JSON and regular expressions

JSON

JSON is how programs send data to each other. It looks almost exactly like a Python dictionary. When you call an AI model's API, the reply is JSON.

json_demo.py

import json

text = '{"name": "Riya", "marks": [78, 85], "passed": true}'

data = json.loads(text)        # Text to Python
print(data["name"], data["marks"][1])

data["city"] = "Pune"
print(json.dumps(data))        # Python back to text

Output

Riya 85
{"name": "Riya", "marks": [78, 85], "passed": true, "city": "Pune"}

Regular expressions

A regular expression describes a pattern. It's the standard tool for cleaning text before it goes into a model.

regex_demo.py

import re

text = "Call 98765 43210 or mail riya@example.com by 5 pm"

print(re.findall(r"\d+", text))                # Groups of digits
print(re.findall(r"\S+@\S+", text))            # Something like an email
print(re.sub(r"\d", "#", text))                # Hide the digits

Output

['98765', '43210', '5']
['riya@example.com']
Call ##### ##### or mail riya@example.com by # pm

Pattern	Matches
\d	A digit
\w	A letter, digit or underscore
\s	A space
+	One or more of the previous
*	Zero or more of the previous
^ and $	Start and end of the text

Finished reading?