How to delete an element in a json file python - python

I am trying to delete an element in a json file,
here is my json file:
before:
{
"names": [
{
"PrevStreak": false,
"Streak": 0,
"name": "Brody B#3719",
"points": 0
},
{
"PrevStreak": false,
"Streak": 0,
"name": "XY_MAGIC#1111",
"points": 0
}
]
}
after running script:
{
"names": [
{
"PrevStreak": false,
"Streak": 0,
"name": "Brody B#3719",
"points": 0
}
]
}
how would I do this in python? the file is stored locally and I am deciding which element to delete by the name in each element
Thanks

I would load the file, remove the item, and then save it again. Example:
import json
with open("filename.json") as f:
data = json.load(f)
f.pop(data["names"][1]) # or iterate through entries to find matching name
with open("filename.json", "w") as f:
json.dump(data, f)

You will have to read the file, convert it to python native data type (e.g. dictionary), then delete the element and save the file. In your case something like this could work:
import json
filepath = 'data.json'
with open(filepath, 'r') as fp:
data = json.load(fp)
del data['names'][1]
with open(filepath, 'w') as fp:
json.dump(data, fp)

Try this:
# importing the module
import ast
# reading the data from the file
with open('dictionary.txt') as f:
data = f.read()
print("Data type before reconstruction : ", type(data))
# reconstructing the data as a dictionary
a_dict = ast.literal_eval(data)
{"names":[a for a in a_dict["names"] if a.get("name") !="XY_MAGIC#1111"]}

import json
with open("test.json",'r') as f:
data = json.loads(f.read())
names=data.get('names')
for idx,name in enumerate(names):
if name['name']=='XY_MAGIC#1111':
del names[idx]
break
print(names)
In order to read the file best approach would be using the with statement after which you could just use pythons json library and convert json string to python dict. once you get dict you can access the values and do your operations as required. you could convert it as json using json.dumps() then save it

This does the right thing useing the python json module, and prettyprints the json back to the file afterwards:
import json
jsonpath = '/path/to/json/file.json'
with open(jsonpath) as file:
j = json.loads(file.read())
names_to_remove = ['XY_MAGIC#1111']
for element in j['names']:
if element['name'] in names_to_remove:
j['names'].remove(element)
with open(jsonpath, 'w') as file:
file.write(json.dumps(j, indent=4))

