Working with Text and Real Projects: Quick Revision

A brief note on Working with Text and Real Projects. Read it before a test. Try each question at the end before you open its answer.

Cut, join and tidy

• split() cuts a string into a list of words, at any run of spaces. split(",") cuts at each comma.
• "-".join(words) glues a list of strings into one string. The separator comes first.
• strip() removes the spaces at the two ends. strip(".,!") removes those marks instead.
• replace(old, new) changes every match. startswith(), endswith() and in answer True or False.
• A string never changes. A method gives a new string, so keep it: name = name.strip().

Clean, then count

• Clean first: lower(), then replace every punctuation mark by a space.
• Then split() gives the words, and a dictionary counts them: counts[w] = counts.get(w, 0) + 1.
• max(words, key=len) is the longest word. max(counts, key=counts.get) is the most common one.
• A tie goes to the first one that max meets. Decide on a tie rule, and write it down.

Letters are numbers

• ord("a") is 97, and chr(97) is "a".
• A Caesar code moves each letter: the new place is (place + shift) % 26. The % 26 wraps z round to a. To decode, use -shift.
• A palindrome reads the same backwards: clean it, then compare it with text[::-1]. Two anagrams give the same sorted() letters.
• In a long English text, e is the most common letter. That cracks a Caesar code.

Looking inside text

• find() gives a position, or -1. index() stops with an error instead. count() says how many.
• isdigit(), isalpha(), isalnum(), isspace(), isupper() and islower() test the kind of characters.
• partition("=") cuts at the first = into three parts. An empty middle part means the separator was missing.
• Check a line before you trust it. Skip a bad line, and go on with the rest.

Building a project

• The method: the goal, the jobs, the data, the expected answers by hand, the functions, a table of tests, the user part, and then grow it.
• Put the thinking in functions without input() and print(), so that a table of tests can check them.
• Look things up by name with a dictionary, and tidy the key with lower().
• A dictionary of functions maps a command word to its function. Write the name without brackets.
• statistics.mean(), median() and stdev() summarise marks. An empty list raises a StatisticsError.
• Catch EOFError, so that a program that reads input ends cleanly.

A worked example

Which word is the most common in " Rain, rain... go AWAY! Rain! "? Follow the cleaning one step at a time.

1. Small letters. lower() gives " rain, rain... go away! rain! ".
2. Marks out. Each of ,, . and ! is replaced by a space. Only spaces and letters are left.
3. Words. split() cuts at every run of spaces: ['rain', 'rain', 'go', 'away', 'rain'].
4. Count. The dictionary becomes {'rain': 3, 'go': 1, 'away': 1}.
5. The winner. max(counts, key=counts.get) gives 'rain'.

Without step 1, Rain and rain would be two different words. Without step 2, rain, and rain would be two more.

Practice with new numbers

A program to try

A small text toolkit: it counts the words of a line, finds the longest and the most common, and codes a message.

Mistakes to avoid

• Not keeping the result. name.strip() on its own changes nothing, because a string never changes. Write name = name.strip().
• split(" ") for words. Two spaces in a row make an empty word, ''. Use split().
• The separator in the wrong place. words.join("-") stops with AttributeError: 'list' object has no attribute 'join'. Write "-".join(words).
• Joining numbers. "-".join([4, 5]) stops with TypeError: sequence item 0: expected str instance, int found. Change them to text first.
• index() for a missing piece. "cat".index("z") stops with ValueError: substring not found. find() gives -1 instead.
• Forgetting % 26. The letter z moved by 2 becomes |, not b.
• A missing key. contacts["Zara"] for a name that is not in the book stops with KeyError: 'Zara'. Ask with in first, or use get().
• Brackets in a table of functions. {"add": do_add()} calls the function at once. Write do_add.

Check yourself

1. What does " apples, 40 kg ".strip().partition(",") give?

2. Why is "12.5".isdigit() False? How else could a program check that a text is a decimal number?

3. Are "Dormitory" and "Dirty room" anagrams? What must a program do before it compares them?

4. A long English message was coded with a Caesar shift. Its commonest letter is h. Which shift was used? (e is at place 4, and h is at place 7.)

5. Write an f-string that shows a price of 1234.5 rupees as Rs 1,234.50.

6. Why should the thinking of a program sit in functions without input() and print()?