I've a very large json file ( like 1,5gb ) and i need to transform it into csv.
The problem is that sometimes there's an extra field like:
[
{
"item": {
"name": "something",
"colors": {
"color_of_something": "something",
"color_of_something2": "something",
"color_of_something3": "something"
},
"dimensions": {
"dimensions1": "something",
"dimensions2": "something",
"dimensions3": "something"
},
"This_field_appears_sometimes": "something",
"description": {
"text": "something"
}
}
}]
I've this code to transform the json file into csv file:
# -*- coding: utf-8 -*-
import json, csv
with open("items.json") as file:
data = json.load(file)
csv_data = csv.writer(open('items.csv','wb+'))
csv_data.writerow(['item_name','item_color','item_dimension','item_random_field','item_description')
for json_parsed in data:
csv_data.writerow([
json_parsed['item']['name'],
json_parsed['item']['colors']['color_of_something'],
json_parsed['item']['dimensions']['dimensions1'],
json_parsed['item']['This_field_appears_sometimes'],
json_parsed['item']['description']['text']
])
When i run the task i'm getting this error:
KeyError: 'This_field_appears_sometimes'
Need some tip or advice to fix this, meanwhile i'll try if a len checkup works on this code.
You can use a "safe get" like this:
json_parsed['item'].get('This_field_appears_sometimes', '')
or check with a condition if that key is inside item
if 'This_field_appears_sometimes' in json_parsed['item'].keys()
The reason is no key 'This_field_appears_sometimes' in some item.
you can use json_parsed['item'].get('This_field_appears_sometimes') or check the json file
Related
With the following simple Python script:
import json
file = 'toy.json'
data = json.loads(file)
print(data['gas']) # example
My data generates the error ...is not JSON serializable.
With this, slightly more sophisticated, Python script:
import json
import sys
#load the data into an element
data = open('transactions000000000029.json', 'r')
#dumps the json object into an element
json_str = json.dumps(data)
#load the json to a string
resp = json.loads(json_str)
#extract an element in the response
print(resp['gas'])
The same.
What I'd like to do is extract all the values of a particular index, so ideally I'd like to render the input like so:
...
"hash": "0xf2b5b8fb173e371cbb427625b0339f6023f8b4ec3701b7a5c691fa9cef9daf63",
"gasUsed": "21000",
"hash": "0xf8f2a397b0f7bb1ff212b6bcc57e4a56ce3e27eb9f5839fef3e193c0252fab26"
"gasUsed": "21000"
...
The data looks like this:
{
"blockNumber": "1941794",
"blockHash": "0x41ee74e34cbf9ef4116febea958dbc260e2da3a6bf6f601bfaeb2cd9ab944a29",
"hash": "0xf2b5b8fb173e371cbb427625b0339f6023f8b4ec3701b7a5c691fa9cef9daf63",
"from": "0x3c0cbb196e3847d40cb4d77d7dd3b386222998d9",
"to": "0x2ba24c66cbff0bda0e3053ea07325479b3ed1393",
"gas": "121000",
"gasUsed": "21000",
"gasPrice": "20000000000",
"input": "",
"logs": [],
"nonce": "14",
"value": "0x24406420d09ce7440000",
"timestamp": "2016-07-24 20:28:11 UTC"
}
{
"blockNumber": "1941716",
"blockHash": "0x75e1602cad967a781f4a2ea9e19c97405fe1acaa8b9ad333fb7288d98f7b49e3",
"hash": "0xf8f2a397b0f7bb1ff212b6bcc57e4a56ce3e27eb9f5839fef3e193c0252fab26",
"from": "0xa0480c6f402b036e33e46f993d9c7b93913e7461",
"to": "0xb2ea1f1f997365d1036dd6f00c51b361e9a3f351",
"gas": "121000",
"gasUsed": "21000",
"gasPrice": "20000000000",
"input": "",
"logs": [],
"nonce": "1",
"value": "0xde0b6b3a7640000",
"timestamp": "2016-07-24 20:12:17 UTC"
}
What would be the best way to achieve that?
I've been thinking that perhaps the best way would be to reformat it as valid json?
Or maybe to just treat it like regex?
