I want to add string dictionary keys like this:
x = "%s-%s-%s %s:%s:00"%(dt.year,dt.month,dt.day,dt.hour,dt.minute)
dict[x] +=a1
But it gives me an error like this:
KeyError: '2015-11-26 8:47:00'
If I try print type(x) it prints str
But if i try this:
dict = {}
x = "abc"
dict[x] = 1
print dict
it print to this:
{'abc': 1}
I don't understand what is the difference.
First error is that you named your dictionary dict. That name's
already being used; it's the name of the dictionary type. Overwriting an
existing name like this is called "shadowing". Don't do it, it will mess
you up.
You're using +=. This implies that there's already a value associated
with the key, which can be incremented. If that key isn't in the dict
yet, you get a KeyError.
You probably want to set a default value of zero. This can be done in
various ways. The simplest is:
d[x] = d.get(x, 0) + a1
Also see the collections standard library, which has a defaultdict
type.
Related
Maybe it is ordinary issue regarding iterating thru a dict. Please find below imovel.txt file, whose content is as follows:
{'Andar': ['primeiro', 'segundo', 'terceiro'], 'Apto': ['101','201','301']}
As you can see this is not a ordinary dictionary, with a key value pair; but a key with a list as key and another list as value
My code is:
#/usr/bin/python
def load_dict_from_file():
f = open('../txt/imovel.txt','r')
data=f.read()
f.close()
return eval(data)
thisdict = load_dict_from_file()
for key,value in thisdict.items():
print(value)
and yields :
['primeiro', 'segundo', 'terceiro'] ['101', '201', '301']
I would like to print a key,value pair like
{'primeiro':'101, 'segundo':'201', 'terceiro':'301'}
Given such txt file above, is it possible?
You should use the builtin json module to parse but either way, you'll still have the same structure.
There are a few things you can do.
If you know both of the base key names('Andar' and 'Apto') you can do it as a one line dict comprehension by zipping the values together.
# what you'll get from the file
thisdict = {'Andar': ['primeiro', 'segundo', 'terceiro'], 'Apto': ['101','201','301']}
# One line dict comprehension
newdict = {key: value for key, value in zip(thisdict['Andar'], thisdict['Apto'])}
print(newdict)
If you don't know the names of the keys, you could call next on an iterator assuming they're the first 2 lists in your structure.
# what you'll get from the file
thisdict = {'Andar': ['primeiro', 'segundo', 'terceiro'], 'Apto': ['101','201','301']}
# create an iterator of the values since the keys are meaningless here
iterator = iter(thisdict.values())
# the first group of values are the keys
keys = next(iterator, None)
# and the second are the values
values = next(iterator, None)
# zip them together and have dict do the work for you
newdict = dict(zip(keys, values))
print(newdict)
As other folks have noted, that looks like JSON, and it'd probably be easier to parse it read through it as such. But if that's not an option for some reason, you can look through your dictionary this way if all of your lists at each key are the same length:
for i, res in enumerate(dict[list(dict)[0]]):
ith_values = [elem[i] for elem in dict.values()]
print(ith_values)
If they're all different lengths, then you'll need to put some logic to check for that and print a blank or do some error handling for looking past the end of the list.
I am trying to sort a list that contain in each index an integer and a string. like the one in the example.
I used sort() and split but I get always the wrong ordered that I expect
def takeSecond(elem):
return elem.split("|")[2]
list = ['|val1: 0|0','|val: 0|80','|val1 0|140','|val1: 0|20','|val1: 0|90']
list.sort(key=takeSecond)
print(list)
that returns:
['|val1: 0|90','|val: 0|80','|val1: 0|20','|val1: 0|0','|val1 0|140']
and I expect to get this:
['|val1: 0|140','|val: 0|90','|val1: 0|80','|val1: 20|0','|val1 0|0']
Where is my mistake in here?
Try this:
l = ['|val1: 0|0','|val: 0|80','|val1 0|140','|val1: 0|20','|val1: 0|90']
l.sort(key=lambda x:int(x.rsplit('|')[-1]), reverse=True)
This will sort your list based on what you need. and the expected output is:
In [18]: l
Out[18]: ['|val1 0|140', '|val1: 0|90', '|val: 0|80', '|val1: 0|20', '|val1: 0|0']
In addition note that:
Do not use list as a variable name. list is a built-in name in python, you will override its functionality .
I have a series of dictionaries which each contain the same keys but their values are different i.e Age in dictionary 1 = 2, Age in dictionary 2 = 4 etc etc but they are broadly identical in structure.
what I would like to do is to randomly select one of these dictionaries and then assign specific values with the dictionary to variables. i.e python randomly chooses Dictionary 1 and then I then want to fill the dictAge variable with the age value from Dictionary 1.
import random
dictList = ['myDict', 'otherDict']
mydict = {
'age' : 10,
'other': "dummy data"
}
.
