File I/O

1# Read the entire file as one string
2with open("notes.txt", "r") as f:
3    content = f.read()
4    print(content)
5
6# Read line by line, memory-efficient for large files
7with open("notes.txt") as f:    # "r" is the default mode
8    for line in f:
9        print(line, end="")     # lines already include \n
10
11# Read all lines into a list
12with open("notes.txt") as f:
13    lines = f.readlines()       # ['line 1\n', 'line 2\n', ...]
14
15# Read one line at a time
16with open("notes.txt") as f:
17    first = f.readline()        # 'line 1\n'

1# Explicit encoding, always do this
2with open("data.txt", encoding="utf-8") as f:
3    content = f.read()
4
5# Common encodings
6# utf-8      universal, handles all Unicode
7# utf-8-sig  UTF-8 with BOM (common in Windows-generated CSVs)
8# latin-1    legacy Western European files

1# Write, creates or overwrites
2with open("output.txt", "w", encoding="utf-8") as f:
3    f.write("Hello, world!\n")
4    f.write("Second line\n")
5
6# writelines, no automatic newlines added
7lines = ["one\n", "two\n", "three\n"]
8with open("output.txt", "w", encoding="utf-8") as f:
9    f.writelines(lines)
10
11# Append, adds to end without truncating
12with open("log.txt", "a", encoding="utf-8") as f:
13    f.write("2024-01-15 server started\n")

1# Write a table of data cleanly
2headers = ["name", "score", "grade"]
3rows    = [
4    ["Alice", "92", "A"],
5    ["Bob",   "78", "C+"],
6    ["Carol", "85", "B"],
7]
8
9with open("results.txt", "w", encoding="utf-8") as f:
10    f.write("\t".join(headers) + "\n")
11    for row in rows:
12        f.write("\t".join(row) + "\n")

1# Copy a file byte-for-byte
2with open("image.png", "rb") as src:
3    with open("copy.png", "wb") as dst:
4        dst.write(src.read())
5
6# Read in chunks, for large files
7with open("large.bin", "rb") as f:
8    while chunk := f.read(8192):    # walrus operator, Python 3.8+
9        process(chunk)

1from pathlib import Path
2
3p = Path("/home/user/documents/report.pdf")
4
5# Parts and properties
6print(p.name)       # report.pdf
7print(p.stem)       # report
8print(p.suffix)     # .pdf
9print(p.parent)     # /home/user/documents
10
11# Building paths with /
12base    = Path("/home/user")
13config  = base / "config" / "settings.json"
14print(config)       # /home/user/config/settings.json
15
16# Current directory and relatives
17here    = Path.cwd()
18home    = Path.home()
19sibling = Path(__file__).parent / "data.csv"

1from pathlib import Path
2
3p = Path("data/output.txt")
4
5# Check existence and type
6print(p.exists())       # True/False
7print(p.is_file())      # True if it's a regular file
8print(p.is_dir())        # True if it's a directory
9
10# Create directories
11Path("logs/2024").mkdir(parents=True, exist_ok=True)
12
13# Read and write directly, no open() needed
14p.write_text("hello\n", encoding="utf-8")
15content = p.read_text(encoding="utf-8")
16
17p.write_bytes(b"\x00\x01\x02")
18data = p.read_bytes()
19
20# Glob for files
21for csv_file in Path("data").glob("*.csv"):
22    print(csv_file.name)
23
24# Recursive glob
25for py_file in Path(".").rglob("*.py"):
26    print(py_file)

1from pathlib import Path
2import shutil
3
4src = Path("report.txt")
5dst = Path("archive/report_backup.txt")
6
7# Rename / move
8src.rename(dst)               # moves within the same filesystem
9shutil.move(str(src), str(dst))  # works across filesystems
10
11# Copy
12shutil.copy2(src, dst)        # copy with metadata
13
14# Delete
15src.unlink()                  # delete file
16src.unlink(missing_ok=True)   # no error if absent (Python 3.8+)
17
18Path("empty_dir").rmdir()     # delete empty dir
19shutil.rmtree("dir_tree")     # delete directory and all contents

1import json
2
3# Write to a file
4config = {"host": "localhost", "port": 5432, "debug": False}
5with open("config.json", "w", encoding="utf-8") as f:
6    json.dump(config, f, indent=2)
7
8# Read from a file
9with open("config.json", encoding="utf-8") as f:
10    loaded = json.load(f)
11print(loaded["port"])    # 5432
12
13# Serialize to/from strings (e.g. for HTTP responses)
14json_str = json.dumps(config, indent=2)
15print(type(json_str))    # <class 'str'>
16
17back = json.loads(json_str)
18print(type(back))        # <class 'dict'>

1# Python  →  JSON
2# dict    →  object  {}
3# list    →  array   []
4# tuple   →  array   []   (round-trips as list)
5# str     →  string  ""
6# int     →  number
7# float   →  number
8# True    →  true
9# False   →  false
10# None    →  null
11
12data = {"items": (1, 2, 3), "flag": True, "value": None}
13print(json.dumps(data))
14# {"items": [1, 2, 3], "flag": true, "value": null}
15
16# json cannot serialize: sets, datetimes, custom classes
17import datetime
18# json.dumps(datetime.date.today())   # TypeError
19
20# Fix: convert manually or use a custom encoder
21print(json.dumps(datetime.date.today(), default=str))
22# "2024-01-15"

1data = {"z": 3, "a": 1, "m": 2}
2
3# sort_keys, deterministic output (good for testing and diffs)
4print(json.dumps(data, sort_keys=True))
5# {"a": 1, "m": 2, "z": 3}
6
7# indent, human-readable formatting
8print(json.dumps(data, indent=2))
9# {
10#   "z": 3,
11#   "a": 1,
12#   "m": 2
13# }
14
15# ensure_ascii=False, keep Unicode characters as-is
16print(json.dumps({"city": "Zürich"}, ensure_ascii=False))
17# {"city": "Zürich"}

1import csv
2
3# Read with csv.reader, rows are lists of strings
4with open("scores.csv", encoding="utf-8") as f:
5    reader = csv.reader(f)
6    header = next(reader)         # skip the header row
7    for row in reader:
8        name, score = row[0], int(row[1])
9        print(f"{name}: {score}")
10
11# Read with DictReader, rows are dicts keyed by header
12with open("scores.csv", encoding="utf-8") as f:
13    for row in csv.DictReader(f):
14        print(row["name"], row["score"])   # row is a dict

1import csv
2
3rows = [
4    ["Alice", 92, "A"],
5    ["Bob",   78, "C+"],
6    ["Carol", 85, "B"],
7]
8
9# csv.writer handles quoting and newlines
10with open("results.csv", "w", newline="", encoding="utf-8") as f:
11    writer = csv.writer(f)
12    writer.writerow(["name", "score", "grade"])   # header
13    writer.writerows(rows)                        # all data rows
14
15# DictWriter, write dicts
16students = [
17    {"name": "Alice", "score": 92},
18    {"name": "Bob",   "score": 78},
19]
20with open("students.csv", "w", newline="", encoding="utf-8") as f:
21    writer = csv.DictWriter(f, fieldnames=["name", "score"])
22    writer.writeheader()
23    writer.writerows(students)