jiahuiding - 8 months ago 45

Python Question

`{'a': [{'c','d'}, {'d'} ],`

'b': [{'c','d'}, set() ],

'c': [set(), {'a','b','d'}],

'd': [{'a','c'}, {'a','b'} ],

'e': [set(), set() ]}

class Graph:

def __init__(self,*args):

self.edges = {}

def degree(self,n):

count = 0

if n not in self.edges.keys():

raise GraphError

for key,value in self.edges.items():

if key == n:

for c in value:

for x in c:

count += 1

return count

def __iter__(self):

l=[]

for key,value in self.edges.items():

for item in value[0]:

l.append((key,item))

return sorted(l, key = degree(l[0]))

Define an

For example, iterating over the original Graph above produces 6 values in the following order: ('b', 'c'), ('b', 'd'), ('a', 'c'), ('a', 'd'), ('d', 'a'), and ('d', 'c'). This is the ordering, because node b has degree 2, so it and its two destination nodes, appearing alphabetically, are first; node a has degree 3, so it and its two destination nodes, appearing alphabetically, are next; and finally node d has degree 4, so it and its two destination nodes, appearing alphabetically, are last.

I am working in the iter function, when it calls

`g = Graph()`

g.edges = {'a': [{'c','d'},{'d'}], 'b': [{'c','d'}, set()], 'c': [set(),{'a','b','d'}], 'd': [{'a','c'},{'a','b'}], 'e': [set(),set()]}

it should return

`[t for t in g]-->[('b', 'c'), ('b', 'd'), ('a', 'c'), ('a', 'd'), ('d', 'a'), ('d', 'c')]`

but when I call me iter function, it produces the following error:

`96 *Error: [t for t in g] raised exception TypeError: iter() returned non-iterator of type 'list'`

99 *Error: [t for t in g] raised exception TypeError: iter() returned non-iterator of type 'list'

102 *Error: [t for t in g] raised exception TypeError: iter() returned non-iterator of type 'list'

can someone help me to fix my iter function?

Answer

You can use generator to implement `__iter__`

for `Graph`

. Then you just need to sort the vertices with tuple key `(degree, name)`

, iterate over them and `yield`

the edges in alphabetical order:

```
EDGES = {
'a': [{'c','d'}, {'d'} ],
'b': [{'c','d'}, set() ],
'c': [set(), {'a','b','d'}],
'd': [{'a','c'}, {'a','b'} ],
'e': [set(), set() ]
}
class Graph:
def __init__(self, edges):
self.edges = edges
def degree(self,n):
if n not in self.edges:
raise GraphError
return sum(len(s) for s in self.edges[n])
def __iter__(self):
# Iterate over vertices, primary key is degree and
# secondary is name
for v in sorted(self.edges, key=lambda x: (self.degree(x), x)):
# Iterate over outgoing edges in alphabetical order
for other in sorted(self.edges[v][0]):
yield v, other
g = Graph(EDGES)
print(list(iter(g)))
```

Output:

```
[('b', 'c'), ('b', 'd'), ('a', 'c'), ('a', 'd'), ('d', 'a'), ('d', 'c')]
```

Source (Stackoverflow)