.
.
randomDict = random.choice(dictList)
dictAge = randomDict['age']
print(dictAge)
In the case of the code above what should happen is:
randomDict is assigned a random value from the distList variable (at the top). This sets which dictionary's values will be used going forward.
I next want the dictAge variable to then be assigned the age value from the selected dictionary. In this case (as mydict is was the only dictionary available) it should be assigned the age value of 10.
The error I am getting is:
TypeError: string indices must be integers
I know this is such a common error but my brain can't quite work out what the best solution is.
(Disclaimer: I haven't used python in ages so I know I am doing something really obviously silly but I can't quite work out what to do).
Right now, you are not actually using the definition of your dicts.
This is because dictList is comprised of strings: ['myDict', 'otherDict'].
So, when doing randomDict = random.choice(dictList), randomDict will either be the string 'myDict', or the string 'otherDict'.
Then you are doing randomDict['age'], which means you are trying to slice a string, with a string. As the error suggests, this can't be done and indices can only be ints.
What you want to do, is move the definition of the dictList to be after the definitions of your dicts, and include references to the dicts themselves, not strings. Something like:
mydict = {
'age' : 10,
'other': "dummy data"
}
.
.
.
dictList = [myDict, otherDict]
In the following piece of code:
dictAge = randomDict['age']
You are trying to index the name of dictionary variable (a string) returned by random.choice function.
To make it work you would need to do it using locals:
locals()[randomDict]['age']
or rather correct the dictList to contain the dictionaries instead of their names:
dictList = [myDict, otherDict]
In the latter case please note that myDict and otherDict should be declared before dictList.
when using a default value for a dictionary the comprehensive loop show empty list when asked to iterate for all key items
from collections import defaultdict
dict = {'whiz':1,'beerus':2,'vegeta':3,'goku':4}
dict = defaultdict(lambda : 'picalo')
print ([key for key in dict])
[]
process finished with exit code 0
this code is run in pycharm
Welcome to SO. Please include a description of expected behavior with your questions, or you'll get answers like this:
That's because there's nothing in the dictionary, it just has a default value!
As far as I can tell the error you're making is you're replacing your dictionary, not giving it a default value.
I think what you're looking for is something like this:
from collections import defaultdict
my_dict = {'whiz': 1,'beerus': 2,'vegeta': 3,'goku': 4}
my_dict = defaultdict(lambda: 'picalo', **my_dict)
print(my_dict)
Notice how I named the variable my_dict instead of dict, that's because dict is the "built-in" dictionary type and generally shouldn't be overwritten (to prevent bugs down the line).
I'm working with python dictionaries and ntlk on some reviews.I have and input (txt)file which is a simple review. In a dictionary all_dict.txt. I have all words (negative and positive) with word polarities and value.
all_dict.txt looks like this
"acceptable":("positive",1),"good":("positive",1),"shame":("negative",2),"bad":("negative",4),...
I want to know how can I get this polarities from a dictionary and a number value for each word so that I can get an output like this:
"acceptable_positive":1,"good_positive":1,"shame_negative":2,"bad_negative":4
I tried with dict.get(), dict.values but I don't get what I want. Is there a method to fetch key and values automatically?:
I tried with my code:
f_all_dict=open('all_dict.txt','r',encoding='utf-8').read()
f = eval(f_all_dict)
result_all = {}
for word in f.items():
suffix, pol=result_all[word] #pol->polarity
result_all[word + "_" + suffix] = pol
But I get KeyError if the word doesn't exist in an input file (review).
Thank you for your help
First off, the dict.items() return a dictitem object contains tuples of key and value and when you want to pass it as a key to your dictionary it raise a KeyError.
suffix, pol=result_all[word]
Secondly you better to use with statement in order to dealing with external objects like files. And use ast.literal_eval() for evaluating your dictionary. Also you can access to your value's items, by using throwaway variables unpacking :-) within a dict comprehension.
from ast import literal_eval
with open('all_dict.txt','r',encoding='utf-8') as f_all_dict:
dictionary = literal_eval(f_all_dict.read().strip())
result_all = {"{}_{}".format(word, suffix): pol for word, (suffix, pol) in dictionary.items()}
After modification my code looks like this. I didn't use with statement and it is working good.
f_all_dict=open('all_dict.txt','r',encoding='utf-8').read()
f = literal_eval(f_all_dict)
result_all = {}
for word in f.items():
result_all = {"{}_{}".format(word, suffix): pol * tokens.count(word) for word, (suffix, pol) in f.items()}
print(result_all)