[Tutor] 'source' in python or preloading variables

[Aditya Lal]

On 10/27/07, John <washakie at gmail.com> wrote:
>
> Ok, so trying to convert it to a function:
>
> def source(ifile, *args):
>         """ takes a file object and a list of variables as input.
>             assumes, source file has variables in format:
>             VAR[i]=variable """
>
>         import re
>         # Read the file
>         lines = fd.readlines()
>         fd.close()
>
>         # Make it python friendly: put all values in 'single quotes'
>         cmd = '\n'.join([re.sub( r'^([^=]+)=(.*)$', r"\1='\2'", v) for v
> in lines])
>
>         # Create variables as list of appropriate size with default
> IM STUCK HERE... HOW DO I DYNAMICALLY CONVERT THE LIST OF VARIABLE NAMES
> TO VARIABLES OF TYPE LIST
>
>         # Execute the command and all your variables are populated ...
>         exec(cmd)
>
>
> Thanks,
>
>
>
>
>
> > Thanks! This is helpful.. I like the RE approach as it's conceivable to
> > write a function...
> >
> > On 10/27/07, Aditya Lal <aditya.n.lal at gmail.com > wrote:
> > >
> > > You can source the file in python provided that you make it python
> > > friendly:-
> > >
> > > STN_id[1]='AAA' instead of STN_id[1]=AAA
> > > ...
> > >
> > >
> > > -----------
> > > import re
> > > # Read the file
> > > fd = open('sitelocations')
> > > lines = fd.readlines()
> > > fd.close()
> > >
> > >
> > > # Make it python friendly: put all values in 'single quotes'
> > > cmd = '\n'.join([re.sub( r'^([^=]+)=(.*)$', r"\1='\2'", v) for v in
> > > lines])
> > >
> > >
> > > # Create variables as list of appropriate size with default values
> > > (assuming 0)
> > > max = 5
> > > STN_id, STNlat, STNlon, STNelv = [0]*max, [0]*max, [0]*max, [0]*max
> > >
> > >
> > > # Execute the command and all your variables are populated ...
> > > exec(cmd)
> > >
> > >
> > > -------------------
> > > Not sure if this is a better way :) But atleast it will make the file
> > > sitelocations order invariant - implies you can interchange the lines
> > > without breaking the logic.
> > >
> > >
> > > Caveat - list in python starts at 0 index so STN_id[1] is not the
> > > first element but the second element. But you can filter unwanted items from
> > > the list if you want as in -
> > > [ x for x in STN_id if x != 0 ]
> > >
> > >
> > > HTH
> > > Aditya
> > >
> > >  On 10/27/07, John < washakie at gmail.com > wrote:
> > >
> > > >  I have a file sitelocations:
> > > >
> > > > STN_id[1]=AAA
> > > > STNlat[1]=58.800000
> > > > STNlon[1]=17.400000
> > > > STNelv[1]=20
> > > > STN_id[2]=BBB
> > > > STNlat[2]=42.450000
> > > > STNlon[2]=25.583333
> > > > STNelv[2]=2925
> > > >
> > > > which in shell scripts I can simple 'source'. In Python I have to:
> > > > sitesFile=file('sitelocations','r')
> > > > sites=sitesFile.readlines()
> > > > i=0;
> > > > for l in sites:
> > > >         if i==0:
> > > >                 STN_id.append(l.split('=')[1].strip('\n')); i+=1;
> > > >         elif i==1:
> > > >                 STNlat.append(l.split('=')[1].strip('\n')); i+=1;
> > > >         elif i==2:
> > > >                 STNlon.append(l.split('=')[1].strip('\n')); i+=1;
> > > >         else:
> > > >                 STNelv.append(l.split('=')[1].strip('\n')); i=0;
> > > >
> > > > Is there a better way??
> > > >
> > > > Thanks!
> > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Configuration
> > > > ``````````````````````````
> > > > Plone 2.5.3-final,
> > > > CMF-1.6.4,
> > > > Zope (Zope 2.9.7-final, python 2.4.4, linux2),
> > > > Five 1.4.1,
> > > > Python 2.4.4 (#1, Jul 3 2007, 22:58:17) [GCC 4.1.1 20070105 (Red Hat
> > > > 4.1.1-51)],
> > > > PIL 1.1.6
> > > >
> > > > _______________________________________________
> > > > Tutor maillist  -   Tutor at python.org
> > > > http://mail.python.org/mailman<http://mail.python.org/mailman/listinfo/tutor>/listinfo/tutor<http://mail.python.org/mailman/listinfo/tutor>
> > > >
> > > >
> > >
> > >
> > > --
> > > Aditya
> > >
> >
> >
> >
> > --
> > Configuration
> > ``````````````````````````
> > Plone 2.5.3-final,
> > CMF-1.6.4 ,
> > Zope (Zope 2.9.7-final, python 2.4.4, linux2),
> > Five 1.4.1,
> > Python 2.4.4 (#1, Jul 3 2007, 22:58:17) [GCC 4.1.1 20070105 (Red Hat
> > 4.1.1-51)],
> > PIL 1.1.6
> >
>
>
>
> --
> Configuration
> ``````````````````````````
> Plone 2.5.3-final,
> CMF-1.6.4,
> Zope (Zope 2.9.7-final, python 2.4.4, linux2),
> Five 1.4.1,
> Python 2.4.4 (#1, Jul 3 2007, 22:58:17) [GCC 4.1.1 20070105 (Red Hat
> 4.1.1-51 )],
> PIL 1.1.6
>


Are you saying you don't know the name of the variables, eh ? For any access
you *must* know the name of the variable somewhere in the program
anyway. Frankly I feel its a lot simpler to just modify the file
'sitelocations' to make it python friendly and import it in the program -
thats what I typically do for most of my code generation programs.


Anyway, for dynamically defining the variables you can do the following
(untested) :

for line in open('sitelocations').readlines() :
  nline = re.sub(r'^([^=]+)=(.*)$', r"\1 = '\2'", line)
  try :
    exec(nline)
  except NameError :
    # define variable name
    vname = nline.split('=')[0]
    exec( 'global ' + vname + ';' + vname + ' = [0]*5')    # to make it
visible globally and set default value
    exec(nline)


-- 
Aditya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tutor/attachments/20071027/cca803be/attachment-0001.htm