10. JSON

.json Let's look at how to load files with extensions into loaders.

• Note: https://python.langchain.com/docs/modules/data_connection/document_loaders/json

Copy

import json
from pathlib import Path
from pprint import pprint


file_path = "data/people.json"
data = json.loads(Path(file_path).read_text())

pprint(data)

Copy

[{'address': {'city':'Seoul','street': '312 times','zipCode': '83795'}, 
  'age': 31, 
  'carOwnership': True, 
  'hobbies': ['cooking','music listening','photo shooting'], 
  'isMarried': True, 
  'name':'boxy', 
  'phoneNumbers': ['483-4639-1933', '947-4179-7976']}, 
 {'address': {'city':'Seoul','street': '877 times','zipCode': '36780'}, 
  'age': 31, 
  'carOwnership': True, 
  'hobbies': ['travel','music appreciation','climbing'], 
  'isMarried': False, 
  'name':'singer', 
  'phoneNumbers': ['337-5721-3227', '387-3768-9586']}, 
 ... 
 (meditation) 
 ... 
 {'address': {'city':'Seoul','street': '940th Street','zipCode': '60335'}, 
  'age': 52, 
  'carOwnership': True, 
  'hobbies': ['cooking','photo shooting','climbing'], 
  'isMarried': True, 
  'name':'Chief Yoon', 
  'phoneNumbers': ['290-8270-9786', '483-1765-4028']}, 
 {'address': {'city':'Seoul','street': '289 times','zipCode': '59793'}, 
  'age': 18, 
  'carOwnership': True, 
  'hobbies': ['Travel','Cooking','Watch the Movie'], 
  'isMarried': True, 
  'name':'Minister', 
  'phoneNumbers': ['460-4533-7245', '344-2344-7362']}] 

Copy

type(data[0])

Copy

dict

JSONLoader

Assuming you want to extract the value below the content field within the message key of the JSON data, you can easily do it through JSONLoader as shown below.

Copy

from langchain_community.document_loaders import JSONLoader

# JSONLoader generation
loader = JSONLoader(
    file_path="data/people.json",
    jq_schema=".[].phoneNumbers",
    text_content=False,
)

# load document
docs = loader.load()

# Output the results
pprint(docs)

Copy

[Document (metadata={'source':'/Users/teddy/Dev/github/langchain-master/07-DocumentLoader/data/people.json','seq_num': 1}, page_content=" [ '483-46 
 Document (metadata={'source':'/Users/teddy/Dev/github/langchain-master/07-DocumentLoader/data/people.json','seq_num': 2}, page_content=" [ '337-5721- 
 Document (metadata={'source':'/Users/teddy/Dev/github/langchain-master/07-DocumentLoader/data/people.json','seq_num': 3}, page_content=" [ '354-563-4 
... 
(meditation) 
... 
 Document (metadata={'source':'/Users/teddy/Dev/github/langchain-master/07-DocumentLoader/data/people.json','seq_num': 19}, page_content=" [ '290-8270 
 Document (metadata={'source':'/Users/teddy/Dev/github/langchain-master/07-DocumentLoader/data/people.json','seq_num': 20}, page_content=" [ '460-453