConfigDictionary():

    config_var_list = (['config_file_name', 
['Initial.sen', STR]],    
    ['mask_file_name',  ['xyz', STR]],
                    ...
    ['stop_time',  ['datetime.time(6, 0, 0)', DAT]],
                        ...
                )
    # load sdict
    sdict = {}
    for j in range(0, len(config_var_list)):
    #    print "j: ", j, "str part: ",
str(config_var_list[j][0]), config_var_list[j][1]
    sdict[str(config_var_list[j][0])] =
config_var_list[j][1][0]
                      ...
    self.sdict = sdict
    self.config_var_list = config_var_list
                      ...
sdict and config_var_list become global to Sentinel_GUI. The first
index of conf_var_list maintains order for sdict. That is, when I want
to output the config file, I use to to fetch from sdict what I need,
which is a value and the "user" type, i.e., STR, DAT, BOO, etc. 

Now in SaveConfigFile, I go merrily along thusly:

                   ...
    # SAVE CONFIG FILE
    items = self.sdict.keys()
    items.sort()
    for (j, conf_entry) in enumerate(self.config_var_list):
    varName = conf_entry[0]
    varType = self.config_var_list[j][1][1]
    # Note, date-time vars are in hh:mm:ss
    varValue = eval('self.' + varName)
    var_assignment = varName + "=" + str(varValue)   
<<--- Beep, beep
    config_file.write(var_assignment + "\n")
                         ...

"Beep, beep" shows the likely problem.  I had coded this loop earlier to
try (and not succeeding) to take into consideration DAT, BOO, ...,
etc, but put aside how to handle the var_assignment statement (Note,
Strictly speaking I should be using TIM and not DAT, for clarity.).  
Here, I suspect that I should put out something like datetime.time(11,
20, 10). The time being 11:00:10. Instead I chose, to put 11:00:10 in
the file. I'm still a bit confused about what the file entry should be
here. However!

I think I may have drifted into the realm of :          :-)   
    UCFformLength = integer(min=50, max=80, default=66)
    FIformLength = integer(min=50, max=80, default=64) 
Maybe things are looking up for ConfigObj!!!

Whoops footnote: I think I got you and Kent confused,
but I think I'm on track here with thinking of the right person. I
posted the two pix to him. 

Marc Tompkins wrote:

  On Tue, Feb 17, 2009 at 11:01 AM, Wayne
Watson 
wrote:
  

Here's the print from the code line below.

Second line from the top.


  
  
There it is - stop_time is a str at this point in the program, so has
no strftime attribute or method.  Step back through your code and see
why...  I find Control-F very helpful in situations like this!
 
  
  


Regarding, ConfigObj, I was aware of it when I decided to go this
route. That's the one that uses has an init file like Windows?  Rather
than have to go through a learning process on it (the Win init
module/object), and some uncertainty about it's acceptability in my
situation, I thought I'd be more exploratory and proceed as I have. So
far it has paid off in many learning dividends. ConfigObj, if not the
Win stuff, may be attractive. Anyway, I'd like to proceed for the
moment with this effort. 


  
  
There's nothing sacred about the ".ini" extension - any text file that
contains "variable = value" pairs is game.  ConfigObj is not
Windows-centric.  The advantage - which is what I was trying to show
when I posted that gosh-awful hunk of code - is that you can define the
format of the file in one central section of your program - it could
even be a separate module if you wanted - and in one swell foop you
tell ConfigObj the name of the variable, its type, acceptable values or
range, and a default value in case it's missing.  Opening, reading,
closing, writing, validating - all handled.  
  
I certainly wouldn't want to dissuade you from writing your own as a
learning exercise - I'm glad I did - but each time I need to update one
of my old programs that still uses my homegrown ConfigFile, that's the
first thing I re-factor.
  
  
  
-- 
  www.fsrtechnologies.com
  

___
Tut