Python Built-in Functions Reference

These functions tell you what a value is or what it contains:

For type checks in real code, prefer isinstance() to type(x) == T. isinstance(True, int) is True because bool subclasses int, and isinstance(value, (int, float)) accepts several types at once. type(x) == int would reject subclasses that should be acceptable.

dir() and help() are exploration tools for the interactive shell: help(str.split) shows a method's parameters without opening a browser.

Aggregation functions reduce a collection to one answer:

quantities = [4, 7, 2]
print(sum(quantities), min(quantities), max(quantities))   # 13 2 7
print(all(q > 0 for q in quantities))                      # True: every one positive
print(any(q > 5 for q in quantities))                      # True: at least one above 5
print(round(sum(quantities) / len(quantities), 2))         # 4.33

• min() and max() accept key= to compare by something else: max(products, key=len) finds the longest name.
• On an empty sequence, min() and max() raise ValueError. Pass default= to return a value instead: max(quantities, default=0).
• sum() starts from 0; sum(prices, start=10) starts from 10. Use math.fsum() for accurate sums of many floats.
• all([]) is True and any([]) is False, which follows logic ("no counter-examples") but surprises many people.

any() and all() stop as soon as the answer is known, so they're efficient with generator expressions.

These built-ins produce a new view or ordering of existing data:

names = ["kale", "bean", "squash"]
counts = [4, 7]
print(list(zip(names, counts)))                 # [('kale', 4), ('bean', 7)]
print(list(zip(names, counts, strict=False)))   # same: extra 'squash' ignored

zip() stops at the shortest input, silently dropping extras. From Python 3.10, zip(..., strict=True) raises ValueError when the lengths differ, which catches mismatched data. itertools.zip_longest() fills the gaps instead. reversed, enumerate, zip, map and filter return lazy iterators; wrap them in list() to see or reuse the results.

Calling a type creates a value of that type:

items = ["kale", "bean", "kale"]
print(list("abc"))                   # ['a', 'b', 'c']
print(set(items))                    # {'kale', 'bean'} (order may vary)
print(dict(zip(["kale", "bean"], [4, 7])))   # {'kale': 4, 'bean': 7}
print(tuple(range(3)))               # (0, 1, 2)

list(), tuple(), set() and dict() build collections from any iterable; int(), float(), str() and bool() convert single values, as covered in type casting.

Construction consumes iterators: list(some_generator) reads every remaining value, leaving the generator empty, and stores them all in memory. That's what you want when you need to reuse or index the data, and a mistake when the input is huge or endless. Two built-ins should almost never be used on outside data: eval() and exec() run arbitrary code, so a malicious string can do anything your program can. Use int(), float(), json.loads() or ast.literal_eval() to parse text instead.

Task: build a seed order summary using inspection, aggregation and paired traversal without eval or exec.

names = ["kale", "bean", "pea"]
raw_counts = ["4", "7", "x"]

counts = []
for name, raw in zip(names, raw_counts, strict=True):
    if raw.isdigit():
        counts.append((name, int(raw)))
    else:
        print(f"Skipping {name}: {raw!r} is not a count")

print(counts)
print("Types ok:", all(isinstance(c, int) for _, c in counts))
print("Total:", sum(c for _, c in counts), "Largest:", max(counts, key=lambda p: p[1])[0])

This prints the skip message for "pea", then [('kale', 4), ('bean', 7)], Types ok: True and Total: 11 Largest: bean. For the worked check, ordinary zip() with unequal lengths stops at the shortest iterable.

The documentation lists all built-in functions with their parameters. Continue with the methods reference and the standard library tour. Practise with the Python exercises and Python quiz.