How to create a flat dictionary in Python? - python-2.7

Question

I have a text which I want to convert into a dictionary.
Here's the format of the text :
Apple 0
orange 5:
text1 : random text
text2 : random text
text3 : random text
text4 : random text
orange 6:
text1 : random text
text2 : random text
text3 : random text
text4 : random text
Apple 1
orange 12:
text1 : random text
text2 : random text
text3 : random text
text4 : random text
orange 13:
text1 : random text
text2 : random text
text3 : random text
text4 : random text
I want to convert the dictionary something like this :
dic_text = {'apple-0-orange-5-text1' : 'random text','apple-0-orange-5-text2' : 'random text','apple-0-orange-5-text3' : 'random text','apple-0-orange-5-text4' : 'random text','apple-0-orange-6-text1' : 'random text','apple-0-orange-6-text2' : 'random text','apple-0-orange-6-text3' : 'random text','apple-0-orange-6-text4' : 'random text','apple-1-orange-12-text1' : 'random text','apple-1-orange-12-text2' : 'random text','apple-1-orange-12-text3' : 'random text','apple-1-orange-12-text4' : 'random text','apple-1-orange-13-text1' : 'random text','apple-1-orange-13-text2' : 'random text','apple-1-orange-13-text3' : 'random text','apple-1-orange-13-text4' : 'random text'}
Can anyone tell me a generic way of making a dictionary something like above?

Answer 1

Assuming the following information that you did not provide (please edit the question clarifying if this holds or not):
That all the elements are on separate lines
That all the elements take at most one line (so random text does not span multiple lines)
That you want the keys in lowercase
That you do not want to preserve the whitespace at beginning/end of the keys and random text
random text cannot be just whitespace
The "Apple X" line does not contain a :
The "orange Y" line is the only kind of line that ends in : (plus eventually whitespace), so random text cannot end in :.
After an "Apple X" line there is always an "orange Y" line (possibly after some empty lines).
Then you can do something like this:
def build_dict(iterable):
result = {}
main_key = None
sub_key = None
for line in iterable:
# remove whitespace at beginning/end of line
line = line.strip()
if not line:
# throw away empty lines
continue
elif ':' not in line:
# we found an "Apple X" line, transform that into apple-X
main_key = '-'.join(line.lower().split())
sub_key = None
elif line[-1] == ':':
# we found an "orange X" line
sub_key = '-'.join(line.lower().split())
else:
# add a `textX : random_text` element
key, value = line.split(':')
result['-'.join([main_key, sub_key, key.strip()])] = value.strip()
return result
So you keep track of which Apple X value is in the main_key, and which orange Y value is in the sub_key and after that all lines text X : random_text are splitted on : and the three keys are combined and the value is saved in the dictionary.
If the assumptions I made do not hold then you have to handle things like multiline values etc, which depends on exactly the format of the file.