6. Files and Exceptions - LAC

Read Python Crash Course Chapter 8
As well as json, similar interfaces exist for other formats like yaml, toml, ...
We can also download things from the web as follows:
import urllib def text_from_url(url): response = urllib.request.urlopen(url) return response.read().decode('utf-8') houn = text_from_url("https://www.gutenberg.org/cache/epub/2852/pg2852.txt")
The main different is that you must decode from bytes to some encoding
And that you need to be connected to the internet!
We will use two new methods for sequences (lists and strings):
- seq.count(item) returns the number of elements with the specified value
  'green banana'.count('na') returns 2
- seq.index(item) returns the position of the first match
  'green banana'.count('na') returns 8

For google colab, files are stored in your virtual machine.

⚠ They will disappear at the end of the session.
You can upload or download them to your own machine
from google.colab import files files.download('/content/spec.txt')
Downloading files to your local file system

Write a program to:
1. Get the text of All Cats are Grey by Andre Norton
  go to gutenberg and search for 'All Cats are Grey'
  find the url of the textfile
  read it in (using text_from_url)
2. Save it as a file called "acag-raw.txt"
3. Read it again and chop off extra text, then save as "acag.txt"
  guess how many letters, increase or decrease until you have the right amount!
4. If you are in colab, download it to your computer.
Read it in again and count characters, words and lines
store these in a dictionary
❂ Also count the number of sentences
Save that dictionary as JSON, YAML and TOML and look at the output
See what happens when you get the URL wrong with text_from_url
- Add exception handling to output a message if the URL cannot be found
Write a function to discard the header and footer of a GUTENBERG file:
everything before the line '*** START' ..., ' ***'
and after the line '*** END OF THE PROJECT GUTENBERG EBOOK', ...
hint: use index
Download a different book and print the first 200 and last 200 characters.

I have prepared a text file extracted from the duolingo wiki for Czech. The file is at skill0.txt (it was made using extract.py --- README, URLs).

Download skill0.txt and extract the vocabulary. The result should be a list of dictionaries, with each dictionary representing a word. The dictionary should indicate which lesson a word was in, the Czech word, and a list of English translations. For example, the first word in the first lesson is::
# ne = no/not { "lesson": "1", "cs": "ne", "en": [ "no", "not" ] }
- First download the file
- The write a function that loops through each line and does something
- If the line is a lesson, store it
- If the line is a word, process it using another function
Finally save the file as json
Note to make the json more readable, you can give json.dump/json.dumps optional parameters:
indent = 2 formats it nicely,
ensure_ascii = False allows it to use unicode.
It is not necessary for the computer, but it can help you to debug.
Read the json file and give some statistics
- How many Czech words are there?
- How many English words are there?
- How many pairs of words are there?
Print this nicely formatted
Write a short program that reads the json file, and quizzes people. It should give a Czech word, and ask them to type the English (or q for quit). If the English is correct it should say 'Well done', and ask the next word. If the English is incorrect it should ask again.
Note, you may notice you want to improve your parsing of the vocabulary, ...

Start on project 1!

You learned how to work with files in Python.
- You can now read the entire contents of a file
- You can process the contents line by line
  good for big files
- You can write to a file
- You can read and write structured data, like JSON, TOML and YAML
- You also learned how to download from the internet
Furthermore, you studied exceptions and how to handle them:
try: with open('filename.txt') as file: contents = file.read() except FileNotFoundError: print("File not found.")