Idf_MSequence - Syntactic Sugar work

Underlying Data structure of again

Let us open a small idf file and look at the underlying data structure.

# assume we have open an IDF file called idf
# let us add three construction objects to it
idf.newidfobject('construction'.upper(), Name='C1')
idf.newidfobject('construction'.upper(), Name='C2')
idf.newidfobject('construction'.upper(), Name='C3')
constructions = idf.idfobjects['construction'.upper()]
print constructions

[
CONSTRUCTION,
    C1;                       !- Name
,
CONSTRUCTION,
    C2;                       !- Name
,
CONSTRUCTION,
    C3;                       !- Name
]

We know that constructions us just syntactic sugar around the underlying data structure. Let us call the underlying data structure real_constructions

# set real_constructions
real_constructions = = idf.model.dt['construction'.upper()]
print real_constructions

[['CONSTRUCTION', 'C1'], ['CONSTRUCTION', 'C2'], ['CONSTRUCTION', 'C3']]

real_constructions -> the underlying data structure
constructions -> syntactic sugar for real_constructions

So any changes made in constructions should reflected in constructions. Let us test this out.

constructions[0].Name = 'New C1'
print constructions

[
CONSTRUCTION,
    New C1;                   !- Name
,
CONSTRUCTION,
    C2;                       !- Name
,
CONSTRUCTION,
    C3;                       !- Name
]

print real_constructions

[['CONSTRUCTION', 'New C1'], ['CONSTRUCTION', 'C2'], ['CONSTRUCTION', 'C3']]

Even though we made the change only in constructions, we can see the changes in both constructions and real_constructions. Ep_Bunch takes care of this for us.

print 'type for constructions', type(constructions)

type for constructions <type 'list'>

since constructions is a list, we can do all the list operations on it. Let us try some of them:

constructions.pop(0)

CONSTRUCTION,
    New C1;                   !- Name

print constructions

[
CONSTRUCTION,
    C2;                       !- Name
,
CONSTRUCTION,
    C3;                       !- Name
]

That makes sense. We poped the first item in the list and now we have only two items.

Is this change reflected in real_constructions ?

print real_constructions

[['CONSTRUCTION', 'New C1'], ['CONSTRUCTION', 'C2'], ['CONSTRUCTION', 'C3']]

Dammit !! Why not ?

We still have 3 items in real_constructions and 2 items in constructions

print 'type for constructions', type(constructions)
print 'id of constructions', id(constructions)
print 'type for real_constructions', type(constructions)
print 'id of real_constructions', id(real_constructions)

type for constructions <type 'list'>
id of constructions 4576898440
type for real_constructions <type 'list'>
id of real_constructions 4535436208
  • Both constructions and real_constructions are lists.
  • But looking at their ids, it is clear that they are two different lists.
  • poping an item in one list will not pop it in the other list :-(
  • In constructions[0].Name = "New C1" we see changes to an item within constructions is reflected within real_constructions
  • EpBunch takes care of that connection
  • We are having problems with the list functions.
  • we see that pop() does not work for us
  • similarly the results of append(), insert(), sort() and reverse() in constructions will not be reflected in real_constructions

This is how it works in eppy version 0.5

We need to fix this. Now we describe how this problem was fixed.

constructions should be a list-like wrapper around real_constructions. Python has an excellent data structure called collections.MutableSequence that works perfectly for this. Alex Martelli has a great discussion of this in this stackoverflow thread Overriding append method after inheriting from a Python List

  • So we make a class eppy.idf_msequence.Idf_MSequence that inherits form collections.MutableSequence
  • constructions is now an instance of eppy.idf_msequence.Idf_MSequence
  • reading the above stackoverflow thread and the code wihtin eppy.idf_msequence.Idf_MSequence should show you how it works
  • version of eppy higher than 0.5 will use eppy.idf_msequence.Idf_MSequence

Let us take a look at how it works (in all versions of eppy newer than 0.5):

# using eppy version greater than 0.5
import sys
# pathnameto_eppy = 'c:/eppy'
pathnameto_eppy = '../../../'
sys.path.append(pathnameto_eppy)
from eppy import modeleditor
from eppy.modeleditor import IDF
iddfile = "../../../eppy/resources/iddfiles/Energy+V7_2_0.idd"
fname1 = "../../../eppy/resources/idffiles/V_7_2/smallfile.idf"
IDF.setiddname(iddfile)
idf = IDF(fname1)

idf.newidfobject('construction'.upper(), Name='C1')
idf.newidfobject('construction'.upper(), Name='C2')
idf.newidfobject('construction'.upper(), Name='C3')
constructions = idf.idfobjects['construction'.upper()]

print constructions

[
CONSTRUCTION,
    C1;                       !- Name
,
CONSTRUCTION,
    C2;                       !- Name
,
CONSTRUCTION,
    C3;                       !- Name
]

real_constructions = idf.model.dt['construction'.upper()]
print real_constructions

[['CONSTRUCTION', 'C1'], ['CONSTRUCTION', 'C2'], ['CONSTRUCTION', 'C3']]

Shall we test pop(0) here ?

constructions.pop(0)

CONSTRUCTION,
    C1;                       !- Name

print constructions

[
CONSTRUCTION,
    C2;                       !- Name
,
CONSTRUCTION,
    C3;                       !- Name
]

print real_constructions

[['CONSTRUCTION', 'C2'], ['CONSTRUCTION', 'C3']]

Awesome !!! both constructions and real_constructions have the same number of items

print type(constructions)
print type(real_constructions)

<class 'eppy.idf_msequence.Idf_MSequence'>
<type 'list'>

what kind of sorcery is this. How did that work. How does Idf.Msequence do this magic ? Let us look at that link in stackoverflow. The question raised in stackovverflow is:

I want to create a list that can only accept certain types. As such, I’m trying to inherit from a list in Python, and overriding the append() method like so: and there is a sample code after this.

Alex Martelli responds:

Not the best approach! Python lists have so many mutating methods that you’d have to be overriding a bunch (and would probably forget some).

Rather, wrap a list, inherit from collections.MutableSequence, and add your checks at the very few “choke point” methods on which MutableSequence relies to implement all others. Alex’s code follows after this point. In eppy.idf_msequence I have included Alex’s code.

Stop here and read through the stackoverflow link

Well ... you don’t really have to. It does go off on some tangents unrelated to what we do in eppy.

The strategy in eppy.idf_msequence.Idf_MSequence is to have two lists, list1 and list2. To play with this I made a simple class TwoLists. Here TwoLists acts just like a list. Any operation list operation on TwoLists will result in a similar operation on both list1 and list2. TwoLists is not used in eppy, I simply use it to flesh out how MutableSequence can be used. I am going to play with TwoLists here to show you how cool it is :-)

from eppy.idf_msequence import TwoLists
twolists = TwoLists()
print twolists

list1 = [], list2 = []

twolists.append(5)
print twolists

list1 = [5], list2 = ['r_5']

twolists.append(dict(a=15))
print twolists

list1 = [5, {'a': 15}], list2 = ['r_5', "r_{'a': 15}"]

twolists.insert(1, 42)
print twolists

list1 = [5, 42, {'a': 15}], list2 = ['r_5', 'r_42', "r_{'a': 15}"]

twolists.pop(-1)

{'a': 15}

print twolists

list1 = [5, 42], list2 = ['r_5', 'r_42']

Isn’t that neat !! Idf_MSequence works in a similar way. Out of sheer laziness I am going to let you figure it out on your own. (ignore Idf_MSequence_old, since that went in the wrong direction)