Related

CSV to JSON of Lists

Currently, I have a CSV file with the following example --
File
skill
experience
overall_experience
1
Java
1.5
3
1
Python
1.0
3
1
SQL
0.5
3
There are multiple entries for many such files but I need to merge the skills and their respective experience into a single value belonging to a single key, something like this -
{
"1": {
"file": "1",
"skill": ["Java", "Python", "SQL"],
"experience": [1.5, 1.0, 0.5]
"Overall_exp": 3.0
},}
I tried a Python Code for this but it is giving me only the value of last skill and last experience (and not the whole thing in a list)
Here is the code I was using --
import csv
import json
# Function to convert a CSV to JSON
# Takes the file paths as arguments
def make_json(csvFilePath, jsonFilePath):
# create a dictionary
data = {}
# Open a csv reader called DictReader
with open(csvFilePath, encoding='utf-8') as csvf:
csvReader = csv.DictReader(csvf)
# Convert each row into a dictionary
# and add it to data
for rows in csvReader:
# Assuming a column named 'file' to
# be the primary key
key = rows['file']
data[key] = rows
# Open a json writer, and use the json.dumps()
# function to dump data
with open(jsonFilePath, 'w', encoding='utf-8') as jsonf:
jsonf.write(json.dumps(data, indent=4))
# Decide the two file paths according to your
# computer system
csvFilePath = 'skill_matrix.csv'
jsonFilePath = 'skill_matrix.json'
# Call the make_json function
make_json(csvFilePath, jsonFilePath)
The output that I get here is this --
{
"1": {
"file": "1",
"skill": "SQL",
"experience": "0.5"
"Overall_exp": "3.0"
},}
How can I convert it to the former json format and not the latter?
You can use pandas to read your csv, group by File and export to json:
df = pd.read_csv(your_csv)
df = df.groupby('File', as_index=False).agg({'skill': list, 'experience': list, 'overall_experience': np.mean})
print(df.to_json(orient='index', indent=4))
Note: you can specify the aggregation functions for your columns in a dictionary
Output:
{
"0":{
"File":1,
"skill":[
"Java",
"Python",
"SQL"
],
"experience":[
1.5,
1.0,
0.5
],
"overall_experience":3.0
}
}
I think that loading into Pandas first and then going from all data to the narrowing strategy is cleaner and easier. You can use the following code for parsing your data into JSON files;
import pandas as pd
import json
# Load the CSV into Pandas
df = pd.read_csv('1.csv', header=0)
data = df.to_dict(orient='list')
# Delete / change as you wish
data['File'] = str(data['File'][0])
data['overall_experience'] = data['overall_experience'][0]
# Save as json
with open('1.json', 'w', encoding='utf-8') as jsonf:
jsonf.write(json.dumps(data, indent=4))
Result (1.json)
{
"File": "1",
"skill": [
"Java",
"Python",
"SQL"
],
"experience": [
1.5,
1.0,
0.5
],
"overall_experience": 3
}
I suppose hat you have multiple file id in a CSV file. Your given example is too minimalistic. Anyhow, then you can create a master dictionary and add your smaller ones as follows;
import pandas as pd
import json
# Load the CSV into Pandas
df = pd.read_csv('1.csv', header=0)
# Master dictionary
master_dict = {}
for idx, file_id in enumerate(df["File"].unique()):
data = df[df['File'] == file_id].to_dict(orient='list')
# Delete / change as you wish
data['File'] = str(data['File'][0])
data['overall_experience'] = data['overall_experience'][0]
master_dict[idx] = data
# Save as json
with open('1.json', 'w', encoding='utf-8') as jsonf:
jsonf.write(json.dumps(master_dict, indent=4))
Result (1.json)
{
"0": {
"File": "1",
"skill": [
"Java",
"Python",
"SQL"
],
"experience": [
1.5,
1.0,
0.5
],
"overall_experience": 3
},
"1": {
"File": "2",
"skill": [
"Java",
"Python"
],
"experience": [
2.0,
2.5
],
"overall_experience": 1
}
}
If you don't want to use Pandas, you could try:
import csv
import json
def make_json(csvfile_path, jsonfile_path):
data = {}
with open(csvfile_path, "r") as csvfile:
next(csvfile) # Skip header line
for row in csv.reader(csvfile):
fdata = data.setdefault(row[0], {"file": row[0]})
fdata.setdefault("skill", []).append(row[1])
fdata.setdefault("experience", []).append(float(row[2]))
fdata.setdefault("overall_experience", []).append(float(row[3]))
with open(jsonfile_path, "w") as jsonfile:
json.dump(data, jsonfile)
The main difference to your approach is the explicit structuring of the inner dicts: values are lists (except for the 'file' key). The dict.setdefault() is great here: You can set a value for a key if it isn't in the dict, and get the value back (either the newly set one or the existing). So you can put a list in the dict, get it back, and can immediately .append() to it.
If you want to use a csv.DictReader:
def make_json(csvfile_path, jsonfile_path):
data = {}
with open(csvfile_path, "r") as csvfile:
for row in csv.DictReader(csvfile):
fdata = data.setdefault(row["file"], {"file": row["file"]})
for key, value in list(row.items())[1:]:
fdata.setdefault(key, []).append(
value if key == "skill" else float(value)
)
with open(jsonfile_path, "w") as jsonfile:
json.dump(data, jsonfile)
(I haven't, since I wasn't sure about the actual column names.)