Your json file is not valid. This data should be a list of dictionaries. You should then separate each dictionary with a comma, Like this:
[
{
"blockNumber":"1941794",
"blockHash": "0x41ee74bf9ef411d9ab944a29",
"hash":"0xf2ef9daf63",
"from":"0x3c0cbb196e3847d40cb4d77d7dd3b386222998d9",
"to":"0x2ba24c66cbff0bda0e3053ea07325479b3ed1393",
"gas":"121000",
"gasUsed":"21000",
"gasPrice":"20000000000",
"input":"",
"logs":[
],
"nonce":"14",
"value":"0x24406420d09ce7440000",
"timestamp":"2016-07-24 20:28:11 UTC"
},
{
"blockNumber":"1941716",
"blockHash":"0x75e1602ca8d98f7b49e3",
"hash":"0xf8f2a397b0f7bb1ff212e193c0252fab26",
"from":"0xa0480c6f402b036e33e46f993d9c7b93913e7461",
"to":"0xb2ea1f1f997365d1036dd6f00c51b361e9a3f351",
"gas":"121000",
"gasUsed":"21000",
"gasPrice":"20000000000",
"input":"",
"logs":[
],
"nonce":"1",
"value":"0xde0b6b3a7640000",
"timestamp":"2016-07-24 20:12:17 UTC"
}
]
Then use this to open the file:
with open('toy.json') as data_file:
data = json.load(data_file)
You can then render the desired output like:
for item in data:
print item['hash']
print item['gasUsed']
If each block is valid JSON data you can parse them seperatly:
data = []
with open('transactions000000000029.json') as inpt:
lines = []
for line in inpt:
if line.startswith('{'): # block starts
lines = [line]
else:
lines.append(line)
if line.startswith('}'): # block ends
data.append(json.loads(''.join(lines)))
for block in data:
print("hash: {}".format(block['hash']))
print("gasUsed: {}".format(block['gasUsed']))
Okay, so I've been banging my head on this for the last 2 days, with no real progress. I am a beginner with python and coding in general, but this is the first issue I haven't been able to solve myself.
So I have this long file with JSON formatting with about 7000 entries from the youtubeapi.
right now I want to have a short script to print certain info ('videoId') for a certain dictionary key (refered to as 'key'):
My script:
import json
f = open ('path file.txt', 'r')
s = f.read()
trailers = json.loads(s)
print(trailers['key']['Items']['id']['videoId'])
# print(trailers['key']['videoId'] gives same response
Error:
print(trailers['key']['Items']['id']['videoId'])
TypeError: string indices must be integers
It does work when I want to print all the information for the dictionary key:
This script works
import json
f = open ('path file.txt', 'r')
s = f.read()
trailers = json.loads(s)
print(trailers['key'])
Also print(type(trailers)) results in class 'dict', as it's supposed to.
My JSON File is formatted like this and is from the youtube API, youtube#searchListResponse.
{
"kind": "youtube#searchListResponse",
"etag": "",
"nextPageToken": "",
"regionCode": "",
"pageInfo": {
"totalResults": 1000000,
"resultsPerPage": 1
},
"items": [
{
"kind": "youtube#searchResult",
"etag": "",
"id": {
"kind": "youtube#video",
"videoId": ""
},
"snippet": {
"publishedAt": "",
"channelId": "",
"title": "",
"description": "",
"thumbnails": {
"default": {
"url": "",
"width": 120,
"height": 90
},
"medium": {
"url": "",
"width": 320,
"height": 180
},
"high": {
"url": "",
"width": 480,
"height": 360
}
},
"channelTitle": "",
"liveBroadcastContent": "none"
}
}
]
}
What other information is needed to be given for you to understand the problem?
The following code gives me all the videoId's from the provided sample data (which is no id's at all in fact):
import json
with open('sampledata', 'r') as datafile:
data = json.loads(datafile.read())
print([item['id']['videoId'] for item in data['items']])
Perhaps you can try this with more data.
Hope this helps.
I didn't really look into the youtube api but looking at the code and the sample you gave it seems you missed out a [0]. Looking at the structure of json there's a list in key items.
import json
f = open ('json1.json', 'r')
s = f.read()
trailers = json.loads(s)
print(trailers['items'][0]['id']['videoId'])
I've not used json before at all. But it's basically imported in the form of dicts with more dicts, lists etc. Where applicable. At least from my understanding.
So when you do type(trailers) you get type dict. Then you do dict with trailers['key']. If you do type of that, it should also be a dict, if things work correctly. Working through the items in each dict should in the end find your error.
Pythons error says you are trying find the index/indices of a string, which only accepts integers, while you are trying to use a dict. So you need to find out why you are getting a string and not dict when using each argument.
Edit to add an example. If your dict contains a string on key 'item', then you get a string in return, not a new dict which you further can get a dict from. item in the json for example, seem to be a list, with dicts in it. Not a dict itself.
I have a json file that I load into python. I want to take a keyword from the file (which is very big), like country rank or review from info taken from the internet. I tried
json.load('filename.json')
but I am getting an error:
AttributeError: 'str' object has no attribute 'read.'
What am I doing wrong?
Additionally, how do I select part of a json file if it is very big?
I think you need to open the file then pass that to json load like this
import json
from pprint import pprint
with open('filename.json') as data:
output = json.load(data)
pprint(output)
Try the following:
import json
json_data_file = open("json_file_path", 'r').read() # r for reading the file
json_data = json.loads(json_data_file)
Access the data using the keys as follows :
json_data['key']
json.load() expects the file handle after it has been opened:
with open('filename.json') as datafile:
data = json.load(datafile)
For example if your json data looked like this:
{
"maps": [
{
"id": "blabla",
"iscategorical": "0"
},
{
"id": "blabla",
"iscategorical": "0"
}
],
"masks": {
"id": "valore"
},
"om_points": "value",
"parameters": {
"id": "valore"
}
}
To access parts of the data, use:
data["maps"][0]["id"]
data["masks"]["id"]
data["om_points"]
That code can be found in this SO answer:
Parsing values from a JSON file using Python?
OK, I am new to python but what I am trying to do is to access specific fields from a json text file
my json text file is like this:
{
"paging": {
"next": "https://graph.facebook.com/search?limit=5000&offset=5000&type=page&q=%26&locale=ar_AR&access_token=CAACEdEose0cBAD7z1vK0aO2Mlb1QZBOb9OwjYZCZBZB56P0MrYnt54WJYZCZBy4ZBv4zaYG0mj9ZCZAMkZBmlP83E885ykZAafog7QbcWwEtvRXfjtVa12DBnW8omWsnC8N6lsmNK7yktI89kBDdrTH9TOIdATHdsX5OewWhzGTpXDelSjE8HAbtcn08zSWsweDc4UZD&__after_id=139433456868"
},
"data": [
{
"category": "\u0627\u0644\u062a\u0639\u0644\u064a\u0645",
"name": "The London School of Economics and Political Science - LSE",
"category_list": [
{
"id": "108051929285833",
"name": "\u0627\u0644\u0643\u0644\u064a\u0629 \u0648\u0627\u0644\u062c\u0627\u0645\u0639\u0629"
},
{
"id": "187751327923426",
"name": "\u0645\u0646\u0638\u0645\u0629 \u062a\u0639\u0644\u064a\u0645\u064a\u0629"
}
],
"id": "6127898346"
},
the filed that I want to access is the 'category_list' filed in order to get the 'id' filed
I have tried some thing like this:
import json
idvalue = []
jsonFile = open('samples0.txt', 'r')
values = json.load(jsonFile)
jsonFile.close()
idValue = values['data'][0]['category_list'][0]['id']
print idvalue
but it keeps telling me that there is a key error.
what I am missing here?
what is the wrong thing I am doing?
any help please
edit :
my code returning null I still cannot understand why?
values['data'][0]['category_list'] is a list, so something like values['data'][0]['category_list'][0]['id'] should work.
No need to declare idValue. Just use it as
idValue = values['data'][0]['category_list'][0]['id']
I have an array of dictionaries like so:
myDict[0] = {'date':'today', 'status': 'ok'}
myDict[1] = {'date':'yesterday', 'status': 'bad'}
and I'm trying to export this array to a json file where each dictionary is its own entry. The problem is when I try to run:
dump(myDict, open("test.json", "w"))
It outputs a json file with a number prefix before each entry
{"0": {"date": "today", "status": "ok"}, "1": {"date": "yesterday", "status": "bad"} }
which apparently isn't legal json since my json parser (protovis) is giving me error messages
Any ideas?
Thanks
Use a list instead of a dictionary; you probably used:
myDict = {}
myDict[0] = {...}
You should use:
myList = []
myList.append({...}
P.S.: It seems valid json to me anyways, but it is an object and not a list; maybe this is the reason why your parser is complaining
You should use a JSON serializer...
Also, an array of dictionaries would better serialize to something like this:
[
{
"date": "today",
"status": "ok"
},
{
"date": "yesterday",
"status": "bad"
}
]
That is, you should just use a JavaScript array.