Skip to content

Latest commit

 

History

History
145 lines (107 loc) · 3.92 KB

lecture5.md

File metadata and controls

145 lines (107 loc) · 3.92 KB
Raw

Day 5: File I/O: Reading and Writing Text, CSV, JSON files

Working with Text Files

  • Used to store data such that it can be retrieved any time
  • To read or write from a file, the following operations take place:
  1. Open a file
  2. Perform Read or write operations
  3. Close the file

Open a file

f = open('hello.txt', 'r')
  • First argument is the string containing file name
  • Second argument is the file opening mode. (r, w, a, etc.)
  • Default, files are opened in text mode.

Reading from file

f.read(5) # read next 5 bytes
f.readline() # read the entire line
f.readlines() # read the entire file and return list of lines
f.tell() # get the current file position
f.seek(0) # bring file cursor to initial position

Writing to file

f.write("hello world\n")

Closing a file

  • Frees up the resources tied with the file
f.close()
  • Need to worry about exceptions if any operation on file is running
  • Use a try...finally block of code

Open file using context manager

  • Need not close the file explicitly
with open('hello.txt') as f:
  # perform file operations

Working with CSV Files

  • CSV (Comma Separated Values) format is one of the most simplest ways to store tabular data.
  • Elements are separated by a delimiter, generally, a comma ,.
  • The python's csv module makes it easier to operate on such files.

Reading CSV File

import csv
with open('marks.csv', 'r') as file:
  reader = csv.reader(file)
  for row in reader:
    print(row)
  • Explore skipinitialspace attribute in the reader function.
  • There can be many other attributes specified and can make code difficult to maintain, so to solve this dialect is provided as another optional parameter, which groups together many formatting patterns.

Writing to CSV File

import csv
with open('marks.csv', 'w') as file:
  writer = csv.writer(file)
  writer.writerow(["Roll No.", "Name", "Marks"])
  writer.writerow([1, "abc", 95])
  writer.writerow([2, "def", 97])
  • You can also try creating rowlist as list of lists and use the writer.writerows function.

Read file as a dictionary

import csv
with open('marks.csv', 'r') as file:
  csv_dict = csv.DictReader(file)
  for row in csv_dict:
    print(row)
  • Similarly, can try exploring DictWriter.
  • Also try to explore the Pandas library which is a popular library for data manipulation and analysis.

Working with JSON Files

  • Saving more complex data types like nested lists and dictionaries, parsing and serializing becomes complicated.
  • Python allows us to use the popular data interchange format JSON (JavaScript Object Notation) through a module json.
  • Converting python data to string is known as serializing.
  • Reconstructing the data from the string representation is called deserializing.

Reading json from file

import json
with open('marks.json') as file:
  data = json.load(file)
  print(data) # dict
  • json.loads can be used to convert string to dict.

Writing json to file

import json

marks_dict = {
  "roll_no": 12,
  "name": "abc",
  "marks": 99
}
with open('marks.json', 'w') as json_file:
  json.dump(marks_dict, json_file)
  • json.dumps can be used to convert dict to string.
  • Can pass indent and sort_keys properties to pretty print the json data.

Example code

Other libraries and documentation