Separate large JSON object into many different files

I have a JSON file with 10000 data entries like below in a file.
{
"1":{
"name":"0",
"description":"",
"image":""
},
"2":{
"name":"1",
"description":"",
"image":""
},
...
}
I need to write each entry in this object into its own file.
For example, the output of each file looks like this:
1.json
{
"name": "",
"description": "",
"image": ""
}
I have the following code, but I'm not sure how to proceed from here. Can anyone help with this?
import json
with open('sample.json', 'r') as openfile:
# Reading from json file
json_object = json.load(openfile)
You can use a for loop to iterate over all the fields in the outer object, and then create a new file for each inner object:
import json
with open('sample.json', 'r') as input_file:
json_object = json.load(input_file)
for key, value in json_object.items():
with open(f'{key}.json', 'w') as output_file:
json.dump(value, output_file)

Replace json nested dictionary value with a file value

I have two files: json and text files.
I would like to replace one of the dictionary value with the value which in the text file.
Let us say, in the text file file.text, I have the following lists. [11, 15, 10].
In the json file, I have the following dictionary.
"aa": {
"bb": [
"25",
"40",
"05"
],
"cc": [
"20"
]
}
I would like to overwrite the cc value with the text file above.
file.json
"aa": {
"bb": [
"25",
"40",
"05"
],
"cc": [
"11", "15", "10"
]
}
I have tried something in Python.
def replace(text_file, json_file):
tex_file_path = 'C:/Documents/file.txt'
with open(os.path.join(tex_file_path, text_file), 'r') as f:
read_text= f.read()
json_file_path = 'C:/Documents/file.json'
with open(os.path.join(json_file_path, json_file), 'r') as f:
read_json = json.load(f)
text_to_be_replaced = read_json.get('aa')
for value in text_to_be_replaced.items():
for element in value:
# statment
I was wondering if someone can really help with this.
Although you've named it .text, the contents of the file appear to be JSON, so you can use json.load() as well. Then convert the integers in the list to strings and insert it into the desired place in the JSON file.
There's no need to loop over the dictionary items. Just address the specific element you want to replace.
def replace(text_file, json_file):
tex_file_path = 'C:/Documents'
with open(os.path.join(tex_file_path, text_file), 'r') as f:
read_text= json.load(f)
read_text = list(map(str, read_text))
json_file_path = 'C:/Documents'
with open(os.path.join(json_file_path, json_file), 'r') as f:
read_json = json.load(f)
read_json["aa"]["cc"] = read_text
with open(os.path.join(json_file_path, json_file), 'w') as f:
json.dump(read_json, f)
Also, your XXX_path variables should just be directories, the filename comes from the function parameter.
Here's a simple example using a StringIO to demonstrate reading / writing from a file-like object:
import json
from io import StringIO
json_file_obj = StringIO("""
{"aa": {
"bb": [
"25",
"40",
"05"
],
"cc": [
"20"
]
}
}
""")
text_file_obj = StringIO("[11, 15, 10]")
def replace(src_file_obj: StringIO, repl_file_obj: StringIO):
# Load file contents into a Python object
data = json.load(src_file_obj)
# Read in txt file contents
new_cc_value = json.load(repl_file_obj)
# But now result will be a list of int, here we want a list of string
new_cc_value = list(map(str, new_cc_value))
# Replace desired value
data['aa']['cc'] = new_cc_value
# Now we write to our file-like object, `src_file_obj`
# This is to demonstrate replacing the original file contents
src_file_obj = StringIO()
json.dump(data, src_file_obj)
# Seek to the start of the file
src_file_obj.seek(0)
return src_file_obj
json_file_obj = replace(json_file_obj, text_file_obj)
print(json_file_obj.read())
Output:
{"aa": {"bb": ["25", "40", "05"], "cc": ["11", "15", "10"]}}
Hint - If you want to write the output to an actual file, you can replace these lines below:
src_file_obj = StringIO()
json.dump(data, src_file_obj)
src_file_obj.seek(0)
With these lines:
with open("file_name.txt", 'w') as out_file:
json.dump(data, out_file)

