combine values of several objects into a single dictionary

Question

I am fetching these rows from db:

blog_id='12', field_name='title', translation='title12 in en', lang='en'
blog_id='12', field_name='desc', translation='desc12 in en', lang='en'

blog_id='13', field_name='title', translation='title13 in en', lang='en'
blog_id='13', field_name='desc', translation='desc13 in en', lang='en'
....

and I want to build a single dictionary for each blog_id: e.g.

[
   {'blog': '12', 'title': 'title12 in en', 'desc': 'desc12 in en'},
   {'blog': '13', 'title': 'title13 in en', 'desc': 'desc13 in en'},
   ....
]

I am trying this way:

res = []
dict_ = {}
for trans in translations: # 'translations' is QuerySet, already filtered by 'en'  
    if trans.blog_id in dict_.values():
        dict_[trans.field_name] = trans.translation
    else:
        dict_['blog'] = trans.blog_id
        dict_[trans.field_name] = trans.translation

    res.append(dict_)

but this is soo wrong, res contains here blog 13 3 times and blog 12 isnot even in the final list. I feel so dumb right now, what am I missing?

Answer 1

First, accumulate all the data corresponding to every blog_id in a dictionary, like this

groups = {}
for trans in translations:
  groups.setdefault(trans.blog_id, {})[trans.field_name] = trans.translation

Now, attach the corresponding blog_id to all the accumulated dictionaries,

for key in groups:
  groups[key]['blog'] = key

Now, just get all the values of the groups, with groups.values() to get the result.

Note: If you want to retain the original order of elements based on the blog_id, then instead of using a normal dictionary, use collections.OrderedDict, like this

from collections import OrderedDict
groups = OrderedDict()
...