Python JSON add Key-Value pair

I'm trying to add key value pairs into the existing JSON file. I am able to concatenate to the parent label, How to add value to the child items?
JSON file:
{
"students": [
{
"name": "Hendrick"
},
{
"name": "Mikey"
}
]
}
Code:
import json
with open("input.json") as json_file:
json_decoded = json.load(json_file)
json_decoded['country'] = 'UK'
with open("output.json", 'w') as json_file:
for d in json_decoded[students]:
json.dump(json_decoded, json_file)
Expected Results:
{
"students": [
{
"name": "Hendrick",
"country": "UK"
},
{
"name": "Mikey",
"country": "UK"
}
]
}
You can do the following in order to manipulate the dict the way you want:
for s in json_decoded['students']:
s['country'] = 'UK'
json_decoded['students'] is a list of dictionaries that you can simply iterate and update in a loop. Now you can dump the entire object:
with open("output.json", 'w') as json_file:
json.dump(json_decoded, json_file)
import json
with open("input.json", 'r') as json_file:
json_decoded = json.load(json_file)
for element in json_decoded['students']:
element['country'] = 'UK'
with open("output.json", 'w') as json_out_file:
json.dump(json_decoded, json_out_file)
opened a json file i.e. input.json
iterated through each of its element
add a key named "country" and dynamic value "UK", to each element
opened a new json file with the modified JSON.
Edit:
Moved writing to output file inside to first with segment. Issue with earlier implemenation is that json_decoded will not be instantiated if opening of input.json fails. And hence, writing to output will raise an exception - NameError: name 'json_decoded' is not defined
This gives [None, None] but update the dict:
a = {'students': [{'name': 'Hendrick'}, {'name': 'Mikey'}]}
[i.update({'country':'UK'}) for i in a['students']]
print(a)

Adding json element top of the other keys

I have json file something like this one.
{
"SomethingA": {
"SomethingB": {
"SomethingC": {
"C-property": "something",
"C-property2": {}
}
}
}
}
I want to add some new data top of the the "Something C" as "NEWSomethingC"
so It should be
{
"SomethingA": {
"SomethingB": {
"NEWSomethingC": {
"NEWC-property": "NEWsomething",
"NEWC-property2": {}
},
"SomethingC": {
"C-property": "something",
"C-property2": {}
}
}
}
}
Okay, here is the problem. I can't add new value top of the keys. Always, NEWSomethingC is going to appear below the SomethingC.
The code I use for adding...
with open(credantials.init['config'], 'r+') as f:
data = json.load(f)
try:
old_data = data['SomethingA'][SomethingB]
append_data = data['SomethingA'][SomethingB]
old_data = {NEWSomethingC :{'C-property':something, 'C-Property2':{}}}
except KeyError:
print ('There is no key you want to search here')
append_data.update(old_data)
print(append_data)
f.seek(0)
json.dump(data,f, indent=4)
f.truncate()
As already pointed out dictionaries in python are unorderd. Therefore we have to use OrderedDict
As explained in this answer we can use the object_pairs_hook
argument in json.loads() to load as OrderdDicts. Then we can add a new dictionary to our "OrderdJsonDictionary" and use the move_to_end function to move our added dictionary to the begin
with open(credantials.init['config'], 'r+') as f:
data = json.load(f, object_pairs_hook=OrderedDict)
new_data = {'newc':{
"NEWC-property": "NEWsomething",
"NEWC-property2": {}
}
}
data["SomethingA"]["SomethingB"].update(new_data)
# last=False moves to the beginning
data["SomethingA"]["SomethingB"].move_to_end(list(new_data.keys())[0], last=False)
f.seek(0)
json.dump(data,f, indent=4)
f.truncate()
So what you would want to do is read the data, search for the data to the point where you wish to make an insertion.
1. Write that data to a new file
2. add your new insertion to the new file
3. add the rest of the file contents to the new file
4. delete the old file
So in order to write to file you would want to insert the following into your code.
outfile = open('file.json')
json.dump(data, outfile)
outfile.close()

Categories